下载文档原格式
指南小班数学领域目标解读As we interpret the goals of the mathematics domain in the small class guide, it is important to recognize the significance of laying a strong foundation for young learners. 小班指南中对数学领域目标的解读，我们要意识到为年幼学习者打下坚实基础的重要性。
Mathematics at this stage should not only focus on numerical skills, but also emphasize the development of critical thinking, problem-solving abilities, and a positive attitude towards learning. 这个阶段的数学不仅应该注重数字技能，还应该强调批判性思维、问题解决能力以及积极的学习态度的培养。
First and foremost, the goal of mathematics instruction in the small class guide should be to cultivate a love for the subject. 首先，小班指南中数学教育的目标应该是培养对这门学科的热爱。
This can be achieved through engaging and interactive activities that make learning fun and enjoyable. 这可以通过有趣互动的活动来实现，让学习变得有趣和令人愉快。
By creating a positive and supportive learning environment, children will be more inclined to participate actively in mathematical tasks and develop a positive attitude towards the subject. 通过营造积极支持的学习环境，孩子们更倾向于积极参与数学任务，并对这门学科产生积极的态度。
介绍学习策略英语作文Learning Strategies。
Learning a new language can be a challenging and rewarding experience. When it comes to learning English, having effective learning strategies is essential. In this essay, I will discuss some effective learning strategies for studying English.First and foremost, it is important to set clear and achievable goals for learning English. Whether it is to improve speaking, listening, reading, or writing skills, having a clear goal will help you stay focused and motivated. Setting short-term and long-term goals can also help you track your progress and stay on track.Another important learning strategy is to practice English as much as possible. This can be done through various activities such as watching English movies, listening to English songs, reading English books, andspeaking with native English speakers. The more you practice, the more confident and proficient you will become in using the language.In addition to practicing, it is also important tobuild a strong foundation in English grammar and vocabulary. This can be achieved through regular study and review of grammar rules and vocabulary words. Flashcards, word games, and language learning apps can be helpful tools for expanding your vocabulary and improving your grammar skills.Furthermore, finding a study method that works best for you is crucial. Some people learn best through visual aids, such as videos and pictures, while others may preferauditory learning through listening to lectures and conversations. Experimenting with different study methods and finding what works best for you can greatly enhanceyour learning experience.Moreover, seeking feedback from teachers, tutors, or language exchange partners can be beneficial for improving your English skills. Constructive feedback can help youidentify areas for improvement and provide you withvaluable insights on how to progress further.Lastly, staying motivated and persistent is key to successful language learning. It is important to stay positive and not be discouraged by setbacks or challenges. Surrounding yourself with a supportive and motivating environment can also help keep you on track and motivatedto continue learning.In conclusion, learning English requires dedication, practice, and effective learning strategies. By settingclear goals, practicing regularly, building a strong foundation in grammar and vocabulary, finding a study method that works for you, seeking feedback, and staying motivated, you can make significant progress in yourEnglish language skills. With the right learning strategies, you can achieve fluency and proficiency in English.。
Attempt to Enter Analysis into AreaIntroductionIn today’s rapidly changing world, the ability to analyze and interpret data has become a crucial skill. The field of analysis encompasses a wide range of disciplines, from business and finance to healthcare and social sciences. The ability to effectively analyze data allows us to gain valuable insights, make informed decisions, and drive innovation. In this article, we will explore the importance of analysis and discuss various approaches to entering this area.The Importance of AnalysisAnalysis plays a pivotal role in many aspects of our lives. It allows us to understand complex phenomena, identify patterns, and make predictions based on available data. Whether it is analyzing financial markets to inform investment decisions or studying consumer behavior to develop effective marketing strategies, analysis provides us with the tools necessary for success.In the business world, analysis helps organizations optimize their operations by identifying inefficiencies and areas for improvement. By analyzing sales data, companies can identify trends and adjust their marketing strategies accordingly. Analysis also enables businesses to forecast demand, manage inventory effectively, and optimize pricing strategies.In healthcare, analysis is crucial for understanding disease patterns, identifying risk factors, and developing effective treatment plans. By analyzing patient data and medical records, healthcare professionals can detect early warning signs of diseases and intervene before they become serious. Analysis also plays a vital role in drug discovery and development by identifying potential therapeutic targets.In the social sciences, analysis helps us understand human behavior and societal trends. Through surveys, interviews, and statistical analysis, researchers can gain insights into various aspects of society such as education levels, income distribution, or voting patterns. This information is invaluable for policymakers in designing effective public policies.Approaches to Entering the Field of AnalysisEntering the field of analysis requires a combination of technicalskills and domain knowledge. Here are some approaches that can help you get started:1. Acquire Technical SkillsTo excel in analysis, it is essential to have a strong foundation in mathematics and statistics. Courses in data analysis, machine learning, and programming languages such as Python or R can provide you with the necessary technical skills. Online platforms like Coursera, edX, or DataCamp offer a wide range of courses on these topics.2. Gain Domain KnowledgeWhile technical skills are important, having domain knowledge in a specific field can give you a competitive edge. Whether it is finance, healthcare, or marketing, understanding the nuances of the industry you wish to analyze will help you ask relevant questions and derive meaningful insights from the data.3. Practice on Real-World DataTheory alone is not enough to become proficient in analysis. To truly understand the challenges and complexities of real-world data analysis, it is crucial to practice on real datasets. Many organizations provide open datasets for analysis purposes, such as Kaggle or UCI Machine Learning Repository. By working on these datasets, you can hone your skills and gain practical experience.4. Collaborate and NetworkCollaborating with others who share your passion for analysis can be immensely beneficial. Joining online communities or attending industry conferences allows you to learn from experts in the field, exchange ideas, and gain exposure to different perspectives. Networking with professionals already established in the field can also open doors to new opportunities.5. Stay Updated with Latest TrendsThe field of analysis is constantly evolving due to advancements in technology and new methodologies. It is crucial to stay updated with the latest trends by reading research papers, following influential blogs orpodcasts, and participating in online forums. This continuous learning will ensure that your skills remain relevant and up-to-date.ConclusionAnalysis is an essential skill that empowers individuals and organizations to make informed decisions based on data-driven insights. Whether it is in business, healthcare, or social sciences, effective analysis enables us to understand complex phenomena, identify patterns, and drive innovation. By acquiring the necessary technical skills, gaining domain knowledge, and staying updated with the latest trends, anyone can enter the field of analysis and contribute to its growth. So, let us embrace the power of analysis and unlock its potential in our lives.Note: This article is for educational purposes only and does not constitute professional advice.。
