2007/04/10
We had the regular meeting in this morning. Afterward I coded a program for the demand of the AutoClass. In this program, I used the technical terms which were extracted from scoring function as the input feature and assigned the boolean value as the feature value for each technical term. Because the AutoClass requires three different documents, I have spent much time for encoding this program. In order to get the high performance on clustering. Thus, I will adopt various feature value such as, tf*idf and score.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment