The objective of this project is to apply classification learning models on retired or terminated customer’s dataset with 60 features for 500,000 customers and therefore to obtain a predictive model for identifying future customers who roll over assets to a competitor. This helps to increase the Rollover retention Rate, revenue, crossover customers, and improve client loyalty.

The objective of this project is to categorize whether a webshop session will end in a purchase or not. Predicting customer behavior increases customer satisfaction and sales, by facilitating an increase in customer experience through personalization, recommendations, and special offers.

Feedback analysis to classify each feedback (as excellent, good, or bad) to understand the customer opinion about the hotel which is useful for hoteliers to improve their performance based on feedback. This is a multilingual problem extracted other language reviews from the data and translated other languages other than English to English and implemented text preprocessing steps like cleaning data, tokenization, and convert text into numbers using TF-IDF Vectorizer. Built multiple models and analyzed the results.

AML Alerts are raised based on transactions done by the customers. The objective of this project is to classify the alerts into high medium and low categories. High alerts are investigated further and raise STR if suspicious whereas low alerts are closed in this way we optimize the alerts for investigation. And in the end our script generates RPA for high-risk customers which are used in investigation process.

ODD alerts are raised when customers do profile changes. ODD Alerts are of two types Periodic and Event Trigger. Alerts are prioritized into high medium and low based on the score. Low alerts are bulk closed and high alerts are investigated further.

Data extraction requires to analyze the trends / pattern of crypto transaction remarks by CIMB Bank customers. After applying various techniques filtered out customers and given for investigation team.

Westfield Insurance is the primary subsidiary of Westfield Group is a multi-line provider of business Property and Liability insurance Personal lines insurance.

Feedback analysis to classify each feedback (as excellent good or bad) to understand the customer opinion about the hotel which is useful for hoteliers to improve their performance based on feedback. This is a multilingual problem extracted other language reviews from the data and translated other languages other than English to English and implemented text preprocessing steps like cleaning data tokenization and convert text into numbers using TF-IDF Vectorizer. Built multiple models and analysed the results.