论文部分内容阅读
This paper presents anew way to extract concept that can beused to improve text classification per-formance (precision and recall). Thecomputational measure will be dividedinto two layers. The bottom layercalled document layer is concernedwith extracting