PROJECTS
Scania Sensor Fault Detection
The Air Pressure System (APS) plays a vital role in heavy-duty vehicles by utilizing compressed air to apply pressure to the brake pads, enabling the vehicle to slow down. Unlike hydraulic systems, APS offers advantages such as the ready availability and sustainable nature of natural air. The task at hand involves Binary Classification, where one class signifies failures attributed to a specific APS component, while the other class represents failures caused by other factors.View here
Insurance Premium Prediction
The purpose of this data is to look into the different features to observe their relationship, ML model based on several features of individual such as age, physical/family condition and location against their existing medical expense to be used for predicting future medical expenses of individuals that help medical insurance to make decision on charging the premium. View here
Flight Fare Prediction
This project aims to predict flight fares based on factors like timing, destination, and duration. It involves machine learning tasks such as data exploration, cleaning, feature engineering, model building, and testing. The goal is to provide people with a basic idea of flight fares, enabling them to save time and money when planning their trips.View here
Butterfly Classification using Deep learning(CNN)
The aim is to effectively classify the remaining 50 butterfly classes using deep learning techniques. By leveraging advanced methods, the project focuses on training a model capable of accurately categorizing butterfly species based on visual characteristics. The objective is to achieve reliable and precise classification results for the remaining 50 butterfly classes using deep learning approaches.View here
Phishing Website Classifier
Develop a classification method to predict if a website is phishing or not, using a set of predictors. It aims to protect users from scams and security threats by accurately identifying phishing sites based on factors like URL structure, content, and user behavior. View here
Fine-tuning ALBERT for Skype Chat Intent Prediction(NLP)
As part of an NLP project, I fine-tuned a custom ALBERT model using Skype chat support conversations to predict intents. This improved model enhances customer support by automating tasks and providing accurate intent predictions based on the content of the conversations. View here
Census Income Prediction Classification
The project involved predicting annual income levels using the Adult Income Census dataset. The main objective was to classify individuals as earning more or less than $50,000 per year. By employing data preprocessing and machine learning techniques, an accurate predictive model was developed. The project demonstrates proficiency in data analysis and machine learning for income prediction.View here