Implementation of Recurrent Network for Emotion Recognition of Twitter Data

Implementation of Recurrent Network for Emotion Recognition of Twitter Data

Anu Kiruthika M., Angelin Gladston
Copyright: © 2020 |Pages: 13
DOI: 10.4018/IJSMOC.2020010101
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

A new generation of emoticons, called emojis, is being largely used for both mobile and social media communications. Emojis are considered a graphic expression of emotions, and users have been widely used to express their emotions in social media. Emojis are graphic unicode symbols used to express perceptions, views, and ideas as a shorthand. Unlike the small number of well-known emoticons carrying clear emotional content, hundreds of emojis are being used in different social networks. The task of emoji emotion recognition is to predict the original emoji in a tweet. Recurrent neural network is used for building emoji emotion recognition system. Glove is a word-embedding method used for obtaining vector representation of words and are used for training the recurrent neural network. This is achieved by mapping words into a meaningful space where the distance between words is related to semantic similarity. Based on the word embedding in the Twitter dataset, recurrent neural network builds the model and finally predicts the emoji associated with the tweets with an accuracy of 83%.
Article Preview
Top

1. Introduction

Developing social network platforms has given people a new way of generating and consuming a lot of web-based information (Dixit et. al., 2017). People used to obtain information from portal websites in the past. A large number of websites today provide information on a long list of subjects that vary from politics to entertainment. These traditional online information sources are always useful, but are less efficient since they often contain redundant information. Due to the arrival of online social network platforms, people tend to get information from them in a faster pace and is more efficient. These platforms are available for users to choose the information source they are interested in. To mention, large number of social network platforms such as, Twitter, Google+, and Facebook provide information to users (Geetha et. al., 2019; Xiong et. al., 2018).

The most popular platform for microblogging is Twitter. It is one of the fastest growing social network platforms and has a dominant microblogging position. Every day, more than 500 million registered users post 340 million twitter messages (Dixit et. al., 2017; Mohammad et. al., 2015), sharing their views and activities every day. Twitter posts are much shorter than those on regular microblogging platforms (Pak et. al., 2010; Pennington et. al., 2014). Only 140 characters or less can be posted in one twitter message (Dixit et. al., 2017). This feature makes twitter easier and keeps it distinct from the massive amount of information available online for people to get the main point. In twitter, communication is made through messages commonly referred as the tweets. In this social website, people are allowed to make posts about different things, thus enabling people to get their required information from the massive amount of information available.

Twitter users can follow whatever people and source of information they prefer, depending on the users' needs. Twitter has therefore become a powerful platform with many kinds of information from worldwide breaking news to buying products at home, with all the benefits mentioned above. The information streams on twitter have experienced an incredible increase in the popularity of social network over the past few years. Users have a huge amount of information on various aspects (Unnisa et. al., 2016). Not all the information is useful to users however, and each user has their own interests and preferences. There is urgency for users to have personalized services. Nowadays, more and more personalized services are provided to benefit the users. People need this personalized service to make their fast-paced lives more efficient.

Every day, users are publishing a large amount of information on the twitter platform. Twitter data is related to the behaviour of the user and therefore many research studies focus on twitter and its collection of data. One of the twitter based research is user modelling. Researchers started to explore rankings and recommendations of twitter-referenced web resources to provide a personalized service. Based on their published tweets, a large amount of research focuses on modelling users and interests. Microblogs such as Twitter and SinaWeibo are a kind of popular social media (Pennington et. al., 2014) in which millions of people express their feelings, emotions, and attitudes. Because a large number of microblog posts are generated on a daily basis, the mining of feelings from this data source helps to perform research on various topics, such as analysing brand reputation, predicting the stock market, and detecting abnormal events.

Complete Article List

Search this Journal:
Reset
Volume 16: 1 Issue (2024)
Volume 15: 1 Issue (2023)
Volume 14: 2 Issues (2022)
Volume 13: 2 Issues (2021)
Volume 12: 2 Issues (2020)
Volume 11: 2 Issues (2019)
View Complete Journal Contents Listing