关于学习策略的报告英语作文Title: Exploring Effective Learning Strategies.In the ever-evolving academic landscape, the importance of effective learning strategies cannot be overstated. These strategies not only enhance academic performance but also foster a life-long passion for learning. In this article, we delve into the intricacies of learning strategies, discussing their significance, types, and practical applications.Firstly, let's delve into the significance of learning strategies. In today's fast-paced world, the traditional rote learning method is becoming increasingly outdated. Instead, students need to adopt innovative strategies that promote critical thinking, problem-solving, and adaptability. Effective learning strategies help students retain information better, improve their comprehension skills, and enhance their motivation to learn. Furthermore, they empower students to become independent learners, ableto navigate new concepts and challenges with confidence.Next, let's explore the various types of learning strategies.1. Active Learning: This strategy involves actively engaging with the material, rather than passively absorbing it. Techniques such as taking notes, summarizing, and discussing concepts with peers foster active learning. By actively participating in the learning process, students are more likely to retain and understand the information.2. Distributed Learning: This strategy advocates spacing out learning over time, rather than cramming all the information in one sitting. By spacing out learning sessions, students give their brains time to process and consolidate information, leading to better retention and comprehension.3. Elaboration: This strategy encourages students to connect new information with what they already know. By making connections and building upon existing knowledge,students are able to understand and remember new concepts more easily.4. Interleaved Practice: This strategy involves mixing up different types of problems or skills during practice sessions. By alternating between different tasks, students are able to generalize their learning and apply it to a wider range of situations.5. Retrieval Practice: This strategy involves retrieving information from memory through quizzes, tests, or self-quizzing. By retrieving information, students are able to巩固 their understanding and identify areas where they need to focus their learning efforts.Now, let's discuss the practical applications of these learning strategies.In the classroom, teachers can incorporate active learning techniques such as group discussions, role-plays, and project-based learning. These techniques encourage students to actively engage with the material, improvingtheir comprehension and motivation to learn. Additionally, teachers can utilize distributed learning by spacing out assignments and assessments over the course of the semester, giving students time to process and consolidate information.Outside the classroom, students can utilize elaboration by connecting new information to their existing knowledge. For example, when learning about a new scientific concept, students can try to connect it to real-world examples or their previous experiences. This will help them understand and remember the concept better.Interleaved practice can be applied when studying for exams or learning new skills. Instead of focusing on one topic or skill at a time, students can alternate between different topics or skills during their practice sessions. This will help them generalize their learning and apply itto a wider range of situations.Finally, retrieval practice can be used on a daily basis. Students can quiz themselves on the material they have learned, identifying areas where they need to focustheir attention. This will help them consolidate their understanding and improve their retention of information.In conclusion, effective learning strategies are crucial for academic success and personal growth. By adopting strategies such as active learning, distributed learning, elaboration, interleaved practice, and retrieval practice, students can enhance their comprehension, retention, and motivation to learn. Teachers and students alike should experiment with different strategies to find the ones that work best for them, fostering a life-long passion for learning.。
写一篇大学学习策略的英语作文Strategies for Effective University LearningIntroductionAttending university is a significant milestone in an individual's academic and personal journey. It presents a unique opportunity to expand one's knowledge, develop critical thinking skills, and prepare for a successful career. However, the transition from high school to university can be challenging, as students are often faced with increased academic demands, a more independent learning environment, and the need to manage their time and resources effectively. In this essay, we will explore various strategies that can help university students maximize their learning potential and achieve academic success.Time ManagementOne of the most crucial aspects of effective university learning is time management. With a multitude of classes, assignments, extracurricular activities, and social engagements, it can be easy for students to feeloverwhelmed and struggle to keep up with their workload. To address this, it is essential for students to develop a well-structured schedule that prioritizes their academic responsibilities. This may involve creating a daily or weekly planner, setting aside dedicated study time, and learning to balance their commitments effectively.Additionally, students should be mindful of their own productivity patterns and learn to manage their time accordingly. For instance, some individuals may be more productive in the morning, while others thrive in the evening. By understanding their personal rhythms, students can optimize their study sessions and ensure that they are making the most of their available time.Active Engagement in the ClassroomPassive listening during lectures is often anineffective learning strategy. Instead, students should strive to actively engage with the course material and participate in class discussions. This can involve asking questions, offering insights, and actively taking notes. By actively engaging with the content, students can deepentheir understanding, retain information more effectively, and develop critical thinking skills.Furthermore, students should take advantage of office hours and seek out opportunities to interact with their professors. These one-on-one interactions can provide valuable feedback, clarification on course concepts, and even guidance on research or career-related matters.Effective Note-taking StrategiesEffective note-taking is a crucial skill for university students. It not only helps them capture and organize important information but also facilitates active engagement with the course material. Students should experiment with different note-taking methods, such as the Cornell method, the outline method, or the mind-mapping technique, to find the approach that works best for their learning style.Moreover, students should review their notes regularly, either immediately after a lecture or during dedicated study sessions. This practice helps to reinforce the information and identify any gaps in their understanding,allowing them to seek clarification or further research the topic.Collaborative LearningUniversity education often involves group projects, discussions, and team-based activities. While these can be challenging, they also present valuable opportunities for collaborative learning. By working with their peers, students can learn from different perspectives, engage in constructive dialogue, and develop essential teamwork and communication skills.Students should actively participate in group discussions, share their ideas, and be open to constructive feedback. They should also be willing to contribute their fair share to group assignments and projects, fostering a sense of shared responsibility and accountability.Self-Reflection and Continuous ImprovementUltimately, the key to effective university learninglies in a student's ability to reflect on their own progress, identify areas for improvement, and implement strategies for continuous growth. This may involveregularly reviewing their academic performance, seeking feedback from professors or peers, and adjusting their study habits or learning approaches as needed.By engaging in self-reflection, students can develop a growth mindset, which enables them to view challenges as opportunities for learning and development. This mindset can help students stay motivated, resilient, and committed to their academic pursuits, ultimately leading to greater success and personal fulfillment.ConclusionNavigating the complexities of university education can be daunting, but with the right strategies and a proactive approach, students can unlock their full potential and achieve academic success. By prioritizing time management, active engagement in the classroom, effective note-taking, collaborative learning, and self-reflection, students can develop the skills and habits necessary to thrive in the university environment and beyond.大学学习策略引言参加大学学习是个人学业和个人发展的重要里程碑。
