The　document　clustering　technique　is　widely　recognized　as　a　useful　tool　for　informationretrieval，　organizing　web　documents，　text　mining　and　so　on．　The　purpose　of　this　paper　is　toreview　various　document　clustering　techniques，　and　to　discuss　research　issues　for　enhancingeffectiveness　or　efficiency　of　the　clustering　methods．　We　explore　extensive　literature　onnon－hierarchical　methods　（single－pass　methods），　hierarchical　methods　（single－link，　completelink，　etc．），　dimensional　reduction　methods　（LSI，　principal　component　analysis，　etc．），　probabilisticmethods，　data　mining　techniques，　and　so　on．　ln　particular，　this　paper　focuses　on　typicaltechniques，　such　as　the　k－means　algorithm，　the　leader－follower　algorithm，　self－organizing　map（SOM），　single一　or　complete－link　methods，　bisecting　k－means　methods，　latent　semantic　indexing（LSI），　Gaussian－Mjxture　model　and　so　on．　After　reviewing　the　techniques　and　algorithms，　wediscuss　research　issues　on　document　clustering；　computational　complexity，　feature　extraction（selection　of　words），　methods　for　defining　term　weights　and　similarity，　and　evaluation　of　results．

展望論文