Repository files navigation

• Live stream Twitter API
• https://github.com/keyreply/Bahasa-Indo-NLP-Dataset
• https://www.kaggle.com/ilhamfp31/indonesian-abusive-and-hate-speech-twitter-text
• Indonesia Corpushttps://dumps.wikimedia.org/idwiki/latest/ (idwiki-latest-pages-articles.xml.bz2)

• Natural Language Processing (NLP) is a hotbed of research in data science these days and one of the most common applications of NLP is sentiment analysis
• From opinion polls to creating entire marketing strategies, this domain has completely reshaped the way businesses work
• Thousands of text documents can be processed for sentiment in seconds, compared to the hours it would take a team of people to manually complete the same task

• Indonesia has so many sub-languages and this is a big homework to build a corpus that can provide all of it for mapping every each word that has common similarity with other words
• When we talk about N-grams bag of words, sometimes stopwords can be meaningful
• Misspelling is the most common issue for Bahasa especially in social media, that's why we need to build a large library that can manage every word so stemming words can be an easy task
• Satire, this is so painful work that our model can't distinguish the meaning of sentences
• Word2Vec and GloVe as the recommend algorithms to build a robust word embedding model