TAILIEUCHUNG - Báo cáo khoa học: "WORD, PHRASE AND SENTENCE"

Among the relative verities of natural language processing are the facts that morphemes and words are primary semantic units, and that their co-ocurrence in phrases and sentences provides cues for selecting sense meanings. At this time the line of history spirals past the points it reached on earlier cycles. Hopefully. | WORD PHRASE AND SENTENCE Rob t F. Simmons Univ of Texas Austin Among the relative verities of natural language processing are the facts that morphemes and words are primary semantic units and that their co-ocurrence in phrases and sentences provides cues for selecting sense meanings. In this session two psycholinguistic studies show some aspects of how human subjects process words while reading. A study of medical vocabulary shows that medical words are highly associated by co-occurrance In medical definitions. Another report shows the effectiveness of keyword Identification and selection of prominent sentences to organize abstracts for retrieval. A fifth study argues that analysis of existing natural language dictionaries can be expected to contribute Importantly to what Is needed for text understanding programs. The final study Is an experiment with a sentence level translator applied to a large German-English translation task. These two studies are primarily concerned with analysis of language at the sentence level. The most glamourous areas of natural language research are at levels above the sentence concerned with dialogues and discourse frequently disdainful of morphological or even grammatical analysis In their search for effective structures for understanding what the discourse Is about. Scripts frames stereotypes schemas are all studied In these areas and often morphological and grammatical analysis Is bypassed In favor of keyword scanning to extract some small relevant portion of the text to be bound as values for slots In these larger data forms. This session reminds US that much can be accomplished with vocabulary analysis with keyword scanning and statistical treatment of text and with semantic analysis at the single sentence level. Yet with regard to most of the topics In this and other sessions there Is a strong sense of deia vu the earliest natural language studies featured automatic extracting and Information retrieval based on statistical lexical

TỪ KHÓA LIÊN QUAN
Đã phát hiện trình chặn quảng cáo AdBlock
Trang web này phụ thuộc vào doanh thu từ số lần hiển thị quảng cáo để tồn tại. Vui lòng tắt trình chặn quảng cáo của bạn hoặc tạm dừng tính năng chặn quảng cáo cho trang web này.