An Optimization of Multi-Class Document Classification with Computational Search Policy

In the era of internet communication, many electronic documents are spread and flow on the platform of website in every splits of seconds. The research interest for the process of knowledge discovery is changed from the traditional data to online data such as online news document classification. Most percentage of the online data is text document and therefore the optimization of multi-class document classification is becoming a challenge for today society. Traditional search policy for feature selection process is degrading with exhaustive search for complex feature in document classification. Therefore, meta-heuristic based computational search is also becoming good solution to overcome the problem of exhaustive search with exploitation process. The search policy of computational algorithm can provide the global optimal solution with random search approach on both exploitation and exploration process, and the selected search results of feature subsets can support the optimal classification results.