Research Article
A Novel Feature Selection Technique for Text Classification Using Naïve Bayes
Table 7
Comparison of proposed method with other classifiers.
| Datasets | NB + FS-CHICLUST | Naïve Bayes | DT | SVM | kNN |
| CNAE-9 | 68% | 19% | 46% | 83% | 77% | Hotel | 87% | 40% | 47% | 60% | 53% | Gender | 69% | 57% | 59% | 67% | 62% | Prosncons | 76% | 50% | 65% | 72% | 71% | CookWare | 80% | 46% | 47% | 73% | 53% | MyMail | 93% | 48% | 78% | 88% | 85% | Reuters | 76% | 12% | 73% | 73% | 56% | Computers | 93% | 38% | 47% | 67% | 60% | Flipkart | 84% | 39% | 63% | 71% | 67% | SpamHam | 95% | 13% | 87% | 90% | 90% | Books | 93% | 40% | 40% | 53% | 47% | DBTotal | 90% | 55% | 75% | 85% | 65% | NYtdm | 48% | .4% | 24% | 6% | 25% |
|
|