Detecting emotions in tweets based on hybrid approach
Date Issued
2018
Author(s)
Najdenkoska, Ivona
Stojanovska, Frosina
Abstract
Emotion detection from text is increasingly popular
nowadays, especially when it comes to human-computer
interaction. It is one of the great areas for recognition of the
human emotional state and it has a potential application in many
other vast areas such as computer vision, psychology, physiology
etc. In this paper, we will try to recognize emotions from posts on
the popular social network Twitter also known as tweets. The
emotions will be represented with four classes of emotions: anger,
fear, joy, and sadness, with additional neutral class, and we will
try to recognize them. For solving the problem, we will use a
hybrid approach. This approach incorporates concepts of two
major areas, natural language processing (NLP) with its
linguistic models and more diverse machine learning (ML)
algorithms.
nowadays, especially when it comes to human-computer
interaction. It is one of the great areas for recognition of the
human emotional state and it has a potential application in many
other vast areas such as computer vision, psychology, physiology
etc. In this paper, we will try to recognize emotions from posts on
the popular social network Twitter also known as tweets. The
emotions will be represented with four classes of emotions: anger,
fear, joy, and sadness, with additional neutral class, and we will
try to recognize them. For solving the problem, we will use a
hybrid approach. This approach incorporates concepts of two
major areas, natural language processing (NLP) with its
linguistic models and more diverse machine learning (ML)
algorithms.
Subjects
File(s)![Thumbnail Image]()
Loading...
Name
Detecting-emotions-in-tweets-based-on-hybrid-approach.pdf
Size
351.47 KB
Format
Adobe PDF
Checksum
(MD5):95904eb033efdd13acd3968b9923353f