This work basically focuses on analyzing the efficiency of different classification techniques on different data sets and impact of applying different feature extraction techniques along with those classification techniques. After experimenting with all the classification algorithms and dimensionality reduction techniques it has been found that the combination of NB and PCA outperforms than other combinations of models. Hence, we can infer NB-PCA is the best model among all the classification models used. In further studies, if more data can be gathered the classifiers can be trained with higher efficiency as well as more accurate results. In future, more number of classifiers along with FE techniques can be used for the comparative analysis in a broader sense. Then we can conclude with some more accurate combinations of models.
展开▼