Sentimen Analisis Pengguna Media Sosial Berdasarkan Metode Ekstraksi Fitur dan Klasifikasi

Abstrak

Sentiment analysis is a combination of various research fields like NLP (natural language processing), data mining, and text mining to find people's opinions expressed in text form. There are several tasks in sentiment analysis like as sentiment extraction, sentiment classification, and summarization. There are several challenges in carrying out sentiment analysis including synonyms and polysemy, sarcasm, compound sentences, and unstructured data. The purpose of this writing is to review other research on Sentiment Analysis based on datasets, feature selection, and classification algorithms as well as the use of multilabel sentiment analysis, as well as evaluation of accuracy results, to get the best approach to the selection of methods used in mining processing.