Text Mining in Bahasa Indonesia - a Review

Peneliti

Nelly Indriani Widiastuti, M.T. Kania Evita Dewi, S.Pd., M.Si

Deskripsi/Abstrak

This study aims to obtain an overview of the research conducted by researchers in text mining in documents Indonesian. Besides, this article also aims to describe what others researches who want to develop text mining technology can do. The research discussed in this article is collected, then extracted to look for similarities, differences, domains of cases handled, methods used, and problems that still arise. Based on reviews from various reputable scientific journal sources, the results show that research in text mining technology has developed rapidly, especially in analytical sentiment. There are some efforts to improve the quality of text mining resources in Indonesian. Even so, there are still many shortcomings that cannot be overcome by text mining researchers. Currently, language skills in an intelligent agent are difficulties to match humans. Besides, language intelligence is computer science objects that are always changing due to increasing or experiencing absorption from foreign languages. That makes the complexity of the algorithm becomes more complicated. Based on the papers that have been compiling, it turns out that there is still much that text mining researchers can do on Indonesian language documents. The problems encountered in network architectures, preprocess, datasets and resources. These results are important to be the object of further research