Publications

Dynamic Traffic Monitoring System.pdf

A review on recent driver safety systems and its emerging solutionsRoad safety and accident prevention are critical concerns in modern transportation. This paper presents a comprehensive survey of driver safety systems, focusing on the latest advancements in this ...

Credit Risk AssessmentThe global financial crisis of 2008 happened due to banks providing mortgages to people having low credit scores which then lead to high default rates. It has become all the more important for lenders to lend money only to people who have a better chance of returning it. This is why credit risk is one of the most important variables in the financial system of today. Credit score speaks to lenders about the likelihood of the consumer repaying the loans given in a timely manner. Organizations use services like CIBIL or they create their own methods of Credit Scoring using the data they have about the borrowers. The objective of this study is to predict the Probability of Default(PD) and to assign a credit score based on PD to minimize loss. We show the model selected meets the expectations with the highest accuracy 89.08% using the HistGradientBoostingClassifier algorithm, 94.54% precision in the prediction tasks and Xscore of 66.16%. The objective is to undertake a detailed analysis of various techniques implemented for credit scoring that allows us to determine the worthiness of the consumers for the lenders and other financial institutions.

Iot Based Home Security Management System (1).pdf

Survey on Virtual Recruitment SystemAfter the recent wave of covid, many companies are yet to shift their focus on recruiting candidates directly through in-campus recruitment. Many companies still do prefer the online way of conducting interviews on platforms like meet, zoom, etc. Even forms are shared to apply to a particular company and many more variable methods. Due to this rising uncertainty, the online recruitment system has started gaining more popularity. However, this procedure presents difficulties for recruiters in managing the flood of applications and maintaining contact with the applicants. Previously, such recruitment systems were inefficient and lacked many parameters like accuracy, not upgraded and not frequently managed. Therefore, there was a need to build a website which is capable of handling all the recruitment related activities. With the help of AI-ML techniques it has become possible to rank students according to the requirements of the companies. We discuss the features and research gaps for each method in applying Machine learning and other techniques for enhancing the recruitment process.

Driver Drowsiness Detection System.pdf

Adaptive smart signal to improve transportation.pdf

Report abuse