I have created a web app for prediction of Diabetes in patients using Machine Learning. You can enter few parameters of patient like the number of pregnancies the patient has had (if female), BMI, insulin level, age, and other things. The model will predict whether or not a patient has diabetes, based on entered diagnostic parameters.

The Machine Learning model has been created using Random Forest algorithm and trained with dataset originally from NIDDK (training data has details of total 768 patients). NIDDK (National Institute of Diabetes and Digestive and Kidney Diseases) research creates knowledge about and treatments for the most chronic, costly, and consequential diseases.

Retail PGP (Unsupervised Learning)

• To analyze transactional data for an online UK-based retail company and create Customer Segmentation.
• Tools used: Python, Pandas, Numpy, K-means clustering, Scikit-learn, Matplotlib, Seaborn, Exploratory Data Analysis (EDA), Tableau.

Real Estate PGP (Regression)

• To predict potential demand of loan for each of the region in the USA for a mortgage bank.
• Tools used: Python, Pandas, Numpy, Linear Regression, Scikit-learn, Matplotlib, Seaborn, Plotly, FactorAnalyzer, Exploratory Data Analysis (EDA), Data Preprocessing, Correlation Analysis, Tableau.

Text Generation (Deep Learning/ NLP)

• To predict text for news headline using Deep learning and Natural Language Processing.
• Tools used: Python, Pandas, Numpy, NLP, Deep Learning, Tensorflow Keras, Recurrent Neural Network (RNN), Long Short Term Memory networks (LSTM), Language Modeling.

To analyze patient data from NIDDK and predict whether patients have diabetes or not.

To analyze product reviews by customers for a leading mobile brand on a leading e-commerce site.

To perform hypothesis testing and statistical analysis in R and create Linear Regression model using R programming language to predict demand for a leading Retail company in US.