Open AccessTwitter Sentiment Analysis as an Evaluation and Service Base On Python Textblob
Author(s) -
I Gede Susrama Mas Diyasa,
Ni Made Ika Marini Mandenni,
Mohammad Idham Fachrurrozi,
Sunu Ilham Pradika,
Kholilul Rachman Nur Manab,
Nyoman Rahadi Sasmita
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1125/1/012034
Subject(s) - social media , sentiment analysis , tag cloud , advertising , computer science , python (programming language) , world wide web , customer base , service (business) , visualization , business , marketing , artificial intelligence , operating system
The development of technology in the current era is very rapid, this is indicated by the many social media that have sprung up. One popular social media is Twitter. Twitter initially became a forum for social media users as a place to preach activities, discuss, and share stories between users. However, now Twitter is even a place for complaints for customers of a company, one of which is PT Telkom Indonesia. Some customers prefer not to contact the call centre that has been provided by the company to be contacted if there is a problem, but prefer to complain via Twitter. According to data taken during a certain period, 3324 tweets were obtained, which included the keywords indihome, myindihome, useetv, and wifi.id. The tweets data that has been collected, if processed properly, will be valuable information for the company. For example, as a reference to assess brand image, customer feedback, and marketing opportunities. This study classifies tweets where the keywords indihome, myindihome, useetv, and wifi.di. Furthermore, several data preprocessing techniques were carried out, sentiment analysis, and visualization in the form of histograms, pie charts, and word clouds. From 3324 tweets that have been analyzed, it is found that there are 34.4% positive tweets, 16.1% negative tweets, and 49.6% neutral tweets.