Leveraging Machine Learning for Predicting Customer Loan Performance

Aim: To build a robust classification model that predicts the binary loan performance label (good or bad) for customers, helping financial institutions reduce risks and make data-driven decisions.

Profiling Users Based on Behaviour and Engagement Metrics Using Clustering Techniques

To segment users into distinct clusters based on key engagement and financial metrics, enabling tailored strategies for customer management and targeted solutions.

Object detection of aerial targets (Drones and Birds)

The purpose of this study is to compare different image classifiers, such as Convolutional Neural Network (CNN), Support Vector Machine (SVM), k-Nearest Neighbour (kNN), and Decision Tree, for identifying aerial targets. The study will evaluate the classifiers' ability to detect aerial targets.

Supplier quality and Performance report

Developed a comprehensive Power Bi report to address supplier quality challenges for Enterprise Manufacturers Ltd • Consolidated data from multiple plants to provide insights into defects quantities, downtime, and supplier performance • Provided actionable insights to management enabling data-driven

Design and development of an analysis tool for automatic detection of brain hemorrhage

The project involves designing and developing an analysis tool using Python and MATLAB for automatic detection of brain hemorrhage. The tool will utilize machine learning algorithms to analyze medical imaging data, such as MRI or CT scans, to identify regions of the brain affected by hemorrhage.

Predicting rainfall

Data was gathered and prepared for machine learning analysis of rainfall data. The data was collected in “.xls” format and was gathered from several weather stations.

Optimizing Taxi Fleet Allocation

By analysing past data, our machine learning algorithm was able to accurately forecast the demand for taxis on December 6, 2016. We have effectively predicted trip counts for several places by utilising features such as day of the week, time of the day, day of the year, and month. An R-squared score

Object detection of aerial targets (Drones and Birds)

The purpose of this study is to compare different image classifiers, such as Convolutional Neural Network (CNN), Support Vector Machine (SVM), k-Nearest Neighbour (kNN), and Decision Tree, for identifying aerial targets. The study will evaluate the classifiers' ability to detect aerial targets.