Application Studies using GRAM++ - CSE, IIT Bombay：使用革兰氏+ + CSE的应用研究,IIT孟买81页PPT

综合理论课程教育研究学法教法研究 277计部门要加强对被审计单位或部门的监督，也可考虑和高校联合起来，共同开展回头看，等等。

通过采取一系列务实、创新的举措，真正将整改工作落到实处。

2.提高质量。

无论是审计工作方式还是撰写审计报告，都要精益求精，经得住推敲。

如果不认真对待审计内容，工作拖拖拉拉、质量不高的话，那么审计出来的结果也是经不起推敲，不能加以运用的。

这样浪费了大量的人力、物力、财力资源，却没有收到实效。

所以，要全方面提高审计工作的质量。

一是端正审计工作的态度，要以高度负责的精神和态度来对待审计，不可马马虎虎地工作。

二是优化审计工作流程，创新审计工作方式。

工作要有逻辑性，按照要求按部就班、不折不扣地做好工作，做到流程清楚，逐项落实；要改变传统的工作方式，按照审计的要求创新工作思路和工作方法，力求将各项工作做得扎实、务实。

三是要撰写高质量的审计工作报告。

审计工作报告是对本次审计工作的总结，具有很强的针对性和指导性，审计工作报告质量的好坏，直接影响了本次审计是否取得成功，这对审计人员提出了更高的要求。

一方面审计人员要懂业务，会正确使用审计术语，了解并掌握审计工作的整个流程，另一方面要有较强的综合分析能力和文字综合能力，能够透过现象看本质，将审计内容进行高度概括和凝练，从而提升审计报告的质量。

3.加强运用。

要将审计后的结果加以应用，防止出现“只审不用”的现象。

被审计单位班子成员要正确认识审计出的问题，对这些问题高度重视，进行积极的整改。

另外，审计者也要加强对审计结果的运用，尤其在审计工作流程结束后，仍然要重视审计结果的运用，还要将运用情况上报给审计部门，这也是对本次审计工作的重视及肯定。

三、结语高校领导干部的经济责任审计对高校经济活动的正常开展具有非常重要的作用。

随着高校的发展，一些新情况新问题也会出现，对经济责任审计将会提出新的更高的要求。

高校领导干部的经济责任审计要紧密结合工作实际，工作者要牢固树立创新意识，不断对审计工作方式进行创新，优化工作流程，提升工作质量，切实取得经济责任审计工作的实效。

Computer Science and Application 计算机科学与应用, 2015, 5, 204-211Published Online June 2015 in Hans. /journal/csa/10.12677/csa.2015.56026Remote Monitoring System for SmartDomestic Floor Heating SystemBased on Node.jsZiran Wu1, Duo Li1, Zhenghui Yang1, Hua Ye1,21School of Automation, Southeast University, Nanjing Jiangsu2Key Laboratory of Measurement and Control of CSE of Ministry of Education, Southeast University, Nanjing JiangsuEmail: zhineng@Received: Jun. 5th, 2015; accepted: Jun. 22nd, 2015; published: Jun. 25th, 2015Copyright © 2015 by authors and Hans Publishers Inc.This work is licensed under the Creative Commons Attribution International License (CC BY)./licenses/by/4.0/AbstractA remote control system is a very common subsystem of a smart homer system. The architectureof the system is usually based on C/S mode or B/S mode, which sets limits on the functionality and performance and has to ignore some user demands. This article is going to design a remote control system based on C/S and B/S mixed mode using Node.js, which is based on a smart domestic floor heating program. The method of building a Node.js server will be described in details in this ar-ticle, and the tools and methods to design a web page and a client software will be introduced briefly.KeywordsNode.js, C/S + B/S Mixed Mode基于Node.js的家庭智能地暖远程监控系统吴子然1，李多1，杨争辉1，叶桦1,21东南大学自动化学院，江苏南京2东南大学复杂工程系统测量与控制教育部重点实验室，江苏南京Email: zhineng@基于Node.js的家庭智能地暖远程监控系统收稿日期：2015年6月5日；录用日期：2015年6月22日；发布日期：2015年6月25日摘要智能家居系统中往往会有远程控制部分，这些远程监控系统通常是基于单一C/S或B/S模式实现的，其功能和性能受到单一模式的限制，无法最大化满足用户需求。

SCI论文摘要中常用的表达方法要写好摘要，需要建立一个适合自己需要的句型库（选择的词汇来源于SCI高被引用论文）引言部分（1）回顾研究背景，常用词汇有review, summarize, present, outline, describe等（2）说明写作目的，常用词汇有purpose, attempt, aim等，另外还可以用动词不定式充当目的壮语老表达（3）介绍论文的重点内容或研究范围，常用词汇有study, present, include, focus, emphasize, emphasis, attention等方法部分（1）介绍研究或试验过程，常用词汇有test study, investigate, examine,experiment, discuss, consider, analyze, analysis等（2）说明研究或试验方法，常用词汇有measure, estimate, calculate等（3）介绍应用、用途，常用词汇有use, apply, application等结果部分（1）展示研究结果，常用词汇有show, result, present等（2）介绍结论，常用词汇有summary, introduce,conclude等讨论部分（1）陈述论文的论点和作者的观点，常用词汇有suggest, repot, present, expect, describe 等（2）说明论证，常用词汇有support, provide, indicate, identify, find, demonstrate, confirm, clarify等（3）推荐和建议，常用词汇有suggest,suggestion, recommend, recommendation, propose,necessity,necessary,expect等。

