INTEGRATING BINARY PARTICLE SWARM OPTIMIZATION AND GENETIC ALGORITHM FOR FEATURE SELECTION IN TEXT CLASSIFICATION
碩士 === 大同大學 === 資訊經營學系(所) === 99 === Feature selection has been proved to be very important for classification. There are some statistical approaches, such as Information Gain, Mutual Information, andχ2, etc. The features are measured one by one, therefore the inference of combination of features is...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2011
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/59314007796932477306 |
| Summary: | 碩士 === 大同大學 === 資訊經營學系(所) === 99 === Feature selection has been proved to be very important for classification. There are some statistical approaches, such as Information Gain, Mutual Information, andχ2, etc. The features are measured one by one, therefore the inference of combination of features is not considered. In recent years, the evolution-based computing algorithms have been involved in feature selection to search the best combination of features. This research presents a feature selection algorithm integrating Binary Particle Swarm Optimization and Genetic Algorithm for Text Classification. The objective is to find the global optimal features for high dimensional data classification, especially for text classification.
|
|---|