I have listed all my projects here on this webpage for your reference. Feel free to go through them. I have linked my GitHub individually and the homepage in the contact section. Some of my projects only required the submission of a final article. Corresponding projects' GitHub will only contain the final article. Some of the group projects that I was part of, will have code in a shared repository and mine will contain the report/ article. If you have any questions about any projects, feel free to contact me.
(All the titles in this section also function as a hyperlink redirecting to my projects section, with the "projects" title taking you to my GitHub homepage)
Website under construction. Certain Links may not work. Need to add more projects! Apologies for the inconvenience caused!
Successfully implemented ML model Long Short-Term Memory.
We deployed the model successfully by first creating a flask model file; set-up an ECS cluster, configured required tasks, and achieved automated deployment using AWS Code-Pipeline.
Effectively tested the working with Docker locally and checked the connections with Postman.
Conducted a comprehensive analysis of appliance energy usage within smart homes. Successfully implemented and optimized predictive models to forecast future energy demand, schedule maintenance, and drive advancements in IoT device development.
Utilized advanced analytics to extract actionable insights, identifying areas for improvement to minimize energy wastage and reduce expenses.
Successfully extracted Legal Entitites from text.
We used the “Bidirectional Encoder Representations from Transformers” (BERT), DistilBERT, and XLNET which were effectively implemented.
Successfully designed three models 1. CNN, 2. LSTM, 3. XGBoost to predict Particulate Matter (PM2.5) for stations that (at the time of designing the models) lacked the ability to predict PM levels.
Used data from AQMD stations in California and used 168 hours’ worth of data to predict the next 6 hours.
(Read the report here)
Integrated Automated Machine Learning in Supply Chain to predict backorders using LightAutoML to integrate with the existing Supply Chain. Successfully achieved a score of 97%.
Reduced the active monitoring time of engineers for supply chain management.
Demonstrated inadequacies in industry supply chain dataset.
Detection of signs of Aging using Machine Learning
Successfully implemented a machine learning model with Tensorflow and OpenCV.
Collaborated with a multidisciplinary team to integrate the aging signs detection system into a broader application.
Demonstrated problem-solving skills and creativity in addressing challenges related to image analysis and machine learning.
COVID cases forecasting
• Analyzed COVID case trends and developed a trend forecasting model using Long Short-Term Memory (LSTM), a variety of recurrent neural network (RNN).
Attendance Tracking system using GPS
• Developed a mobile application for attendance tracking using geolocation
• Led a team of developers in full frontend development and partial backend development
• Used a combination of Adobe XD and Android Studio to develop the GPS-based attendance system