摘要引言部分案例词汇review•Author(s): ROBINSON, TE; BERRIDGE, KC•Title:THE NEURAL BASIS OF DRUG CRA VING - AN INCENTIVE-SENSITIZATION THEORY OF ADDICTION•Source: BRAIN RESEARCH REVIEWS, 18 (3): 247-291 SEP-DEC 1993 《脑研究评论》荷兰SCI被引用1774We review evidence for this view of addiction and discuss its implications for understanding the psychology and neurobiology of addiction.回顾研究背景SCI高被引摘要引言部分案例词汇summarizeAuthor(s): Barnett, RM; Carone, CD; 被引用1571Title: Particles and field .1. Review of particle physicsSource: PHYSICAL REVIEW D, 54 (1): 1-+ Part 1 JUL 1 1996:《物理学评论，D辑》美国引言部分回顾研究背景常用词汇summarizeAbstract: This biennial review summarizes much of Particle Physics. Using data from previous editions, plus 1900 new measurements from 700 papers, we list, evaluate, and average measuredproperties of gauge bosons, leptons, quarks, mesons, and baryons. We also summarize searches for hypothetical particles such as Higgs bosons, heavy neutrinos, and supersymmetric particles. All the particle properties and search limits are listed in Summary Tables. We also give numerous tables, figures, formulae, and reviews of topics such as the Standard Model, particle detectors, probability, and statistics. A booklet is available containing the Summary Tables and abbreviated versions of some of the other sections of this full Review.SCI摘要引言部分案例attentionSCI摘要方法部分案例considerSCI高被引摘要引言部分案例词汇outline•Author(s): TIERNEY, L SCI引用728次•Title:MARKOV-CHAINS FOR EXPLORING POSTERIOR DISTRIBUTIONS 引言部分回顾研究背景，常用词汇outline•Source: ANNALS OF STATISTICS, 22 (4): 1701-1728 DEC 1994•《统计学纪事》美国•Abstract: Several Markov chain methods are available for sampling from a posterior distribution. Two important examples are the Gibbs sampler and the Metropolis algorithm.In addition, several strategies are available for constructing hybrid algorithms. This paper outlines some of the basic methods and strategies and discusses some related theoretical and practical issues. On the theoretical side, results from the theory of general state space Markov chains can be used to obtain convergence rates, laws of large numbers and central limit theorems for estimates obtained from Markov chain methods. These theoretical results can be used to guide the construction of more efficient algorithms. For the practical use of Markov chain methods, standard simulation methodology provides several Variance reduction techniques and also gives guidance on the choice of sample size and allocation.SCI高被引摘要引言部分案例回顾研究背景presentAuthor(s): L YNCH, M; MILLIGAN, BG SC I被引用661Title: ANAL YSIS OF POPULATION GENETIC-STRUCTURE WITH RAPD MARKERS Source: MOLECULAR ECOLOGY, 3 (2): 91-99 APR 1994《分子生态学》英国Abstract: Recent advances in the application of the polymerase chain reaction make it possible to score individuals at a large number of loci. The RAPD (random amplified polymorphic DNA) method is one such technique that has attracted widespread interest.The analysis of population structure with RAPD data is hampered by the lack of complete genotypic information resulting from dominance, since this enhances the sampling variance associated with single loci as well as induces bias in parameter estimation. We present estimators for several population-genetic parameters (gene and genotype frequencies, within- and between-population heterozygosities, degree of inbreeding and population subdivision, and degree of individual relatedness) along with expressions for their sampling variances. Although completely unbiased estimators do not appear to be possible with RAPDs, several steps are suggested that will insure that the bias in parameter estimates is negligible. To achieve the same degree of statistical power, on the order of 2 to 10 times more individuals need to be sampled per locus when dominant markers are relied upon, as compared to codominant (RFLP, isozyme) markers. Moreover, to avoid bias in parameter estimation, the marker alleles for most of these loci should be in relatively low frequency. Due to the need for pruning loci with low-frequency null alleles, more loci also need to be sampled with RAPDs than with more conventional markers, and sole problems of bias cannot be completely eliminated.SCI高被引摘要引言部分案例词汇describe•Author(s): CLONINGER, CR; SVRAKIC, DM; PRZYBECK, TR•Title: A PSYCHOBIOLOGICAL MODEL OF TEMPERAMENT AND CHARACTER•Source: ARCHIVES OF GENERAL PSYCHIATRY, 50 (12): 975-990 DEC 1993《普通精神病学纪要》美国•引言部分回顾研究背景，常用词汇describe 被引用926•Abstract: In this study, we describe a psychobiological model of the structure and development of personality that accounts for dimensions of both temperament and character. Previous research has confirmed four dimensions of temperament: novelty seeking, harm avoidance, reward dependence, and persistence, which are independently heritable, manifest early in life, and involve preconceptual biases in perceptual memory and habit formation. For the first time, we describe three dimensions of character that mature in adulthood and influence personal and social effectiveness by insight learning about self-concepts.Self-concepts vary according to the extent to which a person identifies the self as (1) an autonomous individual, (2) an integral part of humanity, and (3) an integral part of the universe as a whole. Each aspect of self-concept corresponds to one of three character dimensions called self-directedness, cooperativeness, and self-transcendence, respectively. We also describe the conceptual background and development of a self-report measure of these dimensions, the Temperament and Character Inventory. Data on 300 individuals from the general population support the reliability and structure of these seven personality dimensions. We discuss the implications for studies of information processing, inheritance, development, diagnosis, and treatment.摘要引言部分案例•（2）说明写作目的，常用词汇有purpose, attempt, aimSCI高被引摘要引言部分案例attempt说明写作目的•Author(s): Donoho, DL; Johnstone, IM•Title: Adapting to unknown smoothness via wavelet shrinkage•Source: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 90 (432): 1200-1224 DEC 1995 《美国统计学会志》被引用429次•Abstract: We attempt to recover a function of unknown smoothness from noisy sampled data. We introduce a procedure, SureShrink, that suppresses noise by thresholding the empirical wavelet coefficients. The thresholding is adaptive: A threshold level is assigned to each dyadic resolution level by the principle of minimizing the Stein unbiased estimate of risk (Sure) for threshold estimates. The computational effort of the overall procedure is order N.log(N) as a function of the sample size N. SureShrink is smoothness adaptive: If the unknown function contains jumps, then the reconstruction (essentially) does also; if the unknown function has a smooth piece, then the reconstruction is (essentially) as smooth as the mother wavelet will allow. The procedure is in a sense optimally smoothness adaptive: It is near minimax simultaneously over a whole interval of the Besov scale; the size of this interval depends on the choice of mother wavelet. We know from a previous paper by the authors that traditional smoothing methods-kernels, splines, and orthogonal series estimates-even with optimal choices of the smoothing parameter, would be unable to perform in a near-minimax way over many spaces in the Besov scale.Examples of SureShrink are given. The advantages of the method are particularly evident when the underlying function has jump discontinuities on a smooth backgroundSCI高被引摘要引言部分案例To investigate说明写作目的•Author(s): OLTV AI, ZN; MILLIMAN, CL; KORSMEYER, SJ•Title: BCL-2 HETERODIMERIZES IN-VIVO WITH A CONSERVED HOMOLOG, BAX, THAT ACCELERATES PROGRAMMED CELL-DEATH•Source: CELL, 74 (4): 609-619 AUG 27 1993 被引用3233•Abstract: Bcl-2 protein is able to repress a number of apoptotic death programs. To investigate the mechanism of Bcl-2's effect, we examined whether Bcl-2 interacted with other proteins. We identified an associated 21 kd protein partner, Bax, that has extensive amino acid homology with Bcl-2, focused within highly conserved domains I and II. Bax is encoded by six exons and demonstrates a complex pattern of alternative RNA splicing that predicts a 21 kd membrane (alpha) and two forms of cytosolic protein (beta and gamma). Bax homodimerizes and forms heterodimers with Bcl-2 in vivo. Overexpressed Bax accelerates apoptotic death induced by cytokine deprivation in an IL-3-dependent cell line. Overexpressed Bax also counters the death repressor activity of Bcl-2. These data suggest a model in which the ratio of Bcl-2 to Bax determines survival or death following an apoptotic stimulus.SCI高被引摘要引言部分案例purposes说明写作目的•Author(s): ROGERS, FJ; IGLESIAS, CA•Title: RADIATIVE ATOMIC ROSSELAND MEAN OPACITY TABLES•Source: ASTROPHYSICAL JOURNAL SUPPLEMENT SERIES, 79 (2): 507-568 APR 1992 《天体物理学杂志增刊》美国SCI被引用512•Abstract: For more than two decades the astrophysics community has depended on opacity tables produced at Los Alamos. In the present work we offer new radiative Rosseland mean opacity tables calculated with the OPAL code developed independently at LLNL. We give extensive results for the recent Anders-Grevesse mixture which allow accurate interpolation in temperature, density, hydrogen mass fraction, as well as metal mass fraction. The tables are organized differently from previous work. Instead of rows and columns of constant temperature and density, we use temperature and follow tracks of constant R, where R = density/(temperature)3. The range of R and temperature are such as to cover typical stellar conditions from the interior through the envelope and the hotter atmospheres. Cool atmospheres are not considered since photoabsorption by molecules is neglected. Only radiative processes are taken into account so that electron conduction is not included. For comparison purposes we present some opacity tables for the Ross-Aller and Cox-Tabor metal abundances. Although in many regions the OPAL opacities are similar to previous work, large differences are reported.For example, factors of 2-3 opacity enhancements are found in stellar envelop conditions.SCI高被引摘要引言部分案例aim说明写作目的•Author(s):EDV ARDSSON, B; ANDERSEN, J; GUSTAFSSON, B; LAMBERT, DL;NISSEN, PE; TOMKIN, J•Title:THE CHEMICAL EVOLUTION OF THE GALACTIC DISK .1. ANALYSISAND RESULTS•Source: ASTRONOMY AND ASTROPHYSICS, 275 (1): 101-152 AUG 1993 《天文学与天体物理学》被引用934•Abstract:With the aim to provide observational constraints on the evolution of the galactic disk, we have derived abundances of 0, Na, Mg, Al, Si, Ca, Ti, Fe, Ni, Y, Zr, Ba and Nd, as well as individual photometric ages, for 189 nearby field F and G disk dwarfs.The galactic orbital properties of all stars have been derived from accurate kinematic data, enabling estimates to be made of the distances from the galactic center of the stars‘ birthplaces. 结构式摘要•Our extensive high resolution, high S/N, spectroscopic observations of carefully selected northern and southern stars provide accurate equivalent widths of up to 86 unblended absorption lines per star between 5000 and 9000 angstrom. The abundance analysis was made with greatly improved theoretical LTE model atmospheres. Through the inclusion of a great number of iron-peak element absorption lines the model fluxes reproduce the observed UV and visual fluxes with good accuracy. A new theoretical calibration of T(eff) as a function of Stromgren b - y for solar-type dwarfs has been established. The new models and T(eff) scale are shown to yield good agreement between photometric and spectroscopic measurements of effective temperatures and surface gravities, but the photometrically derived very high overall metallicities for the most metal rich stars are not supported by the spectroscopic analysis of weak spectral lines.