From A Property Of The Average Of Fractions To A Text-Processing Interface

摘要: The average of two natural numbers always falls between those numbers. Partitioning a document set into non overlapping subsets, the words in will appear only one subset or on both. These properties can be used to present users with choices that allow them build phrase where chosen terms have context. frequency term study relevance by comparing it same term's relevant and non-relevant subsets. A coefficient variability (VAR) is defined as normalized distance these values. vast majority seem not significant because VAR their relative comparative value minimal. But few show very high This could exploited system find strong word instances represent concepts. It possible imagine an iterative procedure based through which user identifies Such would desirable for diversity text-related computer-based tasks such content analysis, thesauri construction, data mining, indexing feature selection. software instance help context has been developed prototype concept. Knowledge related given requires support structure some cognitive elements other knowledge, data, concepts, information, etc. Using objective subjective measures, derive conceptual relationships. Users gauge topical engaging system, then offer more suggestions. There great advantage reducing load extraneous information. deal directly easier identify words, phrases combinations form information capsules.