Time series prediction on meteorological data from Beijing Airport using Deep learning on Python and Tensorflow June - Sept 2017

• Designed a multivariate time series prediction model based on ARIMA, ARMA for short trends in the data.

• Implemented the times series prediction on multiple features using LSTM neural networks and Deep learning.

Natural Language Processing to identify the source of food-borne diseases and suggesting the next course of action June - Sept 2017

• Designed a Supervised Learning model that incorporated Natural Language Processing for classification using Python.

• The model can be used to detect the food sources responsible for causing certain symptoms, and suggest the next course of actions that can then be taken to alleviate the symptoms

European Language Detection using Natural Language Processing May - July 2017

• Designed a Machine learning model for classifying 21 European Union languages with an accuracy of 97%.

• Logistic regression and Xgboost classifier were used to train the 5 GB dataset using Python.

Stock Market Prediction using Natural Language Processing Feb - May 2017

• Designed a Machine learning model that makes accurate predictions in stock market trends with the help of NLP, scraping data from Yahoo Finance and Reddit.

• Implemented a Random Forest based model and a Neural Network architecture based on the Backpropagation algorithm written in Python using Scikit Learn and Pandas.

Big data? A View from Information Theory Aug - Dec 2017

• Conducted State of the art research on how principles of information theory can solve the problem that relate to Big data technologies. Information coupling problem for reducing higher dimension data to lower dimension data was achieved.