ACL RD-TEC 1.0 Summarization of W95-0110
Paper Title:
INVERSE DOCUMENT FREQUENCY (IDF): A MEASURE OF DEVIATIONS FROM POISSON
INVERSE DOCUMENT FREQUENCY (IDF): A MEASURE OF DEVIATIONS FROM POISSON
Authors: Kenneth Church and William Gale
Primarily assigned technology terms:
- author identification
- bayesian discrimination
- categorization
- disambiguation
- identification
- information retrieval
- information retrieval system
- keyword retrieval
- language modeling
- modeling
- recognition
- retrieval system
- search
- speech recognition
- text categorization
- text compression
- weighting
- word-sense disambiguation
Other assigned terms:
- approach
- bag of words
- case
- community
- concept
- correlations
- density function
- distribution
- document
- document frequency
- entropy
- fact
- genre
- hypothesis
- independence assumption
- information retrieval community
- information theory
- inverse document frequency
- keyword
- measure
- measures
- natural language
- negative binomial
- ngram
- noise
- nouns
- null hypothesis
- poisson distribution
- probabilities
- probability
- process
- search space
- style
- terms
- text
- theory
- word
- word corpus
- word frequency
- words