有关学习策略的英语作文Title: Effective Learning Strategies。
Learning is a journey that requires not only dedication but also effective strategies to maximize comprehension and retention. In this essay, we will explore variousstrategies that can enhance the learning process.Firstly, it is essential to establish clear objectives before embarking on any learning endeavor. Setting specific, achievable goals provides direction and motivation. Whether it's mastering a new language, understanding complex mathematical concepts, or acquiring practical skills,clearly defined objectives serve as roadmaps for success.Secondly, active engagement is crucial for effective learning. Passive methods, such as rote memorization, often yield temporary results. Instead, active learning techniques, such as participating in discussions, solving problems, and teaching others, promote deeper understandingand long-term retention of information.Furthermore, creating a conducive learning environment is paramount. Minimizing distractions and finding a comfortable, well-lit space can significantly enhance concentration and focus. Additionally, utilizing technology wisely, such as educational apps and online resources, can supplement traditional learning methods and cater to diverse learning styles.Moreover, adopting a systematic approach to learning can yield significant benefits. Breaking down complex topics into manageable chunks, creating study schedules, and utilizing mnemonic devices are effective techniques to enhance comprehension and memory retention. Furthermore, regularly reviewing and revising previously learned material reinforces understanding and prevents forgetting.Collaborative learning can also be highly beneficial. Engaging in group discussions, study sessions, or peer teaching allows for the exchange of ideas, perspectives, and feedback. Moreover, explaining concepts to others notonly reinforces one's understanding but also fosters communication and teamwork skills.Additionally, incorporating varied learning modalities can cater to individual preferences and optimize learning outcomes. Visual learners may benefit from diagrams, charts, and videos, while auditory learners may prefer lectures, podcasts, or recorded lectures. Kinesthetic learners, onthe other hand, may benefit from hands-on activities and interactive simulations.Furthermore, embracing a growth mindset is essentialfor continuous improvement. Viewing challenges as opportunities for growth rather than obstacles fosters resilience and perseverance. Embracing mistakes as valuable learning experiences promotes a positive attitude towards learning and encourages experimentation and exploration.In conclusion, effective learning strategies are indispensable for maximizing comprehension, retention, and application of knowledge. By setting clear objectives, actively engaging with the material, creating conducivelearning environments, adopting systematic approaches, fostering collaboration, incorporating varied modalities, and embracing a growth mindset, learners can unlock their full potential and achieve academic and personal success.。
Strategies for English Vocabulary Learning1.Introduction1.1 Research background and purposeWith the openness to the world after decades of isolation and an explosion in commercial, technological, and cultural exchanges with Western countries, China is experiencing transition. Moreover, with China’s joining the market-based global economy, its fast-moving world financial markets appeal to many foreign companies. This financial growth has given rise to a pressing demand for large numbers of competent users of English in a wide range of professions and businesses. Such recent political, economic, and social factors will have a great impact on English language teaching and learning in China. Hence, how to improve learners’ ability of learning English in order to meet new demands seems imperative.In the past 25 years, the field of second language acquisition has seen the reemergence of interest in one area of language study, i.e. vocabulary, and the appearance of a newly recognized aspect learning strategies. V ocabulary is a fundamentally important aspect of language learning. Learning a language is just like constructing a building, in which vocabulary, the building materials, is essential. V ocabulary knowledge enables language use, language use enables the increase of vocabulary knowledge, and knowledge of the world enables the increase of vocabulary knowledge and language use and so on. Rubin& Thompson once said, “One can’t speak, understand, read or write a foreign language without knowing a lot of words.” Therefore, vocabulary is at the heart of mastering a foreign language. In China, the importance of vocabulary in second language acquisition has been greatly recognized. However, teachers put great emphasis on explaining grammar and developing students’reading skill, leaving vocabulary to students themselves. To students, they believe, learning a word means knowing its pronunciation, spelling, and meanings, which can be memorize absolutely by themselves if they want to.However, students cannot achieve a high level after several years of endeavor, and vocabulary is the most challenge and the biggest headache for them. Hence, vocabulary is becoming the neck of a bottle in learning English. According to NewCriteria for English Courses for General High School, students’ English ability after high school should pass Grade 8 with the demand of vocabulary is 3,500~4,000 words, which is similar to that of a college student. Facing to the contradiction between low-efficiency of vocabulary learning and higher requirement on vocabulary size, teachers should realize the necessity of vocabulary learning strategies and help students employ more effective strategies, which will empower students to continue to acquire vocabulary on their own. Students understand the advantages and disadvantages of various vocabulary learning strategies and master the effective vocabulary learning strategies to improve their study efficiency.Previous research has typically associated higher proficiency in the second language with more frequent, appropriate and successful use of language learning strategies, choosing adults or college learners as main participants. Nowadays, however, few studies have been concerned specifically with middle school students’handling of new second language vocabulary and how this relates to vocabulary knowledge. Students in senior middle schools are different from those adults in cognitive processes and English proficiency, which implies they may have their own characteristics in vocabulary learning strategies use. The present paper extends existing research by focusing on the strategies used by high school students who study English as their foreign language. It is expected to draw a clear picture of vocabulary learning strategy use in order to improve their learning efficiency and English proficiency.1.2 The structure of the thesisThe investigation of vocabulary learning strategies have implications for improving vocabulary learning in junior middle school and creating instructional designs that are student-centered and foster self-directed learning. The paper will help students have awareness of their own vocabulary learning strategies which they use. In addition, students will learn more about their strength and weakness in vocabulary learning. This article exposes to more strategies and develops more effective appropriate strategies. The whole paper consists of five parts. First part is a brief introduction which states the background and the purpose of the present paper. Parttwo reviews the key terms on vocabulary learning strategies, and also reviews studies on fields at home. It also gives some