| |
| |
| What This Book Is About | |
| |
| |
| Overview for Instructors | |
| |
| |
| Acknowledgments | |
| |
| |
| |
| Prologue: Encoding Language on Computers | |
| |
| |
| |
| Where do we start? | |
| |
| |
| |
| Encoding language | |
| |
| |
| |
| Writing systems used for human languages | |
| |
| |
| |
| Alphabetic systems | |
| |
| |
| |
| Syllabic systems | |
| |
| |
| |
| Logographic writing systems | |
| |
| |
| |
| Systems with unusual realization | |
| |
| |
| |
| Relation to language | |
| |
| |
| |
| Encoding written language | |
| |
| |
| |
| Storing information on a computer | |
| |
| |
| |
| Using bytes to store characters | |
| |
| |
| |
| Encoding spoken language | |
| |
| |
| |
| The nature of speech | |
| |
| |
| |
| Articulatory properties | |
| |
| |
| |
| Acoustic properties | |
| |
| |
| |
| Measuring speech | |
| |
| |
| |
| Reading a spectrogram | |
| |
| |
| |
| Relating written and spoken language | |
| |
| |
| |
| Language modeling for automatic speech recognition | |
| |
| |
| |
| Writers' Aids | |
| |
| |
| |
| Introduction | |
| |
| |
| |
| Kinds of spelling errors | |
| |
| |
| |
| Nonword errors | |
| |
| |
| |
| Real-word errors | |
| |
| |
| |
| Spell checkers | |
| |
| |
| |
| Nonword error detection | |
| |
| |
| |
| Isolated-word spelling correction | |
| |
| |
| Under the Hood 3: Dynamic programming | |
| |
| |
| |
| Word correction in context | |
| |
| |
| |
| What is grammar? | |
| |
| |
| Under the Hood 4: Complexity of languages | |
| |
| |
| |
| Techniques for correcting words in context | |
| |
| |
| Under the Hood 5: Spell checking for web queries | |
| |
| |
| |
| Style checkers | |
| |
| |
| |
| Language Tutoring Systems | |
| |
| |
| |
| Learning a language | |
| |
| |
| |
| Computer-assisted language learning | |
| |
| |
| |
| Why make CA LL tools aware of language? | |
| |
| |
| |
| What is involved in adding linguistic analysis? | |
| |
| |
| |
| Tokenization | |
| |
| |
| |
| Part-of-speech tagging | |
| |
| |
| |
| Beyond words | |
| |
| |
| |
| An example ICALL system: TAGARELA | |
| |
| |
| |
| Modeling the learner | |
| |
| |
| |
| Searching | |
| |
| |
| |
| Introduction | |
| |
| |
| |
| Searching through structured data | |
| |
| |
| |
| Searching through unstructured data | |
| |
| |
| |
| Information need | |
| |
| |
| |
| Evaluating search results | |
| |
| |
| |
| Example: Searching the web | |
| |
| |
| |
| How search engines work | |
| |
| |
| Under the Hood 6: A brief tour of HTML | |
| |
| |
| |
| Searching semi-structured data with regular expressions | |
| |
| |
| |
| Syntax of regular expressions | |
| |
| |
| |
| Grep: An example of using regular expressions | |
| |
| |
| Under the Hood 7: Finite-state automata | |
| |
| |
| |
| Searching text corpora | |
| |
| |
| |
| Why corpora? | |
| |
| |
| |
| Annotated language corpora | |
| |
| |
| Under the Hood 8: Searching for linguistic patterns on the web | |
| |
| |
| |
| Classifying Documents: From Junk Mail Detection to Sentiment Classification | |
| |
| |
| |
| Automatic document classification | |
| |
| |
| |
| How computers "learn" | |
| |
| |
| |
| Supervised learning | |
| |
| |
| |
| Unsupervised learning | |
| |
| |
| |
| Features and evidence | |
| |
| |
| |
| Application: Spam filtering | |
| |
| |
| |
| Base rates | |
| |
| |
| |
| Payoffs | |
| |
| |
| |
| Back to documents | |
| |
| |
| |
| Some types of document classifiers | |
| |
| |
| |
| The Naive Bayes classifier | |
| |
| |
| Under the Hood 9: Naive Bayes | |
| |
| |
| |
| The perceptron | |
| |
| |
| |
| Which classifier to use | |
| |
| |
| |
| From classification algorithms to context of use | |
| |
| |
| |
| Dialog Systems | |
| |
| |
| |
| Computers that "converse"? | |
| |
| |
| |
| Why dialogs happen | |
| |
| |
| |
| Automating dialog | |
| |
| |
| |
| Getting started | |
| |
| |
| |
| Establishing a goal | |
| |
| |
| |
| Accepting the user's goal | |
| |
| |
| |
| The caller plays her role | |
| |
| |
| |
| Giving the answer | |
| |
| |
| |
| Negotiating the end of the conversation | |
| |
| |
| |
| Conventions and framing expectations | |
| |
| |
| |
| Some framing expectations for games and sports | |
| |
| |
| |
| The framing expectations for dialogs | |
| |
| |
| |
| Properties of dialog | |
| |
| |
| |
| Dialog moves | |
| |
| |
| |
| Speech acts | |
| |
| |
| |
| Conversational maxims | |
| |
| |
| |
| Dialog systems and their tasks | |
| |
| |
| |
| Eliza | |
| |
| |
| Under the Hood 10: How Eliza works | |
| |
| |
| |
| Spoken dialogs | |
| |
| |
| |
| How to evaluate a dialog system | |
| |
| |
| |
| Why is dialog important? | |
| |
| |
| |
| Machine Translation Systems | |
| |
| |
| |
| Computers that "translate"? | |
| |
| |
| |
| Applications of translation | |
| |
| |
| |
| Translation needs | |
| |
| |
| |
| What is machine translation really for? | |
| |
| |
| |
| Translating Shakespeare | |
| |
| |
| |
| The translation triangle | |
| |
| |
| |
| Translation and meaning | |
| |
| |
| |
| Words and meanings | |
| |
| |
| |
| Words and other languages | |
| |
| |
| |
| Synonyms and translation equivalents | |
| |
| |
| |
| Word alignment | |
| |
| |
| |
| IBM Model 1 | |
| |
| |
| Under the Hood 11: The noisy channel model | |
| |
| |
| Under the Hood 12: Phrase-based statistical translation | |
| |
| |
| |
| Commercial automatic translation | |
| |
| |
| |
| Translating weather reports | |
| |
| |
| |
| Translation in the European Union | |
| |
| |
| |
| Prospects for translators | |
| |
| |
| |
| Epilogue: Impact of Language Technology | |
| |
| |
| References | |
| |
| |
| Concept Index | |