•Author(s): PAYNE, MC; TETER, MP; ALLAN, DC; ARIAS, TA; JOANNOPOULOS, JD•Title:ITERA TIVE MINIMIZATION TECHNIQUES FOR ABINITIO TOTAL-ENERGY CALCULATIONS - MOLECULAR-DYNAMICS AND CONJUGA TE GRADIENTS•Source: REVIEWS OF MODERN PHYSICS, 64 (4): 1045-1097 OCT 1992 《现代物理学评论》美国American Physical Society SCI被引用2654 •Abstract: This article describes recent technical developments that have made the total-energy pseudopotential the most powerful ab initio quantum-mechanical modeling method presently available. In addition to presenting technical details of the pseudopotential method, the article aims to heighten awareness of the capabilities of the method in order to stimulate its application to as wide a range of problems in as many scientific disciplines as possible.SCI高被引摘要引言部分案例includes介绍论文的重点内容或研究范围•Author(s):MARCHESINI, G; WEBBER, BR; ABBIENDI, G; KNOWLES, IG;SEYMOUR, MH; STANCO, L•Title: HERWIG 5.1 - A MONTE-CARLO EVENT GENERA TOR FOR SIMULATING HADRON EMISSION REACTIONS WITH INTERFERING GLUONS SCI被引用955次•Source: COMPUTER PHYSICS COMMUNICATIONS, 67 (3): 465-508 JAN 1992:《计算机物理学通讯》荷兰Elsevier•Abstract: HERWIG is a general-purpose particle-physics event generator, which includes the simulation of hard lepton-lepton, lepton-hadron and hadron-hadron scattering and soft hadron-hadron collisions in one package. It uses the parton-shower approach for initial-state and final-state QCD radiation, including colour coherence effects and azimuthal correlations both within and between jets. This article includes a brief review of the physics underlying HERWIG, followed by a description of the program itself. This includes details of the input and control parameters used by the program, and the output data provided by it. Sample output from a typical simulation is given and annotated.SCI高被引摘要引言部分案例presents介绍论文的重点内容或研究范围•Author(s): IDSO, KE; IDSO, SB•Title: PLANT-RESPONSES TO ATMOSPHERIC CO2 ENRICHMENT IN THE FACE OF ENVIRONMENTAL CONSTRAINTS - A REVIEW OF THE PAST 10 YEARS RESEARCH•Source: AGRICULTURAL AND FOREST METEOROLOGY, 69 (3-4): 153-203 JUL 1994 《农业和林业气象学》荷兰Elsevier 被引用225•Abstract:This paper presents a detailed analysis of several hundred plant carbon exchange rate (CER) and dry weight (DW) responses to atmospheric CO2 enrichment determined over the past 10 years. It demonstrates that the percentage increase in plant growth produced by raising the air's CO2 content is generally not reduced by less than optimal levels of light, water or soil nutrients, nor by high temperatures, salinity or gaseous air pollution. More often than not, in fact, the data show the relative growth-enhancing effects of atmospheric CO2 enrichment to be greatest when resource limitations and environmental stresses are most severe.SCI高被引摘要引言部分案例介绍论文的重点内容或研究范围emphasizing •Author(s): BESAG, J; GREEN, P; HIGDON, D; MENGERSEN, K•Title: BAYESIAN COMPUTATION AND STOCHASTIC-SYSTEMS•Source: STATISTICAL SCIENCE, 10 (1): 3-41 FEB 1995《统计科学》美国•SCI被引用296次•Abstract: Markov chain Monte Carlo (MCMC) methods have been used extensively in statistical physics over the last 40 years, in spatial statistics for the past 20 and in Bayesian image analysis over the last decade. In the last five years, MCMC has been introduced into significance testing, general Bayesian inference and maximum likelihood estimation. This paper presents basic methodology of MCMC, emphasizing the Bayesian paradigm, conditional probability and the intimate relationship with Markov random fields in spatial statistics.Hastings algorithms are discussed, including Gibbs, Metropolis and some other variations. Pairwise difference priors are described and are used subsequently in three Bayesian applications, in each of which there is a pronounced spatial or temporal aspect to the modeling. The examples involve logistic regression in the presence of unobserved covariates and ordinal factors; the analysis of agricultural field experiments, with adjustment for fertility gradients; and processing oflow-resolution medical images obtained by a gamma camera. Additional methodological issues arise in each of these applications and in the Appendices. The paper lays particular emphasis on the calculation of posterior probabilities and concurs with others in its view that MCMC facilitates a fundamental breakthrough in applied Bayesian modeling.SCI高被引摘要引言部分案例介绍论文的重点内容或研究范围focuses •Author(s): HUNT, KJ; SBARBARO, D; ZBIKOWSKI, R; GAWTHROP, PJ•Title: NEURAL NETWORKS FOR CONTROL-SYSTEMS - A SURVEY•Source: AUTOMA TICA, 28 (6): 1083-1112 NOV 1992《自动学》荷兰Elsevier•SCI被引用427次•Abstract:This paper focuses on the promise of artificial neural networks in the realm of modelling, identification and control of nonlinear systems. The basic ideas and techniques of artificial neural networks are presented in language and notation familiar to control engineers. Applications of a variety of neural network architectures in control are surveyed. We explore the links between the fields of control science and neural networks in a unified presentation and identify key areas for future research.SCI高被引摘要引言部分案例介绍论文的重点内容或研究范围focus•Author(s): Stuiver, M; Reimer, PJ; Bard, E; Beck, JW;•Title: INTCAL98 radiocarbon age calibration, 24,000-0 cal BP•Source: RADIOCARBON, 40 (3): 1041-1083 1998《放射性碳》美国SCI被引用2131次•Abstract: The focus of this paper is the conversion of radiocarbon ages to calibrated (cal) ages for the interval 24,000-0 cal BP (Before Present, 0 cal BP = AD 1950), based upon a sample set of dendrochronologically dated tree rings, uranium-thorium dated corals, and varve-counted marine sediment. The C-14 age-cal age information, produced by many laboratories, is converted to Delta(14)C profiles and calibration curves, for the atmosphere as well as the oceans. We discuss offsets in measured C-14 ages and the errors therein, regional C-14 age differences, tree-coral C-14 age comparisons and the time dependence of marine reservoir ages, and evaluate decadal vs. single-year C-14 results. Changes in oceanic deepwater circulation, especially for the 16,000-11,000 cal sp interval, are reflected in the Delta(14)C values of INTCAL98.SCI高被引摘要引言部分案例介绍论文的重点内容或研究范围emphasis •Author(s): LEBRETON, JD; BURNHAM, KP; CLOBERT, J; ANDERSON, DR•Title: MODELING SURVIV AL AND TESTING BIOLOGICAL HYPOTHESES USING MARKED ANIMALS - A UNIFIED APPROACH WITH CASE-STUDIES •Source: ECOLOGICAL MONOGRAPHS, 62 (1): 67-118 MAR 1992•《生态学论丛》美国•Abstract: The understanding of the dynamics of animal populations and of related ecological and evolutionary issues frequently depends on a direct analysis of life history parameters. For instance, examination of trade-offs between reproduction and survival usually rely on individually marked animals, for which the exact time of death is most often unknown, because marked individuals cannot be followed closely through time.Thus, the quantitative analysis of survival studies and experiments must be based oncapture-recapture (or resighting) models which consider, besides the parameters of primary interest, recapture or resighting rates that are nuisance parameters. 结构式摘要•T his paper synthesizes, using a common framework, these recent developments together with new ones, with an emphasis on flexibility in modeling, model selection, and the analysis of multiple data sets. The effects on survival and capture rates of time, age, and categorical variables characterizing the individuals (e.g., sex) can be considered, as well as interactions between such effects. This "analysis of variance" philosophy emphasizes the structure of the survival and capture process rather than the technical characteristics of any particular model. The flexible array of models encompassed in this synthesis uses a common notation. As a result of the great level of flexibility and relevance achieved, the focus is changed from fitting a particular model to model building and model selection.SCI摘要方法部分案例•方法部分•（1）介绍研究或试验过程，常用词汇有test，study, investigate, examine,experiment, discuss, consider, analyze, analysis等•（2）说明研究或试验方法，常用词汇有measure, estimate, calculate等•（3）介绍应用、用途，常用词汇有use, apply, application等SCI高被引摘要方法部分案例discusses介绍研究或试验过程•Author(s): LIANG, KY; ZEGER, SL; QAQISH, B•Title: MULTIV ARIATE REGRESSION-ANAL YSES FOR CATEGORICAL-DATA •Source:JOURNAL OF THE ROY AL STA TISTICAL SOCIETY SERIES B-METHODOLOGICAL, 54 (1): 3-40 1992《皇家统计学会志，B辑：统计方法论》•SCI被引用298•Abstract: It is common to observe a vector of discrete and/or continuous responses in scientific problems where the objective is to characterize the dependence of each response on explanatory variables and to account for the association between the outcomes. The response vector can comprise repeated observations on one variable, as in longitudinal studies or genetic studies of families, or can include observations for different variables.This paper discusses a class of models for the marginal expectations of each response and for pairwise associations. The marginal models are contrasted with log-linear models.Two generalized estimating equation approaches are compared for parameter estimation.The first focuses on the regression parameters; the second simultaneously estimates the regression and association parameters. The robustness and efficiency of each is discussed.The methods are illustrated with analyses of two data sets from public health research SCI高被引摘要方法部分案例介绍研究或试验过程examines•Author(s): Huo, QS; Margolese, DI; Stucky, GD•Title: Surfactant control of phases in the synthesis of mesoporous silica-based materials •Source: CHEMISTRY OF MATERIALS, 8 (5): 1147-1160 MAY 1996•SCI被引用643次《材料的化学性质》美国•Abstract: The low-temperature formation of liquid-crystal-like arrays made up of molecular complexes formed between molecular inorganic species and amphiphilic organic molecules is a convenient approach for the synthesis of mesostructure materials.This paper examines how the molecular shapes of covalent organosilanes, quaternary ammonium surfactants, and mixed surfactants in various reaction conditions can be used to synthesize silica-based mesophase configurations, MCM-41 (2d hexagonal, p6m), MCM-48 (cubic Ia3d), MCM-50 (lamellar), SBA-1 (cubic Pm3n), SBA-2 (3d hexagonal P6(3)/mmc), and SBA-3(hexagonal p6m from acidic synthesis media). The structural function of surfactants in mesophase formation can to a first approximation be related to that of classical surfactants in water or other solvents with parallel roles for organic additives. The effective surfactant ion pair packing parameter, g = V/alpha(0)l, remains a useful molecular structure-directing index to characterize the geometry of the mesophase products, and phase transitions may be viewed as a variation of g in the liquid-crystal-Like solid phase. Solvent and cosolvent structure direction can be effectively used by varying polarity, hydrophobic/hydrophilic properties and functionalizing the surfactant molecule, for example with hydroxy group or variable charge. Surfactants and synthesis conditions can be chosen and controlled to obtain predicted silica-based mesophase products. A room-temperature synthesis of the bicontinuous cubic phase, MCM-48, is presented. A low-temperature (100 degrees C) and low-pH (7-10) treatment approach that can be used to give MCM-41 with high-quality, large pores (up to 60 Angstrom), and pore volumes as large as 1.6 cm(3)/g is described.Estimates 介绍研究或试验过程SCI高被引摘要方法部分案例•Author(s): KESSLER, RC; MCGONAGLE, KA; ZHAO, SY; NELSON, CB; HUGHES, M; ESHLEMAN, S; WITTCHEN, HU; KENDLER, KS•Title:LIFETIME AND 12-MONTH PREV ALENCE OF DSM-III-R PSYCHIATRIC-DISORDERS IN THE UNITED-STA TES - RESULTS FROM THE NATIONAL-COMORBIDITY-SURVEY•Source: ARCHIVES OF GENERAL PSYCHIATRY, 51 (1): 8-19 JAN 1994•《普通精神病学纪要》美国SCI被引用4350次•Abstract: Background: This study presents estimates of lifetime and 12-month prevalence of 14 DSM-III-R psychiatric disorders from the National Comorbidity Survey, the first survey to administer a structured psychiatric interview to a national probability sample in the United States.Methods: The DSM-III-R psychiatric disorders among persons aged 15 to 54 years in the noninstitutionalized civilian population of the United States were assessed with data collected by lay interviewers using a revised version of the Composite International Diagnostic Interview. Results: Nearly 50% of respondents reported at least one lifetime disorder, and close to 30% reported at least one 12-month disorder. The most common disorders were major depressive episode, alcohol dependence, social phobia, and simple phobia. More than half of all lifetime disorders occurred in the 14% of the population who had a history of three or more comorbid disorders. These highly comorbid people also included the vast majority of people with severe disorders.Less than 40% of those with a lifetime disorder had ever received professional treatment,and less than 20% of those with a recent disorder had been in treatment during the past 12 months. Consistent with previous risk factor research, it was found that women had elevated rates of affective disorders and anxiety disorders, that men had elevated rates of substance use disorders and antisocial personality disorder, and that most disorders declined with age and with higher socioeconomic status. Conclusions: The prevalence of psychiatric disorders is greater than previously thought to be the case. Furthermore, this morbidity is more highly concentrated than previously recognized in roughly one sixth of the population who have a history of three or more comorbid disorders. This suggests that the causes and consequences of high comorbidity should be the focus of research attention. The majority of people with psychiatric disorders fail to obtain professional treatment. Even among people with a lifetime history of three or more comorbid disorders, the proportion who ever obtain specialty sector mental health treatment is less than 50%.These results argue for the importance of more outreach and more research on barriers to professional help-seekingSCI高被引摘要方法部分案例说明研究或试验方法measure•Author(s): Schlegel, DJ; Finkbeiner, DP; Davis, M•Title:Maps of dust infrared emission for use in estimation of reddening and cosmic microwave background radiation foregrounds•Source: ASTROPHYSICAL JOURNAL, 500 (2): 525-553 Part 1 JUN 20 1998 SCI 被引用2972 次《天体物理学杂志》美国•The primary use of these maps is likely to be as a new estimator of Galactic extinction. To calibrate our maps, we assume a standard reddening law and use the colors of elliptical galaxies to measure the reddening per unit flux density of 100 mu m emission. We find consistent calibration using the B-R color distribution of a sample of the 106 brightest cluster ellipticals, as well as a sample of 384 ellipticals with B-V and Mg line strength measurements. For the latter sample, we use the correlation of intrinsic B-V versus Mg, index to tighten the power of the test greatly. We demonstrate that the new maps are twice as accurate as the older Burstein-Heiles reddening estimates in regions of low and moderate reddening. The maps are expected to be significantly more accurate in regions of high reddening. These dust maps will also be useful for estimating millimeter emission that contaminates cosmic microwave background radiation experiments and for estimating soft X-ray absorption. We describe how to access our maps readily for general use.SCI高被引摘要结果部分案例application介绍应用、用途•Author(s): MALLAT, S; ZHONG, S•Title: CHARACTERIZATION OF SIGNALS FROM MULTISCALE EDGES•Source: IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 14 (7): 710-732 JUL 1992•SCI被引用508次《IEEE模式分析与机器智能汇刊》美国•Abstract: A multiscale Canny edge detection is equivalent to finding the local maxima ofa wavelet transform. We study the properties of multiscale edges through the wavelet。

