Objective: Identify fraudulent transactions for Internet banking channel and reducing false positives.

Technique used: Basic decision tree model (C5.0 classification model)

Objective: Intelligently predict cash withdrawal and prescribe optimal cash replenishments and releases for reducing shadow cost & cash-out events

Technique used: ElasticNet Regression (It is a linear regression model trained with regularization of the coefficients)

Objective:Increase product penetration and customer revenue through tailored product offers (Upsell and Cross-sell products ) by developing Personalized Product Offering tool

Technique used: Propensity Modelling (It is a statistical approach which attempts to estimate the likelihood of products having certain types of behavior (e.g. the purchase of a product))

Objective: Reduce cost-to-serve by moving customers from using high-cost channels to low-cost channels by migrating customers to the most preferred channels based on customer behavior models and optimizing the marketing spend by targeting customers with a high propensity to convert.

Technique used: Propensity Modelling (It is a statistical approach which attempts to estimate the likelihood of products having certain types of behavior (e.g. the purchase of a product))

Objective:Predict customers who are likely to churn and react to potential churn events for reducing customer attrition

Technique used: Basic decision tree model (C5.0 classification model) with RFM