representative classification of vocabulary learning strategies. The third part tells us five vocabulary learning strategies which are mainly suggested in the article. And the next part shows some other vocabulary learning strategies. The last part discusses the results of the study, concludes the major findings and suggesting English vocabulary learning strategies.2. Research on vocabulary learning strategies2.1 Definition of vocabulary learning strategiesThe definition of vocabulary learning strategies derives from language learning strategies. Schmit adopts Rubin’s definition of learning as “the process by which information is obtained, stored, retrieved, and used and extended to vocabulary learning strategies in the following way”. V ocabulary learning strategies could be any which affect this rather broadly-defined process. Rosa Maria and Jimenez Catalan describe an elaborate working definition for vocabulary learning strategy: knowledge about the mechanisms (processes, strategies) used in order to learn vocabulary as well as steps or actions taken by students (a) to retain them in long-term memory, (b) to find out the meaning of unknown words, (c) to recall them at will, and (d) to use them in oral or written mode. It can be inferred that vocabulary learning strategies concerns techniques, approaches of discovery, consolidation and production of words.2.2 Classification of vocabulary learning strategiesA number of attempts have been made to classify learning strategies. Rubin (1981) classified the strategies into two primary groups: direct and indirect learning strategies. The former refers to those directly affecting learning and the latter consisting of those contributing indirectly to learn. Meanwhile, Naiman (1978) proposed five broad categories of learning strategies with numerous secondary categories. Influenced strongly by the findings of cognitive psychology, O’Malley and Chamot (2001) considered learning strategies as cognitive skills. Learning strategies have been differentiated into three categories depending on the level or type of processing involved. They are meta-cognitive strategies, cognitive strategies, and social/affective strategies. Meta-cognitive strategies are higher order executive skillsthat may entail planning for, monitoring, or evaluating the success of a learning activity. Cognitive strategies operate directly on incoming information, manipulating it in ways that enhance learning. Social/affective strategies represent a broad grouping that involves either interaction with another person or ideational control over affect. Gu and Johnson (1996)established two main dimensions of vocabulary learning strategies for their study, met-cognitive regulation and cognitive strategies, which covered six subcategories: guessing, using a dictionary, note-taking, rehearsal, encoding, and activating, all of which were further subcategorized. The total number of strategies in their study was 74.Cohen used second language learner strategies to encompass both second language learning strategies and use strategies. He defined learning strategies as “learning processes which are consciously selected by the learners”. Language use strategies are those used in producing the language. It includes four subsets: retrieval strategies, rehearsal strategies, cover strategies, and communication strategies.2.3 Studies on vocabulary learning strategies in ChinaWestern-based research does not in principle tell us anything about the learning of English vocabulary by Chinese learners.This paper needs research in a Chinese cultural context. And some researchers engage in generalizing vocabulary learning strategies such as Zhang ye, Xing min and Zhou Dajun. Some researchers specialize in some particular strategies like Fan Lin, Wang qinghua and Wu Lilin. Most studies on vocabulary learning strategies investigated the vocabulary learning strategies used by English or non-English major university students such as those researchers: Fan&Wang, 2002; Gu, 1994; Gu &Hu, 2003; Gu&Johnson, 1996; Wang, 1998; Wu& Wang, 1998; Zhang, 2001. There are few studies investigating vocabulary learning strategies used by children or high school students.A study by Yao, Wu and Pang investigated the English vocabulary memorizing strategies of junior middle school students. 134 junior middle school students in Grade Three were given Questionnaire of English V ocabulary Memorizing strategies and vocabulary tests with different degrees of difficulty. The results showed that rote and general memorizing strategies were used primarily and the good students were better than the poor in choosing specific strategies for vocabulary memorizingproperly. The frequency and the number of students were increasing in using more kinds of strategies when facing difficult vocabulary.The previous studies profiles vocabulary learning strategies used by non-English major university learners throws a comprehensive insight in vocabulary teaching and learns especially in universities. However, few people concern with vocabulary learning strategies used by high school students, so more empirical studies in high schools are needed.3.The main vocabulary learning strategies3.1 Meta-cognitive strategiesMeta-cognitive strategies are used by students to control and evaluate their own learning by having an overview of the learning process in general. It is a kind of strategies that facilitate learning by actively involving the learner in conscious efforts to remember new words. The pervasiveness of English-medium books, magazines, newspapers and movies offer an almost endless resource. Besides, the strategy of interacting with native speakers and self-testing also increase input. In practice, one can maximize the effectiv eness of one’s practice time if it is scheduled and organized rather than random. However, learners need to realize that they will never learn all the words and so they should concentrate their limited resources on learning the most useful ones and pass over some unknown words particularly when the main goal is to improve reading speed rather than vocabulary growth. Usually, meta-cognitive strategies divide into plan strategies, control strategies and adjust strategies.Planning strategies include setting learning objectives, browsing the reading materials, producing the questions to answer, and analyzing how to fulfill the tasks. According to the actual process of evaluation and feedback in cognitive goal of cognitive deficiencies, control strategy is the cognitive activities which correct estimation results and the degree of goals. And adjust strategy is based on cognitive activities, such as finding problems, adopting corresponding remedial measures.According to the inspection of the cognitive strategies, adjust strategy modify and adjust timely.Due to the English vocabulary learning is a system of engineering; therefore, it needs to actively adopt meta-cognitive strategies, especially for high school students. Meanwhile it’s important to arrange the reasonable vocabulary learning plans, control the vocabulary learning process constantly, and adjust methods of vocabulary learning. Students are suggested to use the suitable vocabulary learning strategies for themselves.3.2 Semantic field strategiesV ocab ulary is not isolated in people’s memory, but store in human’s memory according to the different kinds. Some words can associate with other words once remember one of the words. On this ground, semantic strategy can be used in vocabulary learning. Semantic paradigmatic relations have known as polymerization (accept-receive; sad-depressed), antonym (long-short; clever-stupid); hyponymy (animal-lion; elephant), etc. For instance, clock, washbasin, stove, mirror, TV, sofa, video, slipper, toothpaste, cupboard shower, towel; overcoat, gloves, raincoat, trousers, suit, T-shirt, sweater can be used in the study.Zhang wen peng used to research and confirm that vocabulary learning strategies (Word list strategy, Context strategy, Semantic field strategy) have an influence on remembering vocabularies. According to the results: From the memory capacitance, context strategy is more than semantic field strategy in short-term memory. As a result, if you want to remember more vocabularies in a short period of time (1-2 weeks) and it’s appropriate to memory though semantic strategy.3.3 Context strategiesContext strategy is a strategy that learners guess the new word through the information which