Unit 1Text II. Writing Skills1.Narration (primary purpose):experience of an interview2.Sequence: time. (It is written in the order which the events occurred)Application ----Fear ----Interview3.Description (not lengthy, does not carry much weight)The description contributes to the main purpose of narration, convincing the reader of the unpleasantness of his first job interview. It helps to make the narrative more vivid.①Journey: awkward/ a long way from home/ inconvenient transportation / hot②Surroundings:Outside the school building: dreary house/ dust/ fumes from a busy main roadInside the school building: smell of the cabbage / dingy/ ink marks/ crumbs on the carpet③Impression of the headmaster: unpleasant appearanceII. Language Points1.advertise①advertise (v.)vt.e.g. Do salesmen ~ their goods?Where do people ~ their products?They ~ their products in newspapers or on TV.vi.~ fore.g. The family will ~ for a servant.People usually ~ for their lost properties in a newspaper or on a billboard②advertiser (n.)③advertisement (n.) = ade.g. The store has an ~ in the newspaper for a special sale.Have you ever received ~s in your mailbox?cf. commercial (n.)---an advertisement on television or radio (电视、广播之) 广告2. in a suburb: a part of the suburbsin the suburbs: all or one of the outlying district of a citye.g. Is our university in the city or in the suburbs?Is Shipai a suburb of Guangzhou?He planned to build a house in the suburb of Shanghai.cf. outskirts (n.) (通常指郊外适合于散步的地方)---(esp. of a town) the outer areas (常用于in the outskirts 片语) 尤指城镇郊区；郊外---the outer limits (常用于on the outskirts 片语中) 边界e.g. We live in the suburbs.The children have been out for the day to some green spot on the outskirts.3. be short of: have not enough; lack ;be in need ofbe short on: be poorer at, be not so good at 在…稍差，在…较欠缺e.g. The book has good information, but rather short on illustration.She is a nice person, but short on looks.4.apply①apply (v.)a) request~ to sb. for sth.e.g. If you want to go abroad, you may ~ to the Consul for a visa.He has applied to the banker for a loan.I will ~ to the boss for the position in person, not by letter.b)put into use or into position to serve its purpose 应用~ sth. to sth.e.g. He applies a sticking plaster to the cut.Do you ~ your knowledge to the practice?The teacher will ~ the rule to every student.~ oneself / one’s energy/ one’s mind to sth./ doing sth.---concentrate one’s thoughts on a taske.g. We must ~ ourselves to our study.Paraphrase:He concentrates himself on learning French .If you try your best to do the job, you will soon finish it.②applicant (n.)③application (n.)make an ~ to sb. for sth.e.g. We made an ~ to the court for inquiry.The manager has received 20 ~s for the position.5.degree①rank or grade given by a university to one who has passed an exame.g. Bachelor of Arts / Science (BA/ BS)Master of Arts /Science (MA/ MS)Doctor of Philosophy (PH.D) 哲学博士Doctor of Science / Medicine / Law②to/ in …degree 等级，程度e.g.. To what degree can he be trusted?He cannot be trusted in the slightest degree.6. land①go, come, put on land; bring to lande.g. The plane ~ed safely.The pilot ~ed the airliner.②bring to reach a positione.g. He ~ed himself in great difficulties.③get (informal)e.g. If I manage to ~ the job, I will stand you all a free drink.He ~ed a valuable prize.7. chances of: (often plural with singular meaning) possibility; likelihood thatsth. will happene.g. You would have more chances of catching the train if you got a bus to the station instead ofwalking.Note: chances of doing sth. = hopes of doing sth.a chance of doing sth. / to do sth. = the opportunity to do sth.8. slim (adj.)e.g. ~ hopes / chances of successa ~ excusea ~ girl: a slender girl9. summon: demand the presence of; call~ sb to sth. / ~ sb. to do sth.e.g. The manager ~ed the shareholders to a general meeting.She ~ed a servant.The teacher ~ed me to the office.They ~ed three women to appear as witness 传唤证人cf.summons (n.) 传票10. prove (proved, proved/ proven)----to show (oneself or itself) afterwards, in the course of time or experience, etc. to be of the quality; turn out to bee.g. He proved (himself) to be a coward.As it happened, my advice proved to be wrong.On the long journey, he proved himself to be an amusing companion.The article has proved (to be) most useful.= The article is proved (to be) most useful. / People proved the article most useful.He proved a real friend.It proved a waste.proof①proof (n.)I believe what you say; I don’t want any proof.②proof (adj.)be ~ againste.g. The tent material is ~ against water.a waterproof coat / a bulletproof car / a soundproof room / a shockproof watch11. as a result : in the ende.g. As a result , he was given a good jog.He broke his leg, as a result, he had to stop his studies.as a result of : because ofe.g. As a result of the war, other events occurred.He is unable to go to work as a result of the fall from his horse.He was late as a result of the snow.12. depress----to lessen the strength of; to sadden; discourage 使不振作；使不景气e.g. Rainy weather always ~es me.Her sad news ~ed me all day.depressed 作表语e.g. Since her baby died, she has been so depressed that she cries at the least thing.He recalled his last evening with his fiancée whenever he felt depressed.depressing 作定语e.g. a depressing weathera depressing incident on my way to work13. survive①survive (v.)a) vi. ---to continue to live, esp. after coming close to deathe.g. Of those wounded in the battle, only 3 ~d.He is the only man who ~d after the explosion.b) vt. ---to continue to live after 经过…活下来，幸存e.g. Only a few soldiers ~d the battle.Only ten of the crew ~d the shipwreck.Did anyone ~ the earthquake?vt. ---live longer thane.g. The man ~d his sister by 3 years.Insurance statistics show that most wives ~ their husbands.②survival (n.)a) [u] ---survivinge.g. The ~ of even a few made people grateful to God.His ~ is still uncertain; he has been very badly hurt and may die.b) [c] ---a person or sth. that remains or survivese.g. The old man is a ~ from my great-grand mother’s days.③survivor (n.) (of accidents)e.g. There were 2 ~s of the air crash.This man was one of the ~s from the earthquake.14. moustache : on the upper lipcf. beard : hair of lower part of the face (excluding moustache)15. a freckled forehead = a forehead with freckles onfreckle (v./n.)e.g. Some people freckle more easily than others.16. look at sb. with an air of …/ do sth. with an air of…air: appearance; mannere.g. He speaks with an air of importance / success / surprise / sadness / excitement.17. to do one’s shoelaces = fasten one’s shoelaces18. smell (smelled / smelt)①vt. e.g. She smelled the meat to see if it was fresh.②vi. e.g. We smell with our nose.③link v. e.g. The dish smells good④vi. ~ of…有…的味道e.g. Your breath smells of brand.The air smelled of paint.19. go: to become (by a natural change, or by changing on purpose)(自然的或故意的)变成，使变成e.g. She is going grey. / Her hair is going grey.The milk went sour.He has gone mad/ blind.He went white with anger.20. proceed①proceed (vi.): go forward to 前进~ to sth.e.g. Please ~ to the next light, and then turn left.Let’s ~ to business. 着手工作Don’t stop! Proceed please.~ to do sth. = begin and continue to do sth.e.g. He ~ed to tell me the matter.After drinking a cup of tea, mother ~ed to cook the dinner.As soon as he came in, he ~ed to tell us all his troubles.~ with = go on withe.g. Let’s ~ with our workThe politician waited for the applause to die down and ~ed with his speech.~ from = result frome.g. The whole trouble ~ed from a misunderstanding.~ against = charge against 控告②proceeding (n.) 活动，进行情况e.g. He wrote an account of the ~s of the meeting.21. GSC = GCE (General Certificate of Education)CSE = Certificate of Secondary Education.Both are public examinations standardized to give national comparability. GCE is usually intended for pupil of an academic basis, usually those who want to enter a university.22. vital (adj.)①connected with livingHe was wounded in a vital part.②supreme, very important23. attach①attach (v.)a) fasten or join one thing to another~ sth. toe.g. I will ~ a label to my luggage.He ~ed a document to the letter.There is a house with a garage ~ed.There is a middle school ~ed to our university.b) consider to have; connect with~ importance / significance toe.g. Do you ~ much importance to what he said?The old man did not ~ much importance to the question.They ~ed significance to his speech.be /become attached to 变得喜欢, 依恋, 对…有感情e.g. She is ~ed to her home.Are you ~ed to this university?②attachment (n.)a) connectione.g. The ~ of a machine is easy.b) part attached 附件e.g. A machine has many~s.c) attached feelingse.g. She has a great ~ to her sister.Do you have a great ~ to your mother?24. singular①singular (adj.)a) uncommon, strangee.g. Don’t make yourself so ~ in your dress. 不要穿奇装异服b) outstanding 非凡的e.g. He is a man of ~ courage.c) (gram.) of the form used in speaking or writing of one person or thing.②singularly adv.)a) particularly 极b) strangely; in an unusual way奇异地③singularity (n.)a) strangeness [u]b) sth. unusual or sth. strange [c]25 .consist of: be made up of (no passive)paraphrase: Our class is made up of 28 students.This book has 18 units.26. range from …to : vary between limitse.g. Prices for books range from 10 cents to 100 dollars.The age of the students ranges from 22 to 25.range in … from…to…e.g. The books range in prices from 10 cents to 100 dollars.The students range in age from 22 to 25.27. appal①appal (U.S. also appall )----fill with fear or terror, shock deeply 惊吓，使惊吓~ sb.e.g. The news appalled me.be appalled at / by…: be shocked bye.g. I was appalled at the news.②appalling (adj.)---terrible; cripple, bad 蹩脚，糟糕e.g. The news was ~.Mary is an ~ cook.③appallingly (adv.): terribly; verye.g. The figures of employment are ~ high.We got ~ hungry working so late.28. split (split, split, splitting)①(vt.) divide into partse.g. They ~ the money between themselves.Let’s ~ the cost of the dinner party.The robbers ~ the loot evenly.The family agreed to ~ the estate.He ~ (up) the book into 5 chapters.I will ~ up the class into groups.②(vt./vi.) break or cause to break, be brokene.g. When he bent over, he ~ his pants.The water pressure ~ the pipe.The lightening ~ the tree.Don’t ~ our organization.Some kinds of wood ~ easily.The board ~ as I sawed it.~ on: give away the secret to one’s disadvantage 告密~ on sth. / sb. to sb. 向…告…的密e.g. The boy ~ on me to the teacher.Don’t ~ on him to his father.~ one’s sides (with laughter): laugh with movements of the sidese.g. They ~ their sides at the joke.I ~ my sides at the story.a ~ second: a brief instant of timee.g. It happened in a ~ second.a splitting headache: so severe that it feels that one’ head may crack29. incompetent: adj. not qualified or able 不合格的,不胜任的be competent/ incompetent at sth./ doing sth.e.g. He is incompetent at his job.He is incompetent at working with his hands.be competent / incompetent to do sth.e.g. He is quite incompetent to be the leader.be competent/ incompetent at sth.e.g. He is competent at all the subjects.be competent/ incompetent for sth. 称职/不称职e.g. He is competent for his job.be competent/ incompetent as sth.e.g. He is competent as a teacher.incompetence (n.)e.g. He was dismissed for incompetence.30. constitute①constitute (v.)a) form/ make up/ be 构成e.g. The best of 3 games ~s a win. 三盘两胜How many departments ~ our university?b) establish建立（制度/机构）e.g. We must ~ good traffic regulations.②constitution (n.)a) the way in which sth. is made upe.g. The ~ of a society is not simple.b) the body of laws and principles according to which a country is governed 宪法e.g. According to American Constitution, presidential elections are held every 4 years.c) the general condition of a person’s body or mind 身心状况（体质）e.g. The old man still has a strong ~.John has a very healthy ~.31. ultimate (adj.)①(not formal) greatest, after which no other can be considerede.g. Ultimate authority is exercised by the King.He has done stupid things before, but to look for the escaping gas with a match was the ~ silliness.②last or farthest distance, being at the end or happening at the ende.g. Becoming president is his ~ goal.Their ~ victory is not in question.32. indignity (antonym: dignity)e.g. The indignity of having to say sorry to a woman is unbearable.the indignities of an illness: such as being washed and dressed by someone else33. diffidentthe antonym of diffident: confidentdiffidence (n.)34. the last straw/ the straw tha t breaks the camel’s back---something in addition to a set of troubles which makes them at lasttoo much to bear 经过一系列打击或不愉快的事又无法忍受的事e.g. The hotel was expensive, the food was poor and the bad weather was the last straw.He couldn’t get the job, being laughed at was the last straw.35. prospect (n.)①[u] reasonable hope of something happening; expectation. 希望，展望，可能性e.g. There is not much ~ of my being able to see you before next month.There is no ~ of my success.②something that is expected 前景，展望之物e.g. Our ~s about the play are good.There is a ~ of a change in the weather.36. leisure①leisure (n.)可作定语“空闲的”e.g. ~ time/ occupationsat leisurea)作表语，“有空”= be freee.g. I only meant that when you are at leisure, you might think the matter over.I am quite at leisure if you want me to do anything for you.b)作状语，“从容地，慢慢地”=without haste; leisurely(adv.)e.g. I put off writing to you till today that I might write at leisure.The doctor directed me to live more leisurely.We talked as we walked leisurely along.②leisurely (adj.)e.g. We had a leisurely meal.He did everything in a leisurely manner.Grammar: 评论性的分词短语(comment participial phrase)A comment participial phrase is used in a sentence as a comment about the statement. In such a structure, the doer of the verb in the phrase is not identical with the subject of the sentence. This is how a comment participial phrase is different from one which acts as an adverbial in the sentence.e.g. Considering the amount she paid, the cat was dear in more ways than one.(If we consider the amount she paid, the cat was dear in more ways than one.)Judging from what you say, he has done his best.(If we judge from what you say, we can find that he has done his best.)There are only 10 apples here, not counting the bad ones.(If we don’t count the bad apples, there are only 10 apples here.)III. Questions on Text I1.What advertisement did the writer see in the newspaper?2.Why did he apply for the job?3.Was he confident that he would get the job?4.When did he receive a reply?5.What was the letter about?6.How was the journey to the school?7.Did the writer like the environment of the school?8.Did the writer like the headmaster?9.Was the headmaster’s appearance pleasant? What did he look like?10.Did the headmaster show his hospitality to the writer? What was the headmaster’s attitudetowards the writer when he saw him?11.Where did the headmaster interview the writer?12.Was the room pleasant?13.What questions did the headmaster ask the writer?14.Did the writer think games were very important in children’s education? How about theheadmaster?15.Did the writer think he had much in common with the headmaster?16.Did the writer get the job? How do you know?17.Was there a playground in the school? Where do the children play?18.Did the writer like the teaching set-up? Why?19.How much was the salary that the headmaster offered to the writer?20.What was the worst thing of the job? Why?IV. Translation1．要一个人专心致力于一份枯燥无味的工作是很艰难的。

