Quantifying lexical influence: Giving direction to context

Quantifying lexical influence: Giving direction to context V Krip~sundar kripa~cs, buffalo, edu CEDAR & Dept. of Computer Science SUNY at Buffalo Buffalo NY 14260, USA Abstract The relevance of context in disambiguat- ing natural language input has been widely acknowledged in the literature. However, most attempts at formalising the intuitive notion of context tend to treat the word and its context symmetrically. We demonstrate here that traditional measures such as mu- tual information score are likely to overlook a significant fraction of all co-occurrence phenomena in natural language. We also propose metrics for measuring directed lex- ical influence and compare performances. Keywords: contextual post-processing, defining context, lexical influence, direc- tionality of context 1 Introduction It is widely accepted that context plays a significant role in shaping all aspects of language. Indeed, com- prehension would be utterly impossible without the extensive application of contextual information. Ev- idence from psycholinguistic and cognitive psycho- logical studies also demonstrates that contextual in- formation affects the activation levels of lexical can- didates during the process of perception (Weinreich, 1980; McClelland, 1987). Garvin (1972) describes the role of context as follows: [The meaning of] a particular text [is] not the system-derived meaning as a whole, but that part of it which is included in the con- textually and situationally derived mean- ing proper to the text in question. (p. 69- 70) In effect, this means that the context of a word serves to restrict its sense. The problem addressed in this research is that of improving the performance of a natural-language recogniser (such as a recognition system for hand- written or spoken language). The recogniser out- put typically consists of an ordered set of candidate words (word-choices) for each word position in the input stream. Since natural language abounds in contextual information, it is reasonable to utilise this in improving the performance of the recogniser (by disambiguating among the word-choices). The word-choices (together with their confidence values) constitute a confusion set. The recogniser may further associate a confidence-value with each of its word choices to communicate finer resolution in its output. The language module must update these confidence values to reflect contextual knowledge. 2 Linguistic post-processing The language module can, in principle, perform several types of "post-processing" on the word- candidate lists that the recogniser outputs for the different word-positions. The most promising possi- bilities are: • re-ranking the confusion set (and assigning new confidence-values to its entries), and, • deleting low-confidence entries from the confu- sion set (after applying contextual knowledge) Several researchers in NLP have acknowledged the relevance of context in disambiguating natural lan- guage input ((Evett et al., 1991); (Zernik, 1991); (Hindle & Rooth, 1993); (Rosenfeld, 1994)). In fact, the recent revival of interest in statistical language processing is partly because of its (comparative) suc- cess in modelling context. However, a theoretically sound definition of context is needed to ensure that such re-ranking and deleting of word-choices helps and not hinders (Gale & Church, 1990). Researchers in information theory have come up with …

Content truncated to 500 words