Title: Opinions cum sentiments of the textual content through ML methods: a smart city-based approach in 21st century using NLP

Authors: Rohit Rastogi; Yati Varshney; Sonali Jaiswal; Markandey Sharma

Addresses: Department of CSE, ABES Engineering College Ghaziabad, UP, India ' Department of CSE, ABES Engineering College Ghaziabad, UP, India ' Department of CSE, ABES Engineering College Ghaziabad, UP, India ' Department of CSE, ABES Engineering College Ghaziabad, UP, India

Abstract: The ambition of this research is to advance a functional classifier for automatic and error-free sentiment classification of an unknown tweet stream. In the past decades, new shape of communication, such as text messaging and microblogging have come up and become universal. Beside there is no range of information to be transmit by text and tweets, often these message are used to give opinions and sentiment that people have around what is going around on the world. The researcher team also worked on the following task. The task was to give a message, so as to categorise whether the message is of positive, negative or neutral sentiment. The corpus-base method was used to find semantic orientation of dictionary-based method and adjectives to find the semantic orientation of adverb and verbs. Algorithms used in this research are Naive Bayes classifier, decision tree, AdaBoost classifier, random forest, support vector machine, KNN classifier, 7XGBOOST and logistic regression. The work in this paper gives a novel approach for sentiment analysis on data. The initial result shows it is a motivating technique.

Keywords: KNN; TF-IDF; vectoriser; logistic regression; classifier; svcl_score; Scikit;df1; python; natural Language processing; NLP; text data.

DOI: 10.1504/IJCRC.2023.133549

International Journal of Creative Computing, 2023 Vol.2 No.1, pp.41 - 72

Received: 24 Dec 2022
Accepted: 13 May 2023
Published online: 20 Sep 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article