Direct Solvers for Sparse MatricesX.Li September2006 Direct solvers for sparse matrices involve much more complicated algorithms than for dense matri-ces.The main complication is due to the need for efficient handling thefill-in in the factors L and U.A typical sparse solver consists of four distinct steps as opposed to two in the dense case:1.An ordering step that reorders the rows and columns such that the factors suffer littlefill,orthat the matrix has special structure such as block triangular form.2.An analysis step or symbolic factorization that determines the nonzero structures of thefactors and create suitable data structures for the factors.3.Numerical factorization that computes the L and U factors.4.A solve step that performs forward and back substitution using the factors.There is a vast variety of algorithms associated with each step.The review papers by Duff[14] (see also[13,Chapter6])and Heath et al.[25]can serve as excellent reference of various algorithms. Usually steps1and2involve only the graphs of the matrices,and hence only integer operations. Steps3and4involvefloating-point operations.Step3is usually the most time-consuming part, whereas step4is about an order of magnitude faster.The algorithm used in step1is quite independent of that used in step3.But the algorithm in step2is often closely related to that of step3.In a solver for the simplest systems,i.e.,symmetric and positive definite systems,the four steps can be well separated.For the most general unsymmetric systems,the solver may combine steps2and3(e.g.SuperLU)or even combine steps1,2and3(e.g.UMFPACK)so that the numerical values also play a role in determining the elimination order.In the past10years,many new algorithms and software have emerged which exploit new architectural features,such as memory hierarchy and parallelism.In Table1,we compose a rather comprehensive list of sparse direct solvers.It is most convenient to organize the software in three categories:the software for serial machines,the software for SMPs,and the software for distributed memory parallel machines.Fair to say,there is no single algorithm or software that is best for all types of linear systems. Some software is targeted for special matrices such as symmetric and positive definite,some is targeted for the most general cases.This is reflected in column3of the table,“Scope”.Even for the same scope,the software may decide to use a particular algorithm or implementation technique, which is better for certain applications but not for others.In column2,“Technique”,we give a high level algorithmic description.For a review of the distinctions between left-looking,right-looking, and multifrontal and their implications on performance,we refer the reader to the papers by Heath et al.[25]and Rothberg[31].Sometimes the best(or only)software is not in public domain,but available commercially or in research prototypes.This is reflected this in column4,“Contact”, which could be the name of a company,or the name of the author of the research code.In the context of shift-and-invert spectral transformation for eigensystem analysis,we need to factorize A−σI,where A isfixed.Therefore,the nonzero structure of A−σI isfixed.Furthermore, for the same shiftσ,it is common to solve many systems with the same matrix and different right-hand sides.(in which case the solve cost can be comparable to factorization cost.)It is reasonable to spend a little more time in steps1and2but speed up steps3and4.That is,one can try different ordering schemes and estimate the costs of numerical factorization and solution based on symbolic factorization,and use the best ordering.For instance,in computing the SVD,one has1Code Technique Scope ContactSerial platformsCHOLMOD Left-looking SPD Davis[8] MA57Multifrontal Sym HSL[17] MA41Multifrontal Sym-pat HSL[1] MA42Frontal Unsym HSL[18] MA67Multifrontal Sym HSL[15] MA48Right-looking Unsym HSL[16] Oblio Left/right/Multifr.sym,out-core Dobrian[12] SPARSE Right-looking Unsym Kundert[27] SPARSPAK Left-looking SPD,Unsym,QR George et al.[20] SPOOLES Left-looking Sym,Sym-pat,QR Ashcraft[5] SuperLLT Left-looking SPD Ng[30] SuperLU Left-looking Unsym Li[10] UMFPACK Multifrontal Unsym Davis[9] Shared memory parallel machinesBCSLIB-EXT Multifrontal Sym,Unsym,QR Ashcraft et al.[6] Cholesky Left-looking SPD Rothberg[33] DMF Multifrontal Sym Lucas[29] MA41Multifrontal Sym-pat HSL[4] MA49Multifrontal QR HSL[3] PanelLLT Left-looking SPD Ng[23] PARASPAR Right-looking Unsym Zlatev[34] PARDISO Left-right looking Sym-pat Schenk[32] SPOOLES Left-looking Sym,Sym-pat Ashcraft[5] SuperLU MT Left-looking Unsym Li[11] TAUCS Left/Multifr.Sym,Unsym,out-core Toledo[7] WSMP Multifrontal SPD,Unsym Gupta[24] Distributed memory parallel machinesDMF Multifrontal Sym Lucas[29] DSCPACK Multifrontal SPD Raghavan[26] MUMPS Multifrontal Sym,Sym-pat Amestoy[2] PaStiX Left-right looking∗SPD CEA[21] PSPASES Multifrontal SPD Gupta[22] SPOOLES Left-looking Sym,Sym-pat,QR Ashcraft[5] SuperLU DIST Right-looking Unsym Li[28] S+Right-looking†Unsym Yang[19] WSMP Multifrontal SPD,Unsym Gupta[24] Table1:Software to solve sparse linear systems using direct methods.∗In spite of the title of the paper†Uses QR storage to statically accommodate any LUfill-inAbbreviations used in the table:SPD=symmetric and positive definiteSym=symmetric and may be indefiniteSym-pat=symmetric nonzero pattern but unsymmetric valuesUnsym=unsymmetricHSL=Harwell Subroutine Library:/Activity/HSL2the choice between shift-and-invert on AA∗,A∗A,and0AA∗0,all of which can have ratherdifferent factorization costs.Some solvers have the ordering schemes built in,but others do not.It is also possible that the built-in ordering schemes are not the best for the target applications.It is sometimes better to substitute an external ordering scheme for the built-in one.Many solvers provide well-defined inter-faces so that the user can make this substitution easily.One should read the solver documentation to see how to do this,as well as tofind out the recommended ordering methods. References[1]P.R.Amestoy and I.S.Duff.Vectorization of a multiprocessor multifrontal code.Int.J.ofSupercomputer Applics.,3:41–59,1989.[2]P.R.Amestoy,I.S.Duff,J.-Y.L’Excellent,and J.Koster.A fully asynchronous multifrontalsolver using distributed dynamic scheduling.SIAM Journal on Matrix Analysis and Applica-tions,23(1):15–41,2001.[3]P.R.Amestoy,I.S.Duff,and C.Puglisi.Multifrontal QR factorization in a multiprocessorenvironment.Numer.Linear Algebra Appl.,3(4):275–300,1996.[4]Patrick R.Amestoy and Iain S.Duff.Memory management issues in sparse multifrontalmethods on multiprocessors.Int.J.Supercomputer Applics,7:64–82,1993.[5]C.Ashcraft and R.Grimes.SPOOLES:An object-oriented sparse matrix library.In Proceed-ings of the Ninth SIAM Conference on Parallel Processing,1999.(/ linalg/spooles).[6]BASLIB-EXT:Sparse matrix software./phantom/bcslib-ext/.[7]D.Chen,V.Rotkin,and S.Toledo.TAUCS:A Library of Sparse Linear Solvers,Tel-AvivUnivesity.(http://www.tau.ac.il/~stoledo/taucs/).[8]Y.Chen,T.A.Davis,W.W.Hager,and S.Rajamanickam.Algorithm8xx:CHOLMOD,supernodal sparse Cholesky factorization and update/downdate.Technical Report TR-2006-005,University of Florida,2006.(/research/sparse/cholmod/).[9]T.A.Davis.Algorithm832:UMFPACK V4.3,an unsymmetric-pattern multifrontal methodwith a column pre-ordering strategy.ACM Trans.Mathematical Software,30(2):196–199,June 2004.(/research/sparse/umfpack/).[10]James W.Demmel,Stanley C.Eisenstat,John R.Gilbert,Xiaoye S.Li,and Joseph W.H.Liu.A supernodal approach to sparse partial pivoting.SIAM J.Matrix Anal.Appl.,20(3):720–755,1999.(/~xiaoye/SuperLU).[11]James W.Demmel,John R.Gilbert,and Xiaoye S.Li.An asynchronous parallel supernodalalgorithm for sparse Gaussian elimination.SIAM J.Matrix Anal.Appl.,20(4):915–952,1999.(/~xiaoye/SuperLU).[12]F.Dobrian and A.Pothen.Oblio:a sparse direct solver library for serial and parallel compu-tations.Technical report,Old Dominion University,2000.3[13]J.J.Dongarra,I.S.Duff,D.C.Sorensen,and H.A.van der Vorst.Numerical Linear Algebrafor High-Performance Computers.SIAM,Philadelphia,PA,1998.[14]Iain S.Duff.Direct methods.Technical Report RAL-98-056,Rutherford Appleton Laboratory,1998.[15]I.S Duffand J.K.Reid.MA47,a Fortran code for direct solution of indefinite sparse symmetriclinear systems.Technical Report RAL-95-001,Rutherford Appleton Laboratory,1995. [16]I.S Duffand J.K.Reid.The design of MA48,a code for the direct solution of sparse unsym-metric linear systems of equations.ACM Trans.Math.Software,22:187–226,1996.[17]I.S Duffand J.K Reid.The multifrontal solution of indefinite sparse symmetric linear equations.ACM Transactions on Mathematical Software,9(3):302–325,September1983.[18]I.S Duffand J.A.Scott.The design of a new frontal code for solving sparse unsymmetricsystems.ACM Trans.Math.Software,22(1):30–45,1996.[19]Cong Fu,Xiangmin Jiao,and Tao Yang.Efficient sparse LU factorization with partial pivotingon distributed memory architectures.IEEE Trans.Parallel and Distributed Systems,9(2):109–125,1998.(/research/S+).[20]A.George and puter Solution of Large Sparse Positive Definite Systems.Prentice-Hall Inc.,Englewood Cliffs,New Jersey,1981.(jageorge@sparse1.uwaterloo.ca).[21]D.Goudin,P.Henon,F.Pellegrini,P.Ramet,and J.Roman.Parallel Sparse matriX pack-age,LaBRI,Universit´e Bordeaux I,Talence,France.(bri.fr/Perso/~ramet/ pastix/).[22]A.Gupta,G.Karypis,and V.Kumar.Highly scalable parallel algorithms for sparse matrixfactorization.IEEE Trans.Parallel and Distributed Systems,8:502–520,1997.(http://www./~mjoshi/pspases).[23]A.Gupta,E.Rothberg,E.Ng,and B.W.Peyton.Parallel sparse Cholesky factorizationalgorithms for shared-memory multiprocessor systems.In R.Vichnevetsky,D.Knight,andG.Richter,editors,Advances in Computer Methods for Partial Differential Equations–VII,pages622–628.IMACS,1992.(egng@).[24]Anshul Gupta.WSMP:Watson Sparse Matrix Package.IBM T.J.Watson Research Center,Yorktown Heights.(/~agupta/wsmp.html).[25]M.Heath,E.Ng,and B.Peyton.Parallel algorithms for sparse linear systems.SIAM Review,33:420–460,1991.[26]Michael T.Heath and Padma Raghavan.Performance of a fully parallel sparse solver.Int.J.Supercomputer Applications,11(1):49–64,1997./~raghavan. [27]Kenneth Kundert.Sparse matrix techniques.In Albert Ruehli,editor,Circuit Analysis,Simulation and Design.North-Holland,1986.(/sparse).[28]Xiaoye S.Li and James W.Demmel.SuperLU DIST:A scalable distributed-memory sparsedirect solver for unsymmetric linear systems.ACM Trans.Mathematical Software,29(2):110–140,June2003.4[29]Robert Lucas.Private communication(rflucas@),2000.[30]Esmond G.Ng and Barry W.Peyton.Block sparse Cholesky algorithms on advanced unipro-cessor computers.SIAM put.,14(5):1034–1056,September1993.(egng@).[31]E.Rothberg.Exploiting the memory hierarchy in sequential and parallel sparse Choleskyfactorization.PhD thesis,Dept.of Computer Science,Stanford University,December1992.[32]O.Schenk,K.G¨a rtner,and W.Fichtner.Efficient sparse LU factorization with left–rightlooking strategy on shared memory multiprocessors.BIT,40(1):158–176,2000.[33]J.P.Singh,W-D.Webber,and A.Gupta.Splash:Stanford parallel applications for shared-puter Architecture News,20(1):5–44,1992.(/ apps/SPLASH).[34]Z.Zlatev,J.Wa´s niewski,P.C.Hansen,and Tz.Ostromsky.PARASPAR:a package forthe solution of large linear algebraic equations on parallel computers with shared memory.Technical Report95-10,Technical University of Denmark,September1995.5。

