COMPARISONALGORITHMCLASSIFICATION NAIVE BAYES, DECISION TREE, ANDNEURAL NETWORK FORANALYSIS SENTIMENT

Ahmad Turmudi Zy, Agung Nugroho

Abstract


Twitter is a social media,that can be accessed by anyone as long as they have a supporting gadget and an internet connection, nowadays everyone can tweet anywhere and anytime using their gadget, whether it is a cellphone, iPhone, tablet, iPad, or tab. Because of the ease of access, everyone from any background can comment and create a status to comment on a product, event or character. This thesis will specialize on tweets with regards to characters, especially on the Governor election of Jakarta on 2017. Every tweets must be different but the content there is only three content in a tweet, positive, negative and neutral. These three contents reflect on the sentiments of the tweeter. Based on the contents of the tweet, it will be divided into it‘s characters. These characters will use feature selection using algorithmic theory ofNaive Bayes Classifier, Decision Tree, Neural Networkto automatically classify the sentiment. This research will use 1200 tweet about the sentiment of the Jakarta Election. These data will be classified manually and divided into 400 data each for positive, negative and neutral. And then 1200 data will be used for testing. Hopefully the comparison will get the best algorithm that is accurate in classifying the sentiments.

Keywords


Twitter, Tweet, Sentiment, Sentiment Analysis, Naive Bayes Classifier, Decision Tree, Neural Network

Full Text:

PDF

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.