uditmishra.com
Projects
Alopecia India
Alopecia India is an independent organization dedicated to supporting individuals affected by alopecia and cancer. Our mission is to boost their confidence and help them face society without feeling self-conscious about their appearance. Through the donation of wigs, we empower people to embrace their beauty and regain a sense of normalcy. To date, we’ve raised $3,200 and proudly donated 18 wigs, with many more campaigns and contributions on the horizon. Our goal is to continue making a positive impact, one wig at a time.
Future City Plantation Drive
Future City Plantation Drive is an environmental project aimed at revitalizing barren land in Phagi, Jaipur, by planting thousands of trees and promoting sustainable land use. Through collaboration with local farmers, volunteers, and environmentalists, the initiative seeks to improve soil quality, enhance biodiversity, and create a greener future for the community. This project is not only about restoring nature but also educating and involving the local population in sustainable practices, making it a long-term solution for the region’s ecological balance.
Research on Artificial Intelligence: Emotion Detection in Virtual Assistants and Chatbots
In my AI research titled 'Emotion Detection in Virtual Assistants and Chatbots,' published in the International Research Journal of Modernization in Engineering, Technology, and Science, I aimed to enhance emotional intelligence in virtual assistants and chatbots. This involved developing systems capable of detecting and responding to user emotions, thereby creating more empathetic and engaging interactions. The research utilized NLP techniques, sentiment analysis, to recognize emotional cues from text and context, with a focus on improving user experience and emotional engagement.
Research on Sports Algorithms: Using Machine Learning to Forecast Football Shot Outcomes
In this research, I used machine learning to analyze football data and predict whether a shot would result in a goal. Starting with data from a Kaggle article by Usama Waheed, I enhanced existing models by introducing advanced algorithms like Decision Trees, LightGBM, and CatBoost. Key features such as shot coordinates, angles, player skill, and shot type were considered to improve predictions. A total of nine algorithms were tested, including XGBoost, Random Forests, and Artificial Neural Networks. Through techniques like nested cross-validation and hyperparameter tuning, the models were optimized for imbalanced datasets. Decision Trees emerged as the top performer with the highest accuracy, offering significant improvements to expected goals (xG) models. This project bridges data science and sports analytics, unlocking new insights to enhance football strategies and performance.