Jurnal Pembangunan Model Klasifikasi Sentimen Teks Bahasa Indonesia Menggunakan Library Spacy

Jurnal Pembangunan Model Klasifikasi Sentimen Teks Bahasa Indonesia Menggunakan Library Spacy

Download Jurnal Disini

Ginting, Jegar Sahaduta
  

(2018)

Development of the Indonesian Text Sentiment Classification Model Using the Spacy Library.

    S1 thesis, UAJY.
  

  

Abstract

Natural Language Processing is one branch of Artificial science
Intelligence that focuses on natural language processing. Natural Language
Processing in the world of information technology is increasing and developing.
Generally the development of NLP occurs rapidly in language processing
English. As for the development of NLP in Indonesian still
too little. One of them is about text classification to find
sentiment sentences.
This research will make an NLP model to classify
Indonesian sentences containing sentiments are positive, negative and neutral
who uses the Spacy library. The data used consists of the corpus
Liepzig University in Germany and Indonesian sentence data
taken from movie reviews, app reviews, netizen comments on Youtube,
Facebook, Instagram and Twitter totaling 2550 sentence data. The data
then divided into 2100 training data and 450 test data.
From the model that was designed then an evaluation was made and found
accuracy of 83.80%. So that it can be concluded that the model is made
it's good enough. This model is then applied to the web service
using the Flask framework so that it can be used on various platforms.

Actions (login required)

 View Item View Item

Download Jurnal Disini