作者: Peter Willett
DOI: 10.1016/0306-4573(85)90107-4
关键词:
摘要: Abstract Term discrimination values have been suggested as an effective means for the selection and weighting of index terms in automatic document retrieval systems. This paper reports algorithm calculation term that is sufficiently fast operation to permit use exact values, rather than approximate studied previous work. Evidence presented show relationship between frequency crucially dependent upon type inter-document similarity measure used values.