Rule-Based Text Simplification for Extractive Automatic Summarization of News Articles in Indonesian
M. Abdi Haryadi. H, Masayu Leylia Khodra, Anggrahita Bayu Sasmita · 2023
Text simplification is one of the techniques to improve the performance of an automatic summarization. Currently, the research about Indonesian text simplification is few. In this paper, we adapted rule-based text simplification for Indonesian language, and integrated it to the extractive automatic summarization of news articles in Indonesian. This integration was conducted to investigate its influence in automatic summarization performance. We adapted a text simplification method that was originally designed for the English language to build the Indonesian text simplification module. Additionally, we built a text simplification dataset based on an existing dataset to evaluate our module. Then, we evaluated the performance of the previous extractive automatic summarization of news articles in Indonesian using our module. We show that the text simplification module for the Indonesian language needs to be adjusted for the cases that are not addressed in the English language. With the text simplification module, the performance of automatic summarization of news articles in Indonesian is decreased in recall, but it is increased in precision.