Learning Indonesian Frequently Used Vocabulary from Large-Scale News

摘要

Frequently used vocabulary of a language plays an important role in language learning. In this work, we introduce how we obtain Indonesian frequently used vocabulary along with a vocabulary level scheme. Techniques in natural language processing and statistics are used to process a large scale of Indonesian news. According to our scheme, an Indonesian language learner may need to learn about 11,200 words to gain adequate comprehension when reading an Indonesian article. In addition, we also compare the vocabulary distribution and usage between English and Indonesian. Our work could be helpful to Indonesian language learners and other relevant research such as language teaching.