Flexibility of Indonesian text pre-processing library

Abstract

This study aimed to achieve and measure flexibility as a software quality factor of text pre-processing libraries with Indonesian text from social media. Library was built based on a review of some text mining applications that did not yet have a special pre-process for Indonesian text. Text pre-processing libraries were designed and built using an object-oriented approach that was modular to achieve flexibility. Flexibility was measured by the Mc Call Cyclomatic Complexity (CC) metric. Flexibility of library was tested by implementing the library into text mining applications. The results of experiment showed that text pre-processing libraries could be flexible and easy to use without much configuration in text mining applications. It was proved by the value of CC of 2.51 which meant the library or software was not too complex, simple enough, and also flexible to use.