It aims at creating a survey tool wherein user can distribute the forms created by admin and record the responses of the people. It also shows detials such as number of resonses and the users who have submitted the response.

Developed Recurrent Neural Networks to translate text from English to French using Keras, a TensorFlow extension. Performed comparative analysis on Long Short-Term Memory (LSTM), Bidirectional Recurrent Neural Networks implementations with text-processing algorithms such as TF-IDF and Continuous Bag of Words.

A project facilitating data management for the grounf work of toilet construction in urban slums across six cities in India

Developed Mastersheet - a customized Python tool for analysis and visualization of large data sets with report generation and data manipulation facilities. Developed Shelter Maps - a facility for geospatial visualization on Google Imagery to monitor the progress of different programs. Developing Shelter Dashboard - a graphical overview of real time data at multiple analytical levels.

Generated multiple choice cloze questions from input English text. Implemented skip-gram model and Continuous Bag of Words with Hierarchical Softmax and Negative Sampling in different permutations. Generated options for the MCQs using word vectorization making them semantically close. Compared the results using Mean Absolute Error and Root Mean Squared Error.