Projects

Price Quoting System

Deloitte has a internal tool for quoting prices to clients. This tool is handled by the Project Manager, where they provide details like, client information, project information, number or resources involved in the project, duration of the project and etc. I am taking multiple such factors and providing few insights to the Project Manager, which enables the manager to provide best quote for the project.

Employee Attrition Prediction

Providing analysis and prediction on Attrition of employees at Deloitte. I am working on more then 40 Employee Factors which can affect Employee Attrition. I am using different Machine Learning techniques to predict attritions and computing analysis of those models using Model Interpretability techniques. (Cannot share further details due to Confidentiality agreement.)

Chatbot – Machine Reading comprehension

At Deloitte we have a internal chatbot for employees, which has multiple feature out of which I have implemented Machine Reading Comprehension (MRC). MRC enables the chatbot to read documents and answer the questions based on those documents. A hint as I am not allowed to talk – I used Transformers.

Human Action Recognition

Creating a solution for recognizing and detecting Human Actions from CCTV Cameras in real-time. This project is par of my M.Tech. thesis. I am still working on the solution, the link will be posted once I am ready with it.

Sukanya Rakshak: Wireless Sensor Network based Security System

Sukanya Rakshak is a technology driven security system for hostels to provide higher security to hostel inmates. The system provides first line of defense which can identify any type of threat. It uses different type of Sensors to identify the threat. It is a WSN-IoT based system which can be integrated with per-existing security systems.

Language Identifier for Hinglish Language

Language Identification for Code-Mixed Language. It classifies Hinglish and English words using RNN Created a Deep Learning Model on Keras for identifying language. The pre-trained model works for Hindi and English Code-mixed language. But the model can be modified for other languages as well.

Abstractive Text Summarization

The objective of this project is to build a seq2seq model that can create relevant summaries for reviews written about fine foods sold on Amazon. This dataset contains above 500,000 reviews, and is hosted on Kaggle. It’s too large to host here, it’s over 300MB.

ChatBot with Human like Emotions

A generative ChatBot based on Sequential Neural Network and Deep Learning which can be trained on any desired dataset for specific purposes. Instead of ordinary ChatBots which are based on hard-coded responses, it can understand context and respond accordingly.