Systematic Review of Text Mining Application Using Apache UIMA

Abstract

Companies are often faced with a number of data and information in the form of unstructured texts. The unstructured data set can be processed / extracted so that it can benefit the company in the decision making process or strategy that must be carried out by the company. Text Mining is one solution to overcome these problems. Text Mining can be defined as the process of retrieving information sourced from several documents. One of the most commonly used text mining tools is Apache UIMA. This study aims to systematically study literature on the implementation of text mining and Apache UIMA by using several related databases, including reviewing text mining, Apache UIMA, and reviewing journal of text mining and Apache UIMA. These journals are reduced using certain criteria. The results obtained are the 20 journals that discuss the implementation of text mining and Apache UIMA. Based on the analysis of these journals, it can be concluded that the application of Text Mining is more widely used in the field of Classification with the method often used is Naive Bayes Classifiers. The average accuracy of the method reaches more than 85%, which means the method is very effective for classification. Specifically, Apache UIMA is more widely implemented in the Information Extraction and NLP fields. The main component of Apache UIMA that is often used is the Annotator Engine and is very effectively implemented for information extraction.