I.J. Education and Management Engineering, 2016, 6, 1-11Published Online November 2016 in MECS ()DOI: 10.5815/ijeme.2016.06.01Available online at /ijemeA Hybrid Approach to Sentiment Analysis of Technical ArticleReviewsBabaljeet Kaur a, Naveen Kumari ba Punjabi University Regional Centre, Mohali, Indiab P unjabi University Regional Centre, Mohali, IndiaAbstractSentiment analysis is similar to opinion mining, which is a popular research problem to search out in the field of NLP. Sentiment analysis determines the perspective of the author and identifies the positive, negative and neutral reviews. It provides the reviews or opinions of people‟s on text, article and product which can be positive, negative or neutral. Reviews on the different websites, social networking sites is an important source to collect the information regarding various brands of product and new features in technology (e.g. Windows, Mobiles). During the sentiment analysis various classification tools within the NLP are used to find out the positivity and negativity of reviews or comments. The paper presents a length aware hybrid approach to analyses the reviews either as positive or negative and present approach is tested on SuperFetch data set. The present approach is a combination of both supervised machine learning techniques that are Support Vector Machine and K-Nearest Neighbor in which SVM is working great for large size review and KNN is working best for small size review.Index Terms: Sentiment analysis, SVM, KNN, SuperFetch review.© 2016 Published by MECS Publisher. Selection and/or peer review under responsibility of the Research Association of Modern Education and Computer Science.1. IntroductionWith the rapid growth of online websites and social networking sites, everybody express their views regarding article, product on these sites [12]. The internet or web becomes a necessity in every person‟s life. On each day everybody is using online websites to express their reviews or comments, providing feedback and asking a question and other business or companies can take decision according to feedback. Sentiment analysis depicts the mood or feeling of individual about various entities and their attributes [11]. Sentiment analysis is a task to express people‟s opinions about an entity, article and product etc.In natural language processing (NLP) different machine learning approaches are utilized to determine the * Corresponding author:E-mail address: babbaljeet001@, naveencse2k4@sentiments of a huge amount of products, services, and text etc. Sentiment analysis provides many facilities to implement new applications which have a major role in business, medical, data mining and evaluate the feedback of different latest and old products.The reviews or comments of individuals in sentiment analysis can be classified into various ways like positive and negative reviews. Positive reviews have high polarity than the negative reviews. According to the reviews by various experts, one can easily find out the quality of technique. The analysis of sentiment work up with few difficulties: Among different things, it must be resolved whether document or segment thereof is subjective or objective and whether or not the sentiment communicated is positive or negative.The main steps involved in the research are Pre-processing of text, analysis of data and classification and evaluation of result. On numerous websites, people share their views in the form of …comments‟. These reviews and comments are the main elements that confirm the sentiment of people as these reviews as original. The sentiment analysis on the website is done in three steps: (a) Determine the sentiment expression. (b) Determining the polarity of reviews (positive, negative and neutral). (c) A classification approach is utilized to classify the sentiments on the website [2]. It has increased much consideration as of late and studies individual‟s emotio ns towards certain substances [1]. Sentiment analysis deal with several challenges. The description of element considered in this area. It is not a cup of tea. For example sentiments about any product can be determined by positive and negative opinions of people on the product.In the present work find out the reviews of a specialized article (e.g. SuperFetch). Determine the positive and negative survey of individuals in this article. On the basis of the reviews of the expert define how much that article technically sound or not. The positive review concerning article gives high polarity. A combined approach of Support Vector Machine and K-Nearest Neighbor is used to classify the reviews. The rest of the paper is organized as follows: Section 2 describes the some related work done in this field. Section 3 explains the proposed approach in identifying the reviews. Section 4 provides the results obtained in the proposed work along with a brief discussion. Section 5 concludes work done in this research and work to be done in the future.1.1. Applications∙Aid in decision making- Decision making is an integral part of our life. It ranges from “which technology‟s feature to use”, “which product to buy”, “which bank insurance policies to go for”, “which restaurant to go”. Sentiment An alysis (SA) can be utilized to decide and select from the available options on the basis of general opinions expressed by other users.∙Business Strategies- Many of business strategies are being guided with regard to response from the public.The various companies‟ aims to satisfy the requirements and demands of the users, so strategic moves of the companies are driven through public views and opinions. With the world connected via technology events have a global impact; issue/failure on one part of the globe has an impact on opposite corners of the world. Therefore it becomes quite necessary to drive products/services per the general public view point.∙Application as sub-component technology- The sentiment analysis has a major role in enabling technology for different systems. In some websites whenever ads are displayed in the sidebars, it‟s helpful to show ads when relevant positive sentiments are detected and nix the ads once negative sentiments are detected. The question-answering is another area where sentiment analysis may be helpful as opinion oriented question might need completely different treatment [13].∙Application to review related websites- The most common application in within the reviews of costumer products, services, and articles. There are several websites that give reviews related to the article. For example is the website that provides numerous reviews related to mobile phones, Windows, DOS, and Memory management. Summarizing opinions or reviews of the public are an important task today, so before using the latest features comes in the technology, people can aware to make decision regarding the technology means good or not for future [10].2. Related WorkThis section presents a comprehensive literature survey of research related to various sentiment analysis classification approaches. Various sentiment analysis related research papers till date have been studied and their brief is presented in this section.Mudinas et al. The paper presented a combination of both lexicon and learning based approaches for sentiment analysis. The paper mainly focused on the anatomy of pSenti- a concept level sentiment analysis and combined approach are tested on two types of data set including CNET software reviews and IMDB movie reviews. In the concept level sentiment analysis, pSenti is originated by combining both lexicon and learning based techniques. The supervised machine learning part is not simply responsible for smaller tasks such as adjusting sentiment values or finding additional sentiment words, however is really responsible for evaluating all the ingredients of sentiment process as well as semantic rules utilized to get the final output. The main advantage of hybrid approach employing a lexicon/machine learning techniques is to achieve more effective of each word- stability also as readability from a correctly designed lexicon, and also the high accuracy of the popular supervised machine learning approach. The of performance of sentiment analysis mainly depends on the sentiment of comments or reviews; if there‟s a clear separation between positive and negative value distribution, the lexicon based technique would work best, otherwise machine learning would well raise the performance. The result shows that even without subjective detection, the combined approach pSenti can achieve 82.50% accuracy that is just slightly below the bag-of-words Support Vector Machine [8].Amira Shoukry et al. The paper presented a sentiment analysis of Egyptian dialect tweets and applied sentence level analysis utilizing a hybrid approach. This approach combines each machine learning (ML) approach using Support Vector Machine (SVM) and semantic orientation (SO) techniques. This approach includes building a classifier utilizing the unigrams, bigrams and trigrams as well as new thebigrams and trigrams), Unigrams produces the best results at threshold 0. It is good to combine the all unigrams, bigrams and trigrams as features to increase the performance of sentiment analysis. The accuracy achieved by Support Vector Machine (SVM) classifier is 82.9% in case of unigrams. When comparing the results of both obtained in ML experiment and semantic orientation, the accuracy (0.806) obtained in SVM learning algorithm greater than achieved using semantic orientation technique‟s accuracy (0.719). After combining both ML and SO accuracy (A) is 0.844 and precision (P) is 0.842, recall (R) is 0.844 and f-measure is 0.842 [3]. Changliang Li et al. The paper introduced a Chinese feeling Treebank over social data because there are not very many resources introduced in sentiment analysis for Chinese and progress is kept down because of expansive, marked corpus and capable models. It identified 13550 marked sentences from movie reviews. The paper introduced a Recursive Neural Deep model (RNDM) figuring out how to sentiment label based on recursive deep learning. The paper considers the sentiment about the sentence and identifying review is positive and negative. For sentence level based sentiment, this model used baselines such as Naïve Bayes, Maximum Entropy and Support Vector Machine. The total accuracy obtained by RNDM is maximum (90.8%) compared to Naïve Bayes (78.65%), Maximum Entropy (87.46%) and Support Vector Machine (84.9%) and RNDM gets the highest performance [7].Grandi et al. In this paper observed that the present sentiment analysis technique is satisfactory for a single entity, but can produce wrong results when with the arrangement of various items. Paper is importing techniques with the help of voting theory and according to the preference aggregation to collect a set of multiple items with high accuracy. The Paper proposed notion of Borda count, which joins public‟sentiment, according to similar comparative preference information and demonstrate this class of standards fulfills various properties which a characteristics understanding in sentiment area. The SP (sentiment preference structure) is used over a set of candidates. Borda always behaves better than random procedure in identifying the winner in complete profile [6].Tan et al. The paper developed sentiment categorization on the basis of Chinese language documents with the size of 1024 documents. The feature selection methods (IG, MI, CHI, and DF) and machine learning methods (K-Nearest Neighbor, Winnow classifier, Centroid classifier, SVM and Naïve Bayes) are conducted to find out the Chinese language sentiment. The results achieved suggest that IG performance best in sentiment phrases selection and Support Vector Machine (SVM) for sentiment classification and it contains the reviews from three different areas like movie, education and house, also found that sentiment classifier mainly dependent on the topics. The dataset contains 507 documents related to education, 248 to house and 266 documents related to the movie. The Precision, Recall and F-Measure parameters are used to evaluate the performance of all the classifiers [4].Guerrero et al.The paper analyzed people‟s emotions, reviews, feelings about the product, services and organizations. Different tools and techniques are implemented for sentiment analysis. Compared some free access web services, analyze their capabilities and score to different pieces of text according to their sentiments. Machine learning and lexicon based approaches are utilized. 