is provided in the context language environment. Context strategy is a popular strategy of vocabulary learning strategies right now. It not only can expand your vocabulary, and can let students understand the cultural knowledge aboutlanguage. In order to understand and learn new vocabulary better, the methods make learners internalize and absorb new words accurately: namely, to learn the phonetic pronunciation, understand a single syllable count and master stress position.3.4 Determination strategiesIf students do not know a word, they have to decide which strategy to choose for the problem. There are four possible opinions for them to solve their problems.1) Guessing the meaning from word part formation such as, derivation, compounding, conversion, abbreviation, backformation, etc;2) Guessing the meaning from a L1 or L2 word similar in sound or meaning (e.g. loaned words from Chinese like kungfu, Confucius, Taoist, etc.);3) Guessing the meaning from context (It is necessary to note that context here should be taken to mean more than just textual context, context clues can also come from a variety of other sources such as pictures, cards, gestures, intonations and so on );4) Guessing the meaning from reference materials (primarily dictionaries).For our high school students, this vocabulary learning strategy is also important.3.5 Social strategiesA second way to discover a new word meaning employs the social strategies of asking someone who knows. Teachers are often in this position, and they can be asked to give help in a variety of ways: giving the L1 translation, giving a synonym, giving an antonym, giving a definition by paraphrase, using the word in sentence, or any combination of these. Of course, classmates and friends also can be asked for meaning in all of ways above. Besides the initial discovery of a word, group work can be used to learn or practice vocabulary. Cooperative learning can prepare the participants for “team activities”. Another social strategy involves students enlisting teachers to check their work for accuracy. Of course, if possible, interacting with native speakers would be an excellent way to gain vocabulary.4.Other vocabulary learning strategiesThere are some other vocabulary learning strategies which used by people in daily life. For instance, Categorization, Memory strategies, Related words, Rhyme and imagery.At first, Categorization (also known as grouping) is an important way to recall, and people often organize words into group naturally according to the meaning, part of speech, formal or informal language forms, alphabetical order, or types of clothing, food, and so on. However, Memory strategy is different from Categorization. Most Memory strategies (traditionally called mnemonics) involve relating the word to be retained with some previously learned knowledge, using some form of imagery, grouping, or actions (Coady and Huckin, 2001: 203 - 20). Four options are proposed here.1). Using the keyword methods: finding a L1 word or phrase with similar sound or spelling, and constructing a visual image that ties the word or phrase to the target word, e.g. while learning the word “deny”in English, a similar sound word in Chinese “dina” will occur to the learner.2). Studying a word’s affixes, root and word class.3). Learning the individual word meanings in more meaningful context such as a phrase, an idiom or a proverb.4). Using semantic feature grids.Students always make use of this kind of vocabulary learning strategies, just like Related words, Rhyme / rhythm, Pictures / imagery.New words can be linked to other words which the students already know, such as coordination (apples- other kinds of fruits like pears, bananas, peaches etc.), synonym (pleased - glad / happy / joyful), or antonym (quick - slow). Some words, particularly gradable adjectives, have meanings relative to the other words in their set. For example, a helpful way to remember the words big, huge, medium-size, etc. is to set them in a scale(huge > big > medium-sized > small > tinny). And making upsongs or short ditties with the words to learn is of great interest and a high efficiency way of gaining new words. For example, one is a pen, two is a duck, three is an ear, four is a flag etc. What’s more, new words can be learned by studying them with pictures of their meanings or creating their own mental images of a w ord’s meaning. They can also be associated with a particular vivid personal experience, for example, a learner mentally connects the word snow to a memory of playing in the snow while he /she was a child.5. ConclusionAbove all, students need to know how to learn vocabularies. In this sense, students learn various kinds of learning strategies and the instruction of vocabulary in classroom. Moreover, students are encouraged to seek the ways they find most helpful since their vocabulary possession is very personal. And their abilities to exploit its elasticity are very individual. Some of them make lists and memorize vocabularies; some of them make associations with words that sound or look similar in Chinese; some students read a large number of materials; some students write down words in short time and some students read dictionaries. In short, it’s important for high students to study more English vocabulary learning strategies. After doing this research, the paper suggests students to choose their own strategies for learning. Learning vocabulary well is the key to learning English and adopting right English vocabulary learning strategies is also the key to learning vocabulary well.ReferencesCoady, J. and Huckin T. 2001. Second Language Vocabulary Acquisition [M].Shanghai: Shanghai Foreign Language Education Press.Cohen, A. 1990. Language Learning: Insights for Learners, Teachers and Researchers [M]. New York: Newbury House∕Harper Row.Gu, Yongqi. & Johnson, R. 1996. V ocabulary learning strategies and language learning outcomes [J]. Language Learning 46(4): 643-679.Nation, I. 2001. Learning vocabulary in another language[M]. Cambridge: Cambridge University Press.O’Malley, J. M. & Chamot. 2001.Learning Strategies in Second Language Acquisition [M]. Shanghai: Shanghai Foreign Language Education Press. Rosa, M. and Jimenez, C. 2003. Six differences in L2 vocabulary learning strategies [J].International Journal of Applied Linguistics 13 (1): 54–77.Schmitt, N. 1997. Vocabulary learning strategies[M]. Cambridge: Cambridge University Press.姚梅林、吴建民、庞晖，2000 ,初中生英语词汇记忆策略的研究[J], 《心理科学》。
AbstractWe present an approach to automatically learningstrategies for natural language question answeringfrom examples composed of textual sources,questions, and answers. Our approach formulatesQA as a problem of first order inference over asuitably expressive, learned representation. Thisframework draws on prior work in learning actionand problem-solving strategies, as well as relationallearning methods. We describe the design of asystem implementing this model in the framework ofnatural language question answering for storycomprehension. Finally, we compare our approachto three prior systems, and present experimentalresults demonstrating the efficacy of our model.1 IntroductionThis paper presents an approach to automatically learning strategies for natural language question answering from examples composed of textual sources, questions, and answers. Our approach is focused on one specific type of text-based question answering known as story comprehension. Most TREC-style QA systems are designed to extract an answer from a document contained in a fairly large general collection [Voorhees, 2003]. They tend to follow a generic architecture, such as the one suggested by [Hirschman and Gaizauskas, 2001], that includes components for document pre-processing and analysis, candidate passage selection, answer extraction, and response generation. Story comprehension requires a similar approach, but involves answering questions from a single narrative document. An important challenge in text-based question answering in general is posed by the syntactic and semantic variability of question and answer forms, which makes it difficult to establish a match between the question and answer candidate. This problem is particularly acute in the case of story comprehension due to the rarity of information restatement in the single document.Several recent systems have specifically addressed the task of story comprehension. The Deep Read reading comprehension system [Hirschman