Sentiment Analysis for Product Reviews using Long Short-Term Memory with Layer Normalization

Sentiment Analysis for Product Reviews using Long Short-Term Memory with Layer Normalization

Sentiment analysis is an important tool in the hands of service providers and sellers to gain insights into customer behaviour. The work reported in this paper aims to classify customer sentiment in a sample of 3,000 customer reviews into positive and negative categories, using two methods: long sho...

Full description

Saved in:
Bibliographic Details
Published in:2023 First International Conference on the Advancements of Artificial Intelligence in African Context (AAIAC) pp. 1 - 6
Main Authors: Raghunathan, Nilaa, Shiwakoti, Prarthana, P, Monika, M, Anbarasi, Shanmugam, Siva
Format: Conference Proceeding
Language:English
Published: IEEE 15-11-2023
Subjects:
Artificial intelligence
Emojis
Long Short-Term Memory
Long Short-Term Memory with layer normalization
Reviews
Sentiment analysis
Tokenization
Virtual assistants
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Sentiment analysis is an important tool in the hands of service providers and sellers to gain insights into customer behaviour. The work reported in this paper aims to classify customer sentiment in a sample of 3,000 customer reviews into positive and negative categories, using two methods: long short-term memory (LSTM), and LSTM with layer normalization. Before applying these two methods, the data is pre-processed using data exploration, emoji conversion, lemmatization, and tokenization. The findings indicate that LSTM with layer normalization gives slightly better results as compared to LSTM.
DOI:10.1109/AAIAC60008.2023.10465282