Pavithran Rajan

An aspiring tech developer with a background in engineering interested in investigating the wide range of opportunities in computer science and technology. A well-organised, detail-oriented, dedicated, and hardworking candidate seeking entry-level opportunities.

.

Power Scheduling in Smart Homes Using Deep Learning Algorithms - Master's Thesis

• Implemented deep learning algorithms as a viable solution to the smart home power scheduling problem. Used LSTM to forecast electricity prices produce an optimised schedule with metaheuristic algorithms such as the genetic algorithm and particle swarm algorithm. The model increased the electricity cost savings by 12%.

Song Popularity and Genre Prediction on Spotify using Machine Learning - Machine Learning Project

• Worked with a team on a module project to build machine learning models to predict song popularity using regression and predict song genre using classification. The model produced a testing accuracy of 97.5% for popularity prediction and 93% for genre prediction.

Automatic Target Recognition in LiDAR Point Clouds - Final Year Project.

• Designed a deep neural network model for automatic target recognition in autonomous vehicles with LiDAR point clouds using convolution neural networks and performed 2D bird-eye projection and a feature extraction block for processing lidar point clouds.