Comparative Analysis of Convolutional Neural Network and LSTM in Text-Based Sentiment Classification

Kalaivani, M. S. and Jayalakshmi, S. (2021) Comparative Analysis of Convolutional Neural Network and LSTM in Text-Based Sentiment Classification. In: 2021 Fifth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC), Palladam, India.

Full text not available from this repository. (Request a copy)

Abstract

Sentiment analysis is a type of natural language processing that categorizes text as positive or negative. It has been used in several industries in recent decades. Analyzing people’s opinions on a subject is a difficult task, because everyone’s perception is different. Sentiment analysis can be used to classify the information about people, services, and products. Every minute, a vast amount of data is uploaded to websites, blogs, and social media platforms. In this data era, new technologies and algorithms are required to extract useful information from online content. Customer expectations, market trends, and people’s attitudes are determined by sentiment analysis. It aids in the expansion of a business and making of critical production decisions. While industries utilize social media to interact with customers, businessmen use customer feedback and reviews to better their services and products. Machine learning and Deep learning models are well suited for text processing. Convolutional neural networks and Recurrent neural networks are two deep learning methods that are utilized to solve numerous challenges in the text processing sector. CNN can get the important features from the input text with pooling operation, but it is complex to get contextual data. RNN can get the contextual detail but the sequence of input words will differ. This study discusses the comparison of Convolution neural networks with LSTM algorithms.

Item Type: Conference or Workshop Item (Paper)
Subjects: Computer Applications > Database Management System
Divisions: Computer Applications
Depositing User: Mr IR Admin
Date Deposited: 09 Oct 2024 12:11
Last Modified: 09 Oct 2024 12:11
URI: https://ir.vistas.ac.in/id/eprint/9615

Actions (login required)

View Item
View Item