An In-Depth Analysis of Text Clustering Techniques for Identifying Potential Insurance Customers on Social Media: A Machine Learning Perspective

Liew, Chun Kin and Goh, Ching Pang (2023) An In-Depth Analysis of Text Clustering Techniques for Identifying Potential Insurance Customers on Social Media: A Machine Learning Perspective. INTI JOURNAL, 2023 (71). pp. 1-6. ISSN e2600-7320

[img] Text
ij2023_71.pdf - Published Version
Available under License Creative Commons Attribution.

Download (214kB)
Official URL: https://intijournal.intimal.edu.my

Abstract

Social media has emerged as a transformative platform for the exchange and dissemination of information. Unlike conventional sources such as online news, social media often offers more real-time and current updates. Effectively harnessing the vast and diverse pool of unstructured data on these platforms requires the extraction of structured information. This research focuses on the development of a social media web crawler, coupled with the implementation of sophisticated algorithms like Web Content Mining, Noisy Text Filtering, Named Entity Extraction, Part-Of-Speech (POS) Tagging, and Text Clustering. The aggregated information will be utilized to train a machine learning model capable of discerning a customer's preferred insurance type—be it accident, health, car, or life insurance. The overarching objective is to provide insurance companies with a swift, precise, and cost-effective means of identifying potential customers within the realm of social media. The result shows that this new technique has successfully identify relevant topic based on the comments and recommend corresponding insurance to the user

Item Type: Article
Uncontrolled Keywords: LDA bag of words, LDA TF-IDF, insurance, machine learning
Subjects: Q Science > Q Science (General)
Q Science > QA Mathematics > QA76 Computer software
Depositing User: Unnamed user with email masilah.mansor@newinti.edu.my
Date Deposited: 12 Dec 2023 06:44
Last Modified: 12 Dec 2023 06:52
URI: http://eprints.intimal.edu.my/id/eprint/1884

Actions (login required)

View Item View Item