A Framework for Illegal Online Loan Risk Using Word Cloud and Big Data Analytics

Abstract

Information technology has provided many conveniences in community activities. Big data on the Internet benefits many things, especially risk information that occurs without being careful in conducting transactions and credit activities online. This study analyzes online credit risk using ten variables related to online credit risk. The data set used in this paper is sourced from the Internet by using keywords that have been determined using the Uniform Resource Locator (URL) from different websites. The research method used in this study is an experimental method by classifying word variables related to online credit risk through data collection, initial data processing with a word cloud generator, and data analysis with python programming, then evaluation and validation of results. Variables analyzed such high loan interest, small loan ceiling, personal data in the App, old approval, the collector is coming, administrative costs, not yet registered with the OJK, unofficial loan institutions, consumer data protection, and cost transparency. Data collection techniques by means of questionnaires were carried out to online loan money borrowers to explore more in-depth information. The results of the analysis that has been carried out with the python programming language using the pandas, matplotlib, and seaborn libraries produce the Small Loan Ceiling variable, which greatly influences the consumer data protection variable with a value of 0.99. An in-depth analysis of these variables found that credit with a ceiling is ineffective.