Objective:

Based on the historical data such as membership, driving data, personal data etc., are used in the analysis to predict how people are likely to claim the amount.

Responsibilities:

• Exploratory Data Analysis on Historical Data
• Identifying the important features and filling missing data with imputation techniques
• Segregate the data into train and test data, and build some classification models.
• Running a clustering model on most important columns and obtained good values of sensitivity and specificity
• Building a regression model with Logistic Regression
• Building a decision tree model using historical data

Tools & Techniques: R, Regression, Clustering, Python, Supervised learning

Objective:

To predict surrounding temperature of brake disc

Responsibilities:

• Surrounding temperature of brake disc is dependent on temperature of brake disc and vehicle speed
• Using Multiple linear regression, surrounding prediction model was built which involves initial exploratory data analysis
• It also involved finding if input variable has any co-linearity problem using scatter plot and correlation matrix
• After verifying, prediction model was built
• Coefficient of determination (R2) was found to be 86 %
• Now this model is been used to calculate surrounding temperature and hence for brake cool down analysis.

Tools & Techniques: R and Excel. : Linear Regression, scatter diagram, Normality check, added variable plot, VIF, Residual plots, Index plot.