Open AccessMachine Classification for Suicide Ideation Detection on Twitter
Author(s) -
Maidam Manisha*,
Anuradha Kodali,
V. Srilakshmi
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.l3655.1081219
Subject(s) - suicidal ideation , social media , naive bayes classifier , machine learning , random forest , automatic summarization , decision tree , computer science , support vector machine , artificial intelligence , feeling , classifier (uml) , sentiment analysis , world wide web , internet privacy , psychology , suicide prevention , poison control , social psychology , medicine , environmental health
The tremendous rise in technology and social media sites enabled everyone to express and share their thoughts and feelings with millions of people in the world. Online social networks like Google+, Instagram, Facebook, twitter, LinkedIn turned into significant medium for communication. With these sites, users can generate, send and receive data among large number of people. Along with the advantages, these platforms are having few issues about its user safety such as the build out and sharing suicidal thoughts. Therefore, in this paper we built a performance report of five Machine Learning algorithms called Support Vector Machine, Random Forest, Decision Tree, Naïve Bayes, and Prism, with the aim of identifying, classifying suicide related text on twitter and providing to the research related to the suicide ideation on communication networks. Firstly, these algorithms identify the most worrying tweets such as suicide ideation, reporting of suicidal thoughts, etc. Also, find outs the flippant to suicide. Along with ML classifiers, One of the most powerful NLP technologies i.e: Opinion summarization is used to classify suicidal and non-suicidal tweets. The outcome of the analysis representing that Prism classifier achieved good accuracy by observing emotions of people and extracting suicidal information from Twitter than other machine learning algorithms