Working as a Senior consultant for global retail customer.+ Using statistical software R to conduct data preparation (Cleaning and Tidying), outlier detection and missing value treatment. + Defining business hypothesis and Validating them by analyzing historical data.+ Creating Logistic regression on train data set, defined the cut-off probability and validating the model on test data.+ Building Decision tree model on train data set based on GINI and Entropy, splitting… Show more Working as a Senior consultant for global retail customer.+ Using statistical software R to conduct data preparation (Cleaning and Tidying), outlier detection and missing value treatment. + Defining business hypothesis and Validating them by analyzing historical data.+ Creating Logistic regression on train data set, defined the cut-off probability and validating the model on test data.+ Building Decision tree model on train data set based on GINI and Entropy, splitting methods and Prune the tree using control parameters so that tree will not over-fit.+ Creating confusion matrix and deriving the Overall accuracy and TPR and TNR.+ Creating Random Forest model. Using the Grid Search mechanism identifying the mtry, node size and sample size to find out the optimal model to minimize the out-of-bag error.+ Using K-means clustering algorithm, an unsupervised machine technique, to cluster the customers.+ Profiling on each cluster to identify the cluster behavioral pattern. Show less

+ Merging various source of data to collect meaningful information and done the missing value imputation and outlier treatment.+ Analyzing the historical data with Descriptive Statistics, doing Uni-variate and Bi-variate analysis on the business-critical variables. + Presenting the graphical view of Sales over all summary, Sales Compensation, Product views (country, region, city wise) etc to business.+ Using Linear regression to predict the future sales and verified the assumptions… Show more + Merging various source of data to collect meaningful information and done the missing value imputation and outlier treatment.+ Analyzing the historical data with Descriptive Statistics, doing Uni-variate and Bi-variate analysis on the business-critical variables. + Presenting the graphical view of Sales over all summary, Sales Compensation, Product views (country, region, city wise) etc to business.+ Using Linear regression to predict the future sales and verified the assumptions of Linear Regression and identifing the business-critical significant variables. Show less