Text Classification Using Genetic Programming with Implementation of Map Reduce and Scraping

Classification of text documents on online media is a big data problem and requires automation. Text classification accuracy can decrease if there are many ambiguous terms between classes. Hadoop Map Reduce parallel processing framework for that has been widely used data. The study presented using genetic programming by pre-processing map-reduce collecting web scraping. Genetic to perform association rule mining (ARM) before analyze patterns. articles from science-direct with the three keywords. This aims ARM-based pattern analysis collection system through web-scraping, map-reduce, programming. Through scraping, collected reducing duplicates as much 17718. Map-reduce tokenized stopped-word removal 36639 5189 unique 31450 common terms. Evaluation ARM different amounts multi-tree produce more longer rules better support. also produces specific performance than single tree. evaluation shows tree (0.7042) decision (0.6892), lowest multi-tree(0.6754). results not in line results, where best result (0.3904) (0.3588), (0.356).