Abstract
In order to describe the user's interest more exactly and provide for him what he needs, a new method to weight and optimize the user's characterized interest in information filtering is presented. The paper defines the weights of the terms according to the order of the proposed sequence, optimizes the interest with synonyms in rough set theory, calculates the semantic similarity between interest and documents to be filtered, sorts them in descending order of their similarity, and submits them to the user. Experiments show that it has the higher precision.
| Original language | English |
|---|---|
| Pages (from-to) | 10-12 |
| Number of pages | 3 |
| Journal | Jisuanji Gongcheng/Computer Engineering |
| Volume | 32 |
| Issue number | 20 |
| Publication status | Published - 20 Oct 2006 |
Keywords
- Information filtering
- Regress
- Rough sets
- Similarity
- Synonym optimization