body, #content, .entry-content, .post-content, .page-content, .post-excerpt, .entry-summary, .entry-excerpt, .widget-area, .widget, .sidebar, #sidebar, footer, .footer, #footer, .site-footer { font-family: "Poppins" !important; } #site-title, .site-title, #site-title a, .site-title a, .entry-title, .entry-title a, h1, h2, h3, h4, h5, h6, .widget-title { font-family: "Poppins" !important; } button, .button, input, select, textarea, .wp-block-button, .wp-block-button__link { font-family: "Poppins" !important; } #site-title, .site-title, #site-title a, .site-title a, #logo, #logo a, .logo, .logo a { font-family: "Poppins" !important; } #site-description, .site-description { font-family: "Poppins" !important; } .menu, .page_item a, .menu-item a { font-family: "Poppins" !important; } .entry-content, .entry-content p, .post-content, .page-content, .post-excerpt, .entry-summary, .entry-excerpt, .excerpt, .excerpt p, .type-post p, .type-page p { font-family: "Poppins" !important; } .entry-title, .entry-title a, .post-title, .post-title a, .page-title, .entry-content h1, #content h1, .type-post h1, .type-page h1 { font-family: "Poppins" !important; } .entry-content h2, .post-content h2, .page-content h2, #content h2, .type-post h2, .type-page h2 { font-family: "Poppins" !important; } .entry-content h3, .post-content h3, .page-content h3, #content h3, .type-post h3, .type-page h3 { font-family: "Poppins" !important; } .entry-content h4, .post-content h4, .page-content h4, #content h4, .type-post h4, .type-page h4 { font-family: "Poppins" !important; } .entry-content h5, .post-content h5, .page-content h5, #content h5, .type-post h5, .type-page h5 { font-family: "Poppins" !important; } .entry-content h6, .post-content h6, .page-content h6, #content h6, .type-post h6, .type-page h6 { font-family: "Poppins" !important; } .widget-title, .widget-area h1, .widget-area h2, .widget-area h3, .widget-area h4, .widgets-area h5, .widget-area h6, #secondary h1, #secondary h2, #secondary h3, #secondary h4, #secondary h5, #secondary h6 { font-family: "Poppins" !important; } .widget-area, .widget, .sidebar, #sidebar, #secondary { font-family: "Poppins" !important; } footer h1, footer h2, footer h3, footer h4, footer h5, footer h6, .footer h1, .footer h2, .footer h3, .footer h4, .footer h5, .footer h6 #footer h1, #footer h2, #footer h3, #footer h4, #footer h5, #footer h6 { font-family: "Poppins" !important; } footer, #footer, .footer, .site-footer { font-family: "Poppins" !important; }


About the customer – Customer is one of the top 10 largest bank 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

About the customer – Customer is one of the top 4 Credit card company in the world

Customer Challenge

  • Create a recommendation model for cross-sell/up-sell recommendations for 80+ Million users
  • Performance requirements of 300-350 TPS
  • Create a Multi-Touch Attribution Model (MTA) for channel optimization
  • Integrate Data from 10+ different sources for building both models

NucleusTeq Solution Approach

  • Developed  a product recommendation engine that personalizes recommendations for 80 million customers
  • Created a high availability big-data platform where the recommendation engine processes the data
  • Created high availability API infrastructure to address the high TPS requirement
  • Created a graphical model for MTA

Outcomes Delivered

  • Savings of millions of dollars in promotional expenses
  • Increased Net New Conversions from  2% to 7%
  • Increase Upgrade Conversions from 8% to 14%
  • Model for channel budget allocation led to increase in ROI


Let’s Connect

    Subscribe to our Newsletter

    Get regular insights and news about how Data and Artificial Intelligence merge to provide an integrated experience for your customers, and how the world of technology is evolving!

    © 2020 NucleusTeq Inc. All Rights Reserved.