viZio – Content Scoring Engine


Recommender systems have become increasingly popular in recent years and are utilized in various areas, including movies, music, news, books, research articles, social tags, and products in general. Most of the existing e-commerce recommender systems have been designed to recommend the right products to users based on the history of previous users' transaction records. Thus, the recommendation system for marketing analytics is a subclass of information filtering system that seeks the similarities between users and items with different combinations.

This work focuses on content-based scoring, which aims to quantify the impact of content quality of promotional emails sent across to healthcare practitioners upon the immediate engagement.

An enhanced version of OCR is being used to extract the content from the digital media (i.e., scanned email images). The process has strived to detect the context and the localization of text information from the email and project it into a higher dimensional space with diffused information. At the same time, the pictorial features have been extracted via Morphological transformations and the HLS threshold algorithm. A semi-supervised K-Nearest Neighbor technique was implemented to extrapolate the distribution of the emails with known impact onto the ones with unknown engagement metrics due to the scarcity of performance data mapping being a significant challenge; this type of scoring system can help the business to provide more appropriate documents to suit a user's personal information need. The simulation results showed that our model could score relevant documents sent to users with higher precision than other non-hybrid information filtering models.

This approach can be extrapolated to various industries trying to optimize their marketing strategy for the digital end users. It will enable the process of getting instantaneous unbiased rating and will act as a guide to the content development team to create and publish relevant, impactful and decision-driven content. It can be used to personalize and aid precision digital marketing on the content while keeping the turn around time to minimum.


Sauradeep Debnath has joined Novartis’s Hyderabad office in Feb 2020 as an Analyst. He has 4 years of experience in NLP, Computer Vision, Image Processing, and other fields of Data Science/analytics. Sauradeep played a crucial role in RAinBOW Clustering project for Japan Hematology–a project which won on the Highest Award for Novartis Global Oncology (BOLD4CURE 2021). He worked in Phase 1 of Deep Learning Implementation of the MATCH Project Sauradeep has worked extensively on the Veeva Survey Analysis for Inclisiran – and worked with Topic Modelling & Keyphrase Extraction there, apart from building the Text Cleaning Pipeline. Prior to joining Novartis, he worked with Oracle. Sauradeep is currently pursuing M.Tech. in Data Science from IIT Hyderabad. He holds a Bachelors degree (in ECE) from NIT Jaipur.

Open Data Science




Open Data Science
One Broadway
Cambridge, MA 02142

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google