15 web services computed for the purpose of sentiment analysis. Some of these services belong to personal companies, however, still they allow to restricted free access to their functionalities and others are completely free services. This fact is interesting to the users who want to incorporate sentiment analysis capability among their own platforms while not having to develop their own algorithms; thus, these tools are particularly interesting for researching purposes and fast prototyping. Besides, because of the actual fact that the chosen services can work as web services, inclusion of them into any platform is basically simple. From the results produced, the services like Alchemy and Semantria might be taken into consideration for any kind of text. Sentiment analysis could also be extremely interesting to the user if the analyzed text is quite high and you would like to classify them [5].3. MethodologyThis section provides the information about the present hybrid approach and various parameters that are used to evaluate the results and also the various steps that are taken into consideration to complete the proposed work.3.1. Data setThe data set is a collection of technical article reviews (e.g. SuperFetch), which is a memory management feature find out in various Windows. The reviews of the SuperFetch have collected from the websites such as , and also from the various professionals of Universities or Colleges. These websites provide detailed information related to positive and negative reviews of different articles given by various professionals. The collected reviews are stored in an excel file, then directly fetch to the MS SQL server for further processing. All the collected reviews are divided into three parts like 50, 100 and 120 to evaluate the result.3.2. PreprocessingThe data preprocessing or cleaning step is more important in sentiment analysis. It is the process of preparing the text before the classification. The preprocessing includes the removal of extra words which are not helpful for identification of sentiments. In case of word level, different words in sentence don‟t have any impact if treated as one dimensional in the classification process [9]. Including those words in the text make the classification more tough. So, there is a process to overcome the noise level in text should help increase the performance of the classifier is called preprocessing. The main step in this process is the removal of stop words is described below. The preprocessing step mainly includes the removal of stop words. In Information Retrieval, stop words removal is a common process to get rid of words that are extremely common which don‟t add substantial value to the classification process. These common words like is, he, it, a, an, and the y‟recollectively known as stop words. Because the inclusion of these words in a review does not give any useful information, they are removed.Fig.1. Flow Chart of Removal of Stop Words3.3. Proposed ApproachThe proposed hybrid approach is a combination of supervised machine learning approaches: SVM and KNN. The present hybrid approach is utilized to classify the reviews of SuperFetch in which Support Vector Machine classifier is working well to evaluate the large size reviews and KNN is working best to evaluate the small size review. Both approaches are combined in such a way that to increase the effectiveness of the present hybrid approach.∙Support Vector MachineSupport Vector Machine is well known for their good generalization performance and has been applied to many sentiment analysis problems. Training and testing data are involved in classification task, which consists of data instances. One class label and various features are contained in each instance in the training set. In basic form, a SVM learns to find a hyper plane that separates both positive and negative examples maximum margins. The Support Vector Machine classifier is most effective classifier to increase the performance using large size sentence during the classification.∙K-Nearest NeighborIt is the classifier that relies on the labels of category and finds a k-nearest neighbor during the classification. Assign the weight to the neighbor based on their distance from the query point. KNN is an effective classifier to classify the small size sentence. The Euclidean distance equation is a main alternative to the similarity measure.∑√ (1)3.4. Performance Evaluation MetricsFinally the result is evaluated using the Accuracy, F-Measure, Precision and Recall. These four parameters are used to evaluate the performance of the classifiers.4. Results and DiscussionThe SuperFetch Reviews is considered as data set to test the performance of the present hybrid approach. The reviews are collected from the websites as well as from the various professionals. The present hybrid approach is tested on a SuperFetch data set in which three cases are considered to evaluate the results. All cases were carried out in Visual Studio 2010 are discussed in this section.Reviews are classified as positive and negative by the hybrid method. The effectiveness of the hybrid method is determined by the following parameters.∙Accuracy = (2) Where;TP is true positiveTN is true negativeFP is false positiveFN is false negative∙Precision = (3) ∙Recall = (4) ∙F-Measure = (5)Fig.2. Accuracy, Positivity and Negativity in case of 50 ReviewsThe Figure 2 shows the graphs which represent the accuracy, positivity and negativity in case of using 50 reviews. The positivity comes in the 50 reviews is much greater than the negativity.Fig.3. Precision, Recall and F-Measure in Case of 50 ReviewsThe figure 3 indicates the Precision, Recall and F-Measure in case of 50 reviews. The value of Recall is greater than the Precision and F-Measure.Fig.4. Accuracy, Positivity and Negativity in Case of 100 ReviewsFig.5. Precision, Recall and F-Measure in Case of 100 ReviewsThe figure 5 and 6 show that the result evaluation using 100 reviews. The results achieved using 100 reviews greater than using 50 reviews in Precision, Accuracy, F-Measure and Positivity. So result evaluation using 100 reviews quite effective than the result evaluation using 50 reviews.Fig.6. Accuracy, Positivity and Negativity in Case of 120 ReviewsFig.7. Precision, Recall and F-Measure in Case of 120 ReviewsThe figures 7 and 8 represent result evaluations using 120 reviews. The Precision, Recall and F-Measure in 120 reviews are at almost same level. The accuracy achieved is better than the previous two cases. So, the hybrid approach is more effective in this case mainly in terms of Accuracy.Table 1. Result Table for All CasesParameters 50 Reviews 100 Reviews 120 ReviewsPrecision 0.84 0.89 0.94Recall 0.97 0.96 0.96Accuracy F-Measure Positive% Negative% 84.310.9062.008.0087.130.9270.005.0090.740.9568.335.005. Conclusion and Future WorkPresent work concluded that the combination of K-Nearest Neighbor and Support Vector Machine produced better results on the basis of Accuracy, Precision, Recall and F-Measure. K-Nearest Neighbor improved the performance in the case of small reviews and Support Vector Machine improved the performance in case of large reviews are working as a single hybrid approach.There are two more parameters positivity and negativity are evaluated that shows most of the reviewers have positive thoughts regarding SuperFetch and the negativity percentage is very less, the remaining reviews are considered as neutral. So the results indicate that SuperFetch is a good feature in the memory management system. Future work includes the comparison of the present technique with existing techniques.AcknowledgementsI am grateful to my guide Assistant Professor Mrs. Naveen Kumari for all help and valuable suggestion provided by her during the study and special thanks to Associate Professor Mr. Michael Swift who help me in collection of reviews.References[1]X. Fang, and J. Zhan, “Sentiment analysis using product review data,” Journal of Big Data, pp.1-14,2015.[2]Y. Sharma, V. Mangat, and M. Kaur, “Sentiment analysis and Opinion mining,” In Proceedings of 21stIRF International conference, March-8, 2015, Pune, India, pp. 35-38.[3] A. Shoukry, and A. Rafea, “A Hybrid Approach for Sentiment Classification of Egyptian DialectTweets,” In Proceedings of First International Conference on Arabic Computational Linguistics, 2015, pp. 78-85.[4]S. Tan, and J. Zhang, “An empirical study of sentiment analysis for Chinese documents,” Expert systemswith applications 34 (2008), pp. 2622-2629.[5]J. S. Guerrero, J. A. Olivas, F. P. Romero, and E. H. Viedma, “Sentiment analysis: A review andcomparative analysis of web services,” Information Sciences, pp. 18-38, 2015.[6]U. Grandi, A. Loreggia, F. R ossi, and V. Saraswat, “A Borda count for collective sentiment analysis,”Annals of Mathematics and Artificial Intelligence, 22 October 2015, DOI: 10.1007/s10472-015-9488-0.[7] C. Li, B. Xu, G. Wu, S. He, G. Tian, and H. Hao, “Recursive Dee p Learning for Sentiment Analysisover Social Data,” In Proceedings of International Joint Conferences on Web Intelligence (WI) and Intelligent Agent Technologies (IAT), IEEE/WIC/ACM, 2014, pp. 180-185.[8] A. Mudinas, D.Zhang, and M. Levene, “Combining Lexico n and Learning based Approaches forConcept-Level Sentiment Analysis,” In Proceedings of the First International Workshop on Issues of Sentiment Discovery and Opinion Mining, ACM, August-12, 2012, pp. 1-8.[9] E. Haddi, X. Liu, and Y. Shi, “The Role of text pre-processing in sentiment analysis,” InformationTechnology and Quantitative Managemen t, pp.26-32, 2013.[10]Y. Sh arma, V. Mangat, and M. Kaur, “A Practical Approach to Sentiment Analysis of Hindi Tweets,”In Proceedings of 1st International Conference on Next Generation Computing Technologies (NGCT-2015), Dehradun, India, 4-5 September 2015, IEEE,DOI: 10.1109/NGCT.2015.7375207, pp. 677-680.[11]J. S. Modha, G. S. Pandi, and S. J. Modha, “Automatic sentiment analysis for unstructured data,”International journal of advanced research in computer science and software engineering, vol.3, pp.91-97, December -2013.[12] A. Tripathi, A. Agrawal, and S. K. Rath, “Classification of Sentimental Reviews using Machine learningTechniques,” InProceedings of 3rd International Conference on Recent Trends in Computing, ICRTC, 2015, pp. 821-829.[13] B. Pang, and L. Lee, “Opinion mining and Sentiment analysis,” Foundation and Trends in InformationRetrieval, vol. 2, no. 1-2, pp. 1-135, 2008.A Hybrid Approach to Sentiment Analysis of Technical Article Reviews 11Authors’ ProfilesBabaljeet Kaur has completed her M.Tech in Computer Science and Engineering fromPunjabi University Regional Centre for Information Technology and Management, Mohali,India. Her research interests include Natural Language Processing, Digital Image Processingand Big Data.How to cite this paper: Babaljeet Kaur, Naveen Kumari,"A Hybrid Approach to Sentiment Analysis of Technical Article Reviews", International Journal of Education and Management Engineering(IJEME), Vol.6, No.6, pp.1-11, 2016.DOI: 10.5815/ijeme.2016.06.01。

