Automatic Categorization of Amharic News Text: a Machine Learning Approach

Currently news items subject classification in Ethiopia is done manually by journalists which is time consuming task (although they are using computer system to store and dispatch information). This research experimented the application of machine learning techniques to automatic categorization of Amharic news items. Machine learning techniques, Naïve Bayes and k Nearest Neighbor classifiers, were used to categorize the Amharic news items. 11, 024 news articles were used to do this research. To come up with good results text preparation and per-processing was done. Stop-word and words that occur in 3 or less documents were removed from the collection. Thirty-three percent of the data was used for testing purposes. The result of this research indicated that such classifiers are applicable to automatically classify Amharic news items. However, the classifiers work well when the categories contain almost evenly distributed news items. The best result obtained is by the naïve Bayes. The result of this research is promising. Nevertheless, additional works are recommended in order to come up with good result.