About the customer – Customer is one of the top 3 largest credit card & payments companies in the world.

Customer Challenge

• Teradata was the only platform that was catering to the enterprise reporting, This added to significant annual costs for the customer
• Data was in silos and each BU had their own process to exchange data with other Bus, This added significant delays in developing products that required enterprise level data.
• Lack of self-service of the enterprise data leading to higher time to market, costs & redundancies.

NucleusTeq Solution Approach

• Centralized, Standardized & Certified Data for Consistency
• Democratize Data through self-service API’s while eliminating redundancy
• Ensuring compliance to Enterprise data security guidelines
• Maximize application/data Reusability & Globalization

Outcomes Delivered

• Comprehensive Metadata Management, Security Data Quality & Governance
• Single source of truth
• Data integrity & quality with financial B&C
• Eliminated redundancy by 47%
• Reduced Attribute SORs by 72%
• 8 Petabyte data migrated
• 18 Months total delivery time
• 35+ M in operations savings per year

About the customer – Customer is one of the top 10 largest banks in the world.

Customer Challenge

• Bank lost billions of dollars in penalties and trading losses
• Challenge posed was to detect the intent to commit fraud proactively by monitoring communications in order to avoid penalties and damage to reputation
• Large team of ~30 people were dedicated to ensure compliance using keyword search but face too many false positives

NucleusTeq Solution Approach

• Delivered a semi-supervised learning-based NLP solution to aid in detecting fraudulent activity
• Coordinated multiple rounds of reviews to improve solution accuracy across 10 categories
• Attained expected precision in SLA within 2 months
• Implemented solution on Stanford NLP, Spark and Tachyon

Outcomes Delivered

• Delivered expected accuracy to detect email messages with scores for fraud categories
• Elimination of a large part of the data via identification of non-relevant messages using learning algorithms
• Defined templates unique to each type of fraud based on presence of different entities (persons, places etc)
• Created an approach to profile all the users in the data by means of graph analytics algorithms