参考文献(人工智能)曹晖目的：对参考文献整理（包括摘要、读书笔记等），方便以后的使用。

分类：粗分为论文（paper）、教程（tutorial）和文摘（digest）。

0介绍 (1)1系统与综述 (1)2神经网络 (2)3机器学习 (2)3.1联合训练的有效性和可用性分析 (2)3.2文本学习工作的引导 (2)3.3★采用机器学习技术来构造受限领域搜索引擎 (3)3.4联合训练来合并标识数据与未标识数据 (5)3.5在超文本学习中应用统计和关系方法 (5)3.6在关系领域发现测试集合规律性 (6)3.7网页挖掘的一阶学习 (6)3.8从多语种文本数据库中学习单语种语言模型 (6)3.9从因特网中学习以构造知识库 (7)3.10未标识数据在有指导学习中的角色 (8)3.11使用增强学习来有效爬行网页 (8)3.12★文本学习和相关智能A GENTS：综述 (9)3.13★新事件检测和跟踪的学习方法 (15)3.14★信息检索中的机器学习——神经网络，符号学习和遗传算法 (15)3.15用NLP来对用户特征进行机器学习 (15)4模式识别 (16)4.1JA VA中的模式处理 (16)0介绍1系统与综述2神经网络3机器学习3.1 联合训练的有效性和可用性分析标题：Analyzing the Effectiveness and Applicability of Co-training链接：Papers 论文集\AI 人工智能\Machine Learning 机器学习\Analyzing the Effectiveness and Applicability of Co-training.ps作者：Kamal Nigam, Rayid Ghani备注：Kamal Nigam (School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, knigam@)Rayid Ghani (School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213 rayid@)摘要：Recently there has been significant interest in supervised learning algorithms that combine labeled and unlabeled data for text learning tasks. The co-training setting [1] applies todatasets that have a natural separation of their features into two disjoint sets. We demonstrate that when learning from labeled and unlabeled data, algorithms explicitly leveraging a natural independent split of the features outperform algorithms that do not. When a natural split does not exist, co-training algorithms that manufacture a feature split may out-perform algorithms not using a split. These results help explain why co-training algorithms are both discriminativein nature and robust to the assumptions of their embedded classifiers.3.2 文本学习工作的引导标题：Bootstrapping for Text Learning Tasks链接：Papers 论文集\AI 人工智能\Machine Learning 机器学习\Bootstrap for Text Learning Tasks.ps作者：Rosie Jones, Andrew McCallum, Kamal Nigam, Ellen Riloff备注：Rosie Jones (rosie@, 1 School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213)Andrew McCallum (mccallum@, 2 Just Research, 4616 Henry Street, Pittsburgh, PA 15213)Kamal Nigam (knigam@)Ellen Riloff (riloff@, Department of Computer Science, University of Utah, Salt Lake City, UT 84112)摘要：When applying text learning algorithms to complex tasks, it is tedious and expensive to hand-label the large amounts of training data necessary for good performance. This paper presents bootstrapping as an alternative approach to learning from large sets of labeled data. Instead of a large quantity of labeled data, this paper advocates using a small amount of seed information and alarge collection of easily-obtained unlabeled data. Bootstrapping initializes a learner with the seed information; it then iterates, applying the learner to calculate labels for the unlabeled data, and incorporating some of these labels into the training input for the learner. Two case studies of this approach are presented. Bootstrapping for information extraction provides 76% precision for a 250-word dictionary for extracting locations from web pages, when starting with just a few seed locations. Bootstrapping a text classifier from a few keywords per class and a class hierarchy provides accuracy of 66%, a level close to human agreement, when placing computer science research papers into a topic hierarchy. The success of these two examples argues for the strength of the general bootstrapping approach for text learning tasks.3.3 ★采用机器学习技术来构造受限领域搜索引擎标题：Building Domain-specific Search Engines with Machine Learning Techniques链接：Papers 论文集\AI 人工智能\Machine Learning 机器学习\Building Domain-Specific Search Engines with Machine Learning Techniques.ps作者：Andrew McCallum, Kamal Nigam, Jason Rennie, Kristie Seymore备注：Andrew McCallum (mccallum@ , Just Research, 4616 Henry Street Pittsburgh, PA 15213)Kamal Nigam (knigam@ , School of Computer Science, Carnegie Mellon University Pittsburgh, PA 15213)Jason Rennie (jr6b@)Kristie Seymore (kseymore@)摘要：Domain-specific search engines are growing in popularity because they offer increased accuracy and extra functionality not possible with the general, Web-wide search engines. For example, allows complex queries by age-group, size, location and cost over summer camps. Unfortunately these domain-specific search engines are difficult and time-consuming to maintain. This paper proposes the use of machine learning techniques to greatly automate the creation and maintenance of domain-specific search engines. We describe new research in reinforcement learning, information extraction and text classification that enables efficient spidering, identifying informative text segments, and populating topic hierarchies. Using these techniques, we have built a demonstration system: a search engine forcomputer science research papers. It already contains over 50,000 papers and is publicly available at ....采用多项Naive Bayes 文本分类模型。