'Recommendations' beyond boundaries: Prof. Bidyut Patra heading for ECIR 2019

'Recommendations' beyond boundaries: Prof. Bidyut Patra heading for ECIR 2019

Animesh Pradhan Harsh Mohan | Mar 04, 2019

  • 0

In yet another achievement for the Department of Computer Science and Technology, Prof. Bidyut Kumar Patra would be presenting a tutorial talk on “Concept to Code: Neural Networks for Sequence Learning” at the “European Conference on Information Retrieval” (ECIR) to be held in Germany later in April this year. For the same Prof. Patra has teamed up with Flipkart who would sponsor his trip to Germany. Team MM caught up with him to get a glimpse of the timeline which led him to one of the most prestigious Data Science Conferences.


Recommendation System (or Recommender system) refers to the science of understanding the preferences and making predictions based on a user’s present access to the data and webpage. It facilitates to provide a recommendation that would interest the customers, for example, if you purchase a product from an e-commerce site, the recommendation system makes valid guesses the preference, rating and interests of the customer.

This field of Deep Learning and Sequential Learning is growing immense popularity in the world of Big data and Computer Science since it finds its use in various sectors including E-commerce, movies, music, search engines, social media.


Prof. Bidyut Kumar Patra’s credentials in this particular research area date back to 2015, where Prof. Patra presented his research paper along with Apoorva Pathak at CODS Conference held at Bangalore (2015). Prof. Patra worked on ‘Recommender Systems’ along with Mr. Pathak (Graduated from NITR in 2013) who pitched in the idea for the research while Prof. Patra was carrying out his post-doctoral researches in Finland under Marie-Curie Fellowship.

Out of 64 research papers sent at the prestigious conference, only 7 were accepted including the one authored by Prof. Patra and Mr. Pathak. CODS-COMAD is one of the few prestigious conferences that witness the congregation of research enthusiasts from institutions along with various coveted industries like Flipkart, Adobe, Microsoft, American Expressway, Amazon. The paper was hugely appreciated by the participating members in general and the representatives from Flipkart in particular.


In the year 2018, T. Dharas, a Dual Degree student from the Computer Science branch who was working with Prof. Korra Sathya Babu collaborated with Prof. Bidyut Kr. Patra focused on ‘Recommender System’ focused on ‘Multi-criteria Recommendation’ using Deep Learning. Working on this idea, Prof. Patra submitted a short research paper in RecSys, a top-notch conference exclusively on Recommender Systems. The paper was accepted by RecSys which was held in Vancouver, Canada (October 2018). Since T. Dharas had graduated during that time, the paper was presented by the Ph.D. a scholar working under Prof. Patra, Miss Rama Syamala Sreepada at RecSys-2018. Being impressed with the work, Flipkart provided the sponsorship worth 1.3 Lakh for presenting the paper in the prestigious conference.


In the year 2019, Prof. Bidyut Kumar Patra has teamed up with Flipkart to present a tutorial titled