Risk management credit scoring prediction using sentiment analysis

Chew, Chun Phang (2024) Risk management credit scoring prediction using sentiment analysis. Final Year Project, UTAR.

Preview

PDF Download (2861Kb) | Preview

Abstract

In changing dynamic financial risk management, this project seeks to advance credit scoring predictions through the sentiment analysis into the data science framework, with a specific focus on Natural Language Processing (NLP) and classification algorithms. Traditional credit scoring models, they rely on the traditional historical financial data, normally cannot capture the real-time dynamics and external factors that can impact the borrower’s creditworthiness. The objective is to use the power of sentiment analysis, originate from the diverse textual sources such as social media and financial reports to increase the accuracy and flexibility of credit risk assessments. The project adopts a development-based approach with field of data science, leveraging NLP techniques and classification algorithms to seamlessly integrate sentiment-derived features with conventional credit scoring attributes. The methodology emphasizes the fusion of sentiment-derived insights with established credit data, ensuring a comprehensive understanding of credit risk factors. The methodology involves five steps to process data; those are data collection, text preparation, sentiment detection, sentiment classification, and presentation of output. This is to ensure the accuracy of the borrower’s creditworthiness will increase compared to the traditional credit scoring models. This proposal will discuss a few relevant topics such as literature reviews, research analysis, and conclusion of the project work.

Actions (login required)