Emotion Recognition and detection using Sentiment Analysis

Shilpi Chaubey

Assistant Professor

shilpi.chaubey@kalingauniversity.ac.in

Sentiment analysis involves obtaining and examining people’s opinions, ideas, and first impressions about a variety of issues, goods, subjects, and services. Text mining and NLP are used in sentiment analysis to locate and extract subjective information from the text. A vital NLP task that has significant potential for advancements in a variety of fields, including AI, human-computer interaction, etc. is emotion recognition from text.

Human responses to the events are accompanied by physiological ideas known as emotions. It is crucial to analyze these feelings without taking into account speech and facial expressions, and these calls for a supervisory method. As people increasingly interact with one another through the use of abusive text on social media platforms like Facebook, Twitter, and others, it is important to recognize the range of human emotions.

To categories expressing sentiments into positive or negative emotions, various deep learning approaches are applied. Convolutional NN, Long Short Term Memory networks, and Recurrent NN (RNN) will all be used in this example to demonstrate how to attain high emotion classification accuracy through experimentation and evaluation on various data sets.

Sentiment analysis can be characterized as a classification approach used to text data after extensive pre processing. An analysis of the sentiment in a text using computation is called sentiment analysis. It is a potent machine learning application for categorizing text data into several groups. It is used to determine feelings or the polarity of information or to analyze customer feedback.

Data collection, pre-processing, feature selection, and sentiment analysis are the steps in the work.

Focus is placed on various deep learning models used in sentiment analysis, including CNN, LSTM, and RNN. The text’s emotion classification labels into positive and negative sentiments are trained and predicted using the extracted features that are input into CNN, LSTM, and RNN. Under the umbrella of NLP, sentiment analysis is a vast statistical analytical subfield. In many different applications, including feedback analytics, a recommendation system, and many others, sentiment analysis is quite useful. We should obtain more precise results for emotion detection and recognition based on the various deep learning models that are being applied to the featured text.

References:

1)    Mayur Wankhade1,2 · Annavarapu Chandra Sekhara Rao1,2  · Chaitanya Kulkarni1,2 A survey on sentiment analysis methods, applications, and challenges Artifcial Intelligence Review (2022) 55:5731–5780 https://doi.org/10.1007/s10462-022-10144-1

2)    Dipak R. Kawade#1, Dr.Kavita S. Oza*2 # Department of ComputerScience, Sangola College, Sangola Dist-Solapur (MS) India 1 dipakkavade@gmail.com * Department of ComputerScience, Shiveji University, Kolhapur (MS) India 2 skavita.oza@gmail.com Sentiment Analysis: Machine Learning Approach ISSN (Print) : 2319-8613 ISSN (Online) : 0975-4024

3)    Raheesa Safrin1, K.R.Sharmila2, T.S.Shri Subangi3, E.A.Vimal4 SENTIMENT ANALYSIS ON ONLINE PRODUCT REVIEW e-ISSN: 2395 -0056 Volume: 04 Issue: 04 | Apr -2017 www.irjet.net p-ISSN: 2395-0072

4)    Shilpa P C ,Rissa Shereen ,Susmi Jacob ,Vinod P Computer Science & Engineering SCMS School Of Engineering and Technology vinodp@scmsgroup.org  SENTIMENT ANALYSIS USING DEEP LEARNING Proceedings of the Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV 2021). IEEE Xplore Part Number: CFP21ONG-ART; 978-0-7381-1183-4