Solar Energy Production Forecast:
o Building Regression and Time Series methods.
o Forecastenergyproductionbybuilding ahybridmodel.[RNNandGBM]
o Forecast the amount of insurance claim received by insurance company.

• Data insights and Data preparation
• Classifying PO (use full form of PO) whether it will be in market or not.
• DelayPrediction-PredictingweatheraPOwillbelateorintime.Ifitisdelayed, predicting how many weeks it would be delayed.

• Workbench as a desktop widget: An interactive and essential tool for data analysts. This is a desktop version which includes regression (10 algorithms), classification (7 algorithms), clustering, time series (7 algorithms), dimensions reduction, association, few solutions like anomaly detection.
• Workbench as a web based: The desktop UI was migrated to web based along with other features like visualisation, big data, NLP, data base connection.
• Forecasting number of calls : forecasting the number of calls thats are made in a call centre for credit TL to support Mobiloans.
• Modularizing Forecasting tool: converting the R forecasting code into python and using oop concepts to modularise the forecasting tool.
• Forecasting the number of accidents: Forecasting the number of accidents happyening with the data by Vantage agora.
• Forecasting the number of claims: forecasting the number of insurance claims made for every month.
• Forecasting the amount of claims: forecasting the amount of claims per month using time series models/ regression models.
• Classification of customers : forecasting if a customer claims in the next year or not based on given data.
• Noise analysis: analysis of the noise present in the time series data.

Skills: Python, Machine Learning, Computer Vision, C++