et al., 1999] uses a statistical bag-of-words approach, matching the question with the lexically most similar sentence in the story. Quarc [Riloff and Thelen, 2000] utilizes manually generated rules that selects a sentence deemed to contain the answer based on a combination of syntactic similarity and semantic correspondence (i.e., semantic categories of nouns). The Brown University statistical language processing class project systems [Charniak, et al., 2000] combine the use of manually generated rules with statistical techniques such as bag-of-words and bag-of-verb matching, as well as deeper semantic analysis of nouns. As a rule, these three systems are effective at identifying the sentence containing the correct answer as long as the answer is explicit and contained entirely in that sentence. They find it difficult, however, to deal with semantic alternations of even moderate complexity. They also do not address situations where answers are split across multiple sentences, or those requiring complex inference.Our framework, called QABLe (Question-Answering Behavior Learner), draws on prior work in learning action and problem-solving strategies [Tadepalli and Natarajan, 1996; Khardon, 1999]. We represent textual sources as sets of features in a sparse domain, and treat the QA task as behavior in a stochastic, partially observable world. QA strategies are learned as sequences of transformation rules capable of deriving certain types of answers from particular text-question combinations. The transformation rules are generated by instantiating primitive domain operators in specific feature contexts. A process of reinforcement learning [Kaebling, et al., 1996] is used to select and promote effective transformation rules. We rely on recent work in attribute-efficient relational learning [Khardon et al., 1999; Cumby and Roth, 2000; Even-Zohar and Roth, 2000] to acquire natural representations of the underlying domain features. These representations are learned in the course of interacting with the domain, and encode the features at the levels of abstraction that are found to be conducive to successful behavior. This selection effect is achieved by a fusion of abstraction space generalization [Sacerdoti, 1974; Knoblock, 1992] and reinforcement learning elements.The rest of this paper is organized as follows. Section 2 presents the details of the QABLe framework. In section 3 we describe preliminary experimental results which indicate promise for our approach. In section 4 we summarize and draw conclusions.Learning Strategies for Open-Domain Natural Language Question AnsweringEugene Grois, David C. Wilkins, Dan RothBeckman InstituteUniversity of Illinois405 N. Mathews AvenueUrbana, IL 61801e-grois@, dcw@, danr@2 QABLe – Learning to Answer Questions2.1 OverviewFigure 1 shows a diagram of the QABLe framework. The bottom-most layer is the natural language textual domain. It represents raw textual sources, questions, and answers. The intermediate layer consists of processing modules that translate between the raw textual domain and the top-most layer, an abstract representation used to reason and learn.This framework is used both for learning to answer questions and for the actual QA task. While learning, the system is provided with a set of training instances, each consisting of a textual narrative, a question, and a corresponding answer. During the performance phase, only the narrative and question are given.At the lexical level, an answer to a question is generated by applying a series of transformation rules to the text of the narrative. These transformation rules augment the original text with one or more additional sentences, such that one of these explicitly contains the answer, and matches the form of the question.On the abstract level, this is essentially a process of searching for a path through problem space that transforms the world state, as described by the textual source and question, into a world state containing an appropriate answer. This process is made efficient by learning answer-generation strategies. These strategies store procedural knowledge regarding the way in which answers are derived from text, and suggest appropriate transformation rules at each step in the answer-generation process. Strategies (and the procedural knowledge stored therein) are acquired by explaining (or deducing) correct answers from training examples. The framework’s ability to answer questions is tested only with respect to the kinds of documents it has seen during training, the kinds of questions it has practiced answering, and its interface to the world (domain sensors and operators).In the next two sections we discuss lexical pre-processing, and the representation of features and relations over them in the QABLe framework. In section 2.4 we look at the structure of transformation rules and describe how they are instantiated. In section 2.5, we build on this information and describe details of how strategies are learned and utilized to generate answers. In section 2.6 we explain how candidate answers are matched to the question, and extracted.2.2 Lexical Pre-ProcessingSeveral levels of syntactic and semantic processing are required in order to generate structures that facilitate higher order analysis. We currently use MontyTagger 1.2, an off-the-shelf POS tagger based on [Brill, 1995], for POS tagging. At the next tier, we utilize a Named Entity (NE) tagger for proper nouns a semantic category classifier for nouns and noun phrases, and a co-reference resolver (that is limited to pronominal anaphora). Our taxonomy of semantic categories is derived from the list of unique beginners for WordNet nouns [Fellbaum, 1998]. We also have a parallel stage that identifies phrase types. Table 1 gives a list of phrase types currently in use, together with the categories of questions each phrase type can answer. In the near future, we plan to utilize a link parser to boost phrase-type tagging accuracy. For questions, we have a classifier that identifies the semantic category of information requested by the question. Currently, this taxonomy is identical to that of semantic categories. However, in the future, it may be expanded to accommodate a wider range of queries. A separate module reformulates questions into statement form for later matching with answer-containing phrases.2.3 Representing the Question-Answering DomainIn this section we explain how features are extracted from raw textual input and tags which are generated by pre-processing modules.A sentence is represented as a sequence of words 〈w 1, w 2,…, w n 〉, where word (w i , word) binds a particular word to its position in the sentence. The k th sentence in a passage isgiven a unique designation s k . Several simple functions capture the syntax of the sentence. The sentence Main (e.g ., main verb) is the controlling element of the sentence, and is recognized by main (w m , s k ). Parts of speech are recognized by the function pos , as in pos (w i , NN) and pos (w i , VBD). The relative syntactic ordering of words is captured by the predicate before (w i , w j ). It can be applied recursively, as before (w i , before (w j , w k )) to generate the entire sentence starting with an arbitrary word, usually the sentence Main . The integrity of the sentence is checked by the function inSentence (w i , s i ) ⇒ main (w m , s k ) ∧ (before (w i , w m ) ∨ before (w m , w i )) for each word w i in the sentence. A consecutive sequence of words is a phrase entity or simply entity . It is given the designation e x and declared by a binding function, such as entity (e x , NE ) for a named entity, and entity (e x , NP ) for a syntactic group of type noun phrase . Each phrase entity is identified by its head , as head (w h , e x ), and we say that the phrase head controls the entity. A phrase entity is defined as head (w h , e x ) ∧ inPhrase (w i , e x ) ∧ … ∧ inPhrase (w j , e x ).We also wish to represent higher-order relations such as functional roles and semantic categories. Functional dependency between pairs of words is encoded as, for example, subj (w i , w j ) and aux (w j , w k ). Functional groups are represented just like phrase entities. Each is assigned a designation r x , declared for example, as func_role (r x , SUBJ ), and defined in terms of its head and members (which may be individual words or composite entities). Semantic categories are similarly defined over the set of words and syntactic phrase