NORMA eResearch @NCI Library

Financial Complaints: Sentiment Analysis: Final Technical Report

Moleka, Arlon Junior (2017) Financial Complaints: Sentiment Analysis: Final Technical Report. Undergraduate thesis, Dublin, National College of Ireland.

[thumbnail of Bachelor of Science]
Preview
PDF (Bachelor of Science)
Download (5MB) | Preview

Abstract

In this project, we are going to be analysing a large file of dataset (consumers’ complaints) to make sure that consumers are treated fairly by financial companies such as banks, lenders and so on. This data is made available by the Consumer Financial Protection Bureau, this dataset provides consumers’ experience in their own words explaining what happened in different sectors of finance like in mortgage services, prepaid card services, student loan and so on.

To do so, we analyse consumers’ complaints on specific services (Prepaid Cards and Student Loan) so by adding their consents, consumers help to improve the financial marketplace for better experience in the future and companies have a clear understanding in what must be done to improve their services or products to meet consumers’ need.

Measuring sentiment will help us to understand the overall feeling of consumers on a specific subject, in this scenario student loan and prepaid card services. This helps in creating a complete image of the consumers’ feedback on the service. Sentiment is a point of view of an individual towards a specific subject, that point of view can be either positive, negative or neutral.

Two algorithms (Score sentiment algorithm and Naïve Bayes algorithm) are used in order to produce sentiment from the consumers’ complaints dataset (Prepaid Cards and Student Loan) and to then classify the data of each service into “positive”, “Neutral” and “negative” sentiment. For this analysis 9,403 consumers’ complaints have been collected on student loan service and 1,271 consumers’ complaints on prepaid cards.

Both of datasets have been collected on the 28th April 2017, they both successfully classified using the score sentiment approach and the machine learning approach.

Item Type: Thesis (Undergraduate)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
H Social Sciences > HG Finance > Financial Services
Divisions: School of Computing > Bachelor of Science (Honours) in Business Information Systems
Depositing User: Caoimhe Ní Mhaicín
Date Deposited: 25 Oct 2017 15:24
Last Modified: 25 Oct 2017 15:24
URI: https://norma.ncirl.ie/id/eprint/2648

Actions (login required)

View Item View Item