Project: Credit Risk Analysis

Description:

Risk is delinquency and defaults of consumer credit. Using predictive modeling to investigate the delinquent and non-delinquent customers. Increase in accuracy of identifying high-risk loans could prevent huge losses.

Role Description:

Analyzing data and understanding the requirements, converting the data problem into business problem. Applied the logistic regression, random forest, SVM (kernel-based classification approach) models to identify risk assessment algorithms. Evaluating the model performance of the model.

Project: Sentiment Analysis

Description:

KBC wants to understand the sentiment of the customers towards the Bank. For this they are mainly focussing the comments made over the social media and interactions made over the Live Chat. The scope of the project is to analyze the sentiment of the customers and the areas where Bank needs to focus more.

Role Description:

Used NLTK in data pre-processing like stemming, lemmatization, and noise reduction conceptualized and implemented a sentiment analysis to find the areas where customers are mainly focusing based on subjective customer comments. Reporting statistical summaries and develop predictive models that helps Bank in determining the customer satisfaction and enhance customer experience.

Identify clusters of customers having similar characteristics for the purpose of targeting different segments with different types of promotional offers and benefits to have a significant increase in Return on Investment (ROI).

Role Description:

Understanding the requirements and objectives of the business problem. Cluster Analysis was done using K-Means Clustering. 4 clusters based on travel frequency finalized and multiple marketing techniques were shared.