entities – for example, sem_cat (c x , PERSON ) ∧ head (w h , c x ) ∧ pos (w i , NNP) ∧ word (w h , “John”).Semantically, sentences are treated as events defined by their verbs. A multi-sentential passage is represented by tying the member sentences together with relations over their verbs. We declare two such relations – seq and cause . The seq relation between two sentences, seq (s i , s j ) ⇒ prior (main (s i ), main (s j )), is defined as the sequential ordering in time of the corresponding events. The cause relation cause (s i , s j ) ⇒ cdep (main (s i ), main (s j )) is defined such that the second event is causally dependent on the first.Phrase TypeCommentsSUBJ“Who” and nominal “What”questionsVERB event “What” questionsDIR-OBJ“Who” and nominal “What”questionsINDIR-OBJ“Who” and nominal “What”questionsELAB-SUBJdescriptive “What” questions(eg. what kind)ELAB-VERB-TIME ELAB-VERB-PLACE ELAB-VERB-MANNER ELAB-VERB-CAUSE “Why” questionELAB-VERB-INTENTION“Why” as well as “What for”questionELAB-VERB-OTHERsmooth handling of undefinedverb phrase typesELAB-DIR-OBJdescriptive “What” questions(eg. what kind)ELAB-INDIR-OBJdescriptive “What” questions(eg. what kind)VERB-COMPL WHERE/WHEN/HOWquestions concerning state orstatus Table 1. Phrase types used by QABLe framework.2.4 Primitive Operators and Transformation RulesThe system, in general, starts out with no procedural knowledge of the domain (i.e ., no transformation rules). However, it is equipped with 9 primitive operators that define basic actions in the domain. Primitive operators are existentially quantified. They have no activation condition, but only an existence condition – the minimal binding condition for the operator to be applicable in a given state.A primitive operator has the form AC E ˆ→, where E C is the existence condition and Aˆ is an action implemented in the domain. An example primitive operator isprimitive-op-1 : ∃ w x , w y → add-word-after(w y , w x ) Other primitive operators delete words or manipulate entire phrases. Note that primitive operators act directly on the syntax of the domain. In particular, they manipulate wordsand phrases. A primitive operator bound to a state in the domain constitutes a transformation rule. The procedure forinstantiating transformation rules using primitive operators is given in Figure 2. The result of this procedure is auniversally quantified rule having the form A G C R →∧. Amay represent either the name of an action in the world or an internal predicate. C represents the necessary condition for rule activation in the form of a conjunction over the relevant attributes of the world state. R G represents the expected effect of the action. For example, turn_on_x2→∧∧221g x x indicates that when 1x is on and2x is off, this operator is expected to turn 2x on.An instantiated rule is assigned a rank composed of:• priority rating• level of experience with rule• confidence in current parameter bindingsThe first component, priority rating, is an inductively acquired measure of the rule’s performance on previous instances. The second component modulates the priority rating with respects to a frequency of use measure. The third component captures any uncertainty inherent in the underlying features serving as parameters to the rule. Each time a new rule is added to the rule base, an attempt is made to combine it with similar existing rules to produce more general rules having a wider relevance and applicability. Given a rule 1A g g c c Ry R x b a →∧∧∧covering a set of example instances 1E and another rule2A g g c c R z R y c b →∧∧∧covering a set of examples 2E , we add a more general rule 3A g c Ry b →∧ to the strategy. The new rule 3A is consistent with 1E and 2E . In addition it will bind to any state where the literal b c is active. Therefore the hypothesis represented by the triggering condition islikely an overgeneralization of the target concept. This means that rule 3A may bind in some states erroneously. However, since all rules that can bind in a state compete to fire in that state, if there is a better rule, then 3A will be preempted and will not fire.2.5 Generating AnswersReturning to Figure 1, we note that at the abstract level the process of answer generation begins with the extraction of features active in the current state. These features represent low-level textual attributes and the relations over them described in section 2.3.Immediately upon reading the current state, the system checks to see if this is a goal state . A goal state is a state who’s corresponding textual domain representation containsan explicit answer in the right form to match the questions. In the abstract representation, we say that in this state all of the goal constraints are satisfied.If the current state is indeed a goal state, no further inference is required. The inference process terminates andthe actual answer is identified by the matching technique described in section 2.6 and extracted.If the current state is not a goal state and more processing time is available, QABLe passes the state to theInference Engine (IE). This module stores strategies in theform of decision lists of rules. For a given state, each strategy may recommend at most one rule to execute. For each strategy this is the first rule in its decision list to fire. The IE selects the rule among these with the highest relative rank, and recommends it as the next transformation rule tobe applied to the current state. If a valid rule exists it is executed in the domain. This modifies the concrete textual layer. At this point, the pre-processing and feature extraction stages are invoked, a new current state is produced, and the inference cycle begins anew. If a valid rule cannot be recommend by the IE, QABLe passes the current state to the Search Engine (SE). The SEuses the current state and its set of primitive operators to instantiate a new rule, as described in section 2.4. This rule is then executed in the domain, and another iteration of the process begins. If no more primitive operators remain to be applied to the current state, the SE cannot instantiate a new rule. At this point, search for the goal state cannot proceed, processing terminates, and QABLe returns failure. When the system is in the training phase and the SE instantiates a new rule, that rule is generalized against theexisting rule base. This procedure attempts to create moregeneral rules that can be applied to unseen exampleinstances.Once the inference/search process terminates(successfully or not), a reinforcement learning algorithm isapplied to the entire rule search-inference tree. Specifically,rules on the solution path receive positive reward, and rules that fired, but are not on the solution path receive negativereinforcement.2.6 Candidate Answer Matching and Extraction As discussed in the previous section, when a goal state is generated in the abstract representation, this corresponds toa textual domain representation that contains an explicit answer in the right form to match the questions. Such a candidate answer may be present in the original text, or may be generated by the inference/search process. In either case, the answer-containing sentence must be found, and the actual answer extracted. This is accomplished by the Answer Matching and Extraction procedure.The first step in this procedure is to reformulate the question into a statement form. This results in a sentence containing an empty slot for the information being queried. For example, “How far is the drive to Chicago?” becomes “The drive to Chicago is ______.” Recall further that QABLe’s pre-processing stage analyzes text with respect to various syntactic and semantic types. In addition to supporting abstract feature generation, these tags can be used to analyze text on a lexical level. Thus, the question above is marked up as [ELAB-VERB <quantity-distance> (WRB How) (RB far)] [VERB (VBZ is)] [SUBJ <action> (DT the) (NN drive)] [VERB-COMPL (TO to) (NNP <place> Chicago)]. Once reformulated into statement form, this becomes [ELAB-VERB <quantity-distance> ______ ] [VERB (VBZ is)] [SUBJ <action> (DT the) (NN drive)] [VERB-COMPL (TO to) (NNP <place> Chicago)]. The goal now is to find a sentence who’s syntactic and semantic analysis matches that of the reformulated question’s as closely as possible. Thus, for example the text may contain the sentence “The drive to Chicago is 2 hours” with the corresponding analysis [SUBJ <action> (DT the) (NN drive)] [VERB-COMPL (TO to) (NNP <place> Chicago)] [VERB (VBZ is)] [VERB-COMPL <quantity-time> (CD 2) (NNS hours)]. Notice that all of the elements of this candidate answer match the corresponding elements of the question, with the exception of the semantic category of the ELAB-VERB phrase. This is likely not the answer we are seeking. The text may contain a second sentence “The drive to Chicago is 130 miles”, that analyses as [SUBJ <action> (DT the) (NN drive)] [VERB-COMPL (TO to) (NNP <place> Chicago)] [VERB (VBZ is)] [VERB-COMPL <quantity-distance> (CD 130) (NNS miles)]. In this case, all of the elements match their counterparts in the reformulated question. Thus, the second sentence can be matched as the correct answer with high confidence. 3 Experimental Evaluation3.1 Experimental SetupWe evaluate our approach to open-domain natural language question answering on the Remedia corpus. This is a collection of 115 children’s stories provided by Remedia Publications for reading comprehension. The comprehension of each story is tested by answering five who, what, where, and why questions.The Remedia Corpus was initially used to evaluate the Deep Read reading comprehension system [Hirschman et al., 1999], and later also other systems, including Quarc [Riloff and Thelen, 2000] and the Brown University statistical language processing class project [Charniak, et al., 2000].The corpus includes two answer keys. The first answer key contains annotations indicating the story sentence that is lexically closest to the answer found in the published answer key (AutSent). The second answer key contains sentences that a human judged to best answer each question (HumSent). Examination of the two keys shows the latter to be more reliable. We trained and tested using the HumSent answers. We also compare our results to the HumSent results of prior systems. In the Remedia corpus, approximately 10% of the questions lack an answer. Following prior work, only questions with annotated answers were considered.We divided the Remedia corpus into a set of 55 tests used for development, and 60 tests used to evaluate our model, employing the same partition scheme as followed by the prior work mentioned above. With five questions being supplied with each test, this breakdown provided 275 example instances for training, and 300 example instances to test with. However, due to the heavy reliance of our model on learning, many more training examples were necessary. We widened the training set by adding story-question-answer sets obtained from several online sources. With the extended corpus, QABLe was trained on 262 stories with 3-5 questions each, corresponding to 1000 example instances.3.2 Discussion of ResultsTable 2 compares the performance of different versions of QABLe with those reported by the three systems describedSystem who what when where why Overall DeepRead 48% 38% 37% 39% 21% 36% Quarc 41% 28% 55% 47% 28% 40% Brown 57% 32% 32% 50% 22% 41% QABLe-N/L 48% 35% 52% 43% 28% 41%QABLe-L 56% 41% 56% 45% 35% 47%QABLe-L+ 59% 43% 56% 46% 36% 48%Table 2. Comparison of QA accuracy by question type.System # rules learned # rules on solution path average # rules per correct answer QABLe-L 3,463 426 3.02 QABLe-L+ 16,681 411 2.85Table 3. Analysis of transformation rule learning and use.above. We wish to discern the particular contribution of transformation rule learning in the QABLe model, as well as the value of expanding the training set. Thus, the QABLe-N/L results indicate the accuracy of answers returned by the QA matching and extraction algorithm described in section 2.6 only. This algorithm is similar to prior answer extraction techniques, and provides a baseline for our experiments. The QABLe-L results include answers returned by the full QABLe framework, including the utilization of learned transformation rules, but trained only on the limited training portion of the Remedia corpus. The QABLe-L+ results are for the version trained on the expanded training set.As expected, the accuracy of QABLe-N/L is comparable to those of the earlier systems. The Remedia-only training set version, QABLe-L, shows an improvement over both the baseline QABLe, and most of the prior system results. This is due to its expanded ability to deal with semantic alternations in the narrative by finding and learning transformation rules that reformulate the alternations into a lexical form matching that of the question.The results of QABLe-L+, trained on the expanded training set, are for the most part noticeably better than those of QABLe-L. This is because training on more example instances leads to wider domain coverage through the acquisition of more transformation rules. Table 3 gives a break-down of rule learning and use for the two learning versions of QABLe. The first column is the total number of rules learned by each system version. The second column is the number of rules that ended up being successfully used in generating an answer. The third column gives the average number of rules each system needed to answer an answer (where a correct answer was generated). Note that QABLe-L+ used fewer rules on average to generate more correct answers than QABLe-L. This is because QABLe-L+ had more opportunities to refine its policy controlling rule firing through reinforcement and generalization.Note that the learning versions of QABLe do significantly better than the QABLe-N/L and all the prior systems on why-type questions. This is because many of these questions require an inference step, or the combination of information spanning multiple sentences. QABLe-L and QABLe-L+ are able to successfully learn transformation rules to deal with a subset of these cases.4 ConclusionIn this paper we present an approach to automatically learn strategies for natural language questions answering from examples composed of textual sources, questions, and corresponding answers. The strategies thus acquired are composed of ranked lists transformation rules that when applied to an initial state consisting of an unseen text and question, can derive the required answer. We compare our approach to three prior systems, and present experimental results demonstrating the efficacy of our model. In particular, we show that our approach is effective given a sufficiently large training corpus, and reasonably accurate pre-processing of raw input data.References[Brill, 1995] E. Brill. Transformation-based error driven learning and natural language processing: A case studyin part of speech tagging. In Computational Linguistics, 21(4):543-565, 1995.[Charniak, et al., 2000] Charniak, Y. Altun, R. de Salvo Braz, B. Garrett, M. Kosmala, T. Moscovich, L. Pang,C. Pyo, Y. Sun, W. Wy, Z. Yang, S. Zeller, and L. Zorn.Reading comprehension programs in a statistical-language-processing class. ANLP/NAACL-00, 2000. [Cumby and Roth, 2000] C. Cumby and D. Roth.Relational representations that facilitate learning. KR-00, pp. 425-434, 2000.[Even-Zohar and Roth, 2000] Y. Even-Zohar and D. Roth.A classification approach to word prediction. NAACL-00, pp. 124-131, 2000.[Fellbaum, 1998] C. Fellbaum (ed.) WordNet: An Electronic Lexical Database. The MIT Press, 1998. [Hirschman and Gaizauskas, 2001] L. Hirschman and R.Gaizauskas. Natural language question answering: The view from here. Natural Language Engineering,7(4):275-300, 2001.[Hirschman et al., 1999] L. Hirschman, M. Light, and J.Burger. Deep Read: A reading comprehension system.ACL-99, 1999.[Kaebling, et al., 1996] L. P. Kaebling, M. L. Littman, andA. W. Moore. Reinforcement learning: A survey. J.Artif. Intel. Research, 4:237-285, 1996.[Khardon et al., 1999] R. Khardon, D. Roth, and L. G.Valiant. Relational learning for nlp using linearthreshold elements, IJCAI-99, 1999.[Khardon, 1999] R. Khardon. Learning to take action.Machine Learning 35(1), 1999.[Knoblock, 1992] C. Knoblock. Automatic generation of abstraction for planning. Artificial Intelligence, 68(2):243-302, 1992.[Riloff and Thelen, 2000] E. Riloff and M. Thelen. A rule-based question answering system for readingcomprehension tests. ANLP/NAACL-2000, 2000. [Sacerdoti, 1974] E. D. Sacerdoti. Planning in a hierarchy of abstraction spaces. Artificial Intelligence, 5(2):115-135, 1974.[Tadepalli and Natarajan, 1996] P. Tadepalli and B.Natarajan. A formal framework for speedup learningfrom problems and solutions. J. Artif. Intel. Research,4:445-475, 1996.[Voorhees, 2003] E. M. Voorhees Overview of the TREC 2003 question answering track. TREC-12, 2003.。
