What is computational linguistics?
Computational linguistics is a branch of applied linguistics, dealing with computer processing of human language.
(1) It includes the analysis of language data so as to establish the order in which learners acquire various grammatical rules or the frequency of occurrence of some particular item.
(2) It includes electronic production of artificial speech and the automatic recognition of human speech.
(3) It includes research on automatic translation between natural languages.
(4) It also includes text processing and communication between people and computers.
10.1 Computer-assisted language learning (CALL)
10.1.1 CAL / CAI vs. CALL
CAI (Computer-assisted instruction) means the use of a computer in a teaching program. CAL (Computer-assisted learning) refers to the use of a computer in teaching and learning and in order to help achieve educational objectives. CAI aims at seeing educational problems on the part of the teacher, whereas CAL emphasizes the use of a computer in both teaching and learning. CALL (Computer-assisted language learning) means the use of a computer in the teaching or learning of a second or foreign language. If CAI or CAL deals with teaching and learning problems in general, CALL deals with language teaching and learning in particular.
10.1.2 Phases of CALL development (4 periods)
1. During this period, computers were large mainframe machines kept in research institutions.
2. Small computers appeared and cost cheaper than before, which made a generation of programs possible.
3. The learning was not so much supplied by the language of the text itself as by the cognitive problem-solving techniques and the interaction between students in the group.
4. Instead of writing specific programs for language teaching, word-processing has adapted to language teaching by enabling students to compose and try our their writings in a non-permanent form.
10.1.3 Technology
1. Customizing, template, and authoring programs.
2. Computer networks.
3. Compact disk technology
4. Digitized sound.
10.2 Machine translation (MT)
10.2.1 History of development
1. The independent work by MT researchers
2. Towards good quality output
3. The development of translate tools
10.2.2 Research methods
1. Linguistic approach
2. The practical approaches
(1) The transfer approach
(2) The inter-lingual approach
(3) Knowledge-based approach
10.2.3 MT Quality
10.2.4 MT and the Internet
10.2.5 Spoken language translation
10.2.6 MT and human translation
At the beginning of the new century, it is apparent that MT and human translation can and will co-exist in relative harmony. Those skills which the human translators can contribute will always be in demand.
(1) When translation has to be of “publishable” quality, both human translation and MT have their roles. MT plays an important part in large scale and rapid translation of boring technical documentation, highly repetitive software localization manuals, and many other situations where the costs of human translation are much higher than the ones of MT. By contrast, the human translators are and will remain unrivalled for non-repetitive linguistically sophisticated texts (e.g. in literature and law), and even for one-off texts in specific highly-specialized technical subjects.
(2) For the translation of texts where the quality of output is much less important, MT is often an ideal solution.
(3) For the one-to-one interchange of information, there will probably always be a role for the human translators. But for the translation of personal letters, MT systems are likely to be increasingly used; and, for electronic mail and for the extraction of information from web pages and computer-based information services, MT is the only feasible solution.
(4) As for spoken translation, there must surely always be a market for the human translators. But MT systems are opening up new areas where human translation has never featured: the production of draft versions for authors writing in a foreign language, who need assistance in the translation of information from databases; and no doubt, more such new applications will appear in the future as the global communication networks expand and as the realistic usuality of MT becomes familiar to a wider public.
10.3 Corpus linguistics
10.3.1 Definition
1. Corpus (pl. corpora): A collection of linguistic data, either compiled as written texts or as a transcription of recorded speech. The main purpose of a corpus is to verify a hypothesis about language – for example, to determine how the usage of a particular sound, word, or syntactic construction varies.
2. Corpus linguistics: Corpus linguistics deals with the principles and practice of using corpora in language study. A computer corpus is a large body of machine-readable texts.
10.3.2 Criticisms and the revival of corpus linguistics
10.3.3 Concordance
10.3.4 Text encoding and annotation
1. It should be possible to remove the annotation from an annotated corpus in order to revert to the raw corpus.
2. It should be possible to extract the annotation by themselves from the text.
3. The annotation scheme should be based on guidelines which are available to the end user.
4. It should be made clear how and by whom the annotation was carried out.
5. The end user should be made aware that the corpus annotation is not infallible, but simply a potentially useful tool.
6. Annotation schemes should be based as far as possible on widely agreed and theory-neutral principles.
7. No annotation scheme has a priori right to be considered as a standard.
10.3.5 The roles of corpus data
1. Speech research
2. Lexical studies
3. Semantics
4. Sociolinguistics
5. Psycholinguistics
10.4 Information retrieval (IR)
10.4.1 Scope defined
Data retrieval vs. information retrieval
| Data retrieval | Information retrieval | |
| Matching | Exact match | Partial or best match |
| Inference | Deduction | Induction |
| Data retrieval | Information retrieval | |
| Model | Deterministic | Probabilistic |
| Classification | Monothetic | Polythetic |
| Query language | Artificial | Natural |
| Query specification | Complete | Incomplete |
| Items wanted | Matching | Relevant |
| Error response | Sensitive | Insensitive |
10.4.3 Three main areas of research
1. Content analysis
2. Information structure
3. Evaluation
10.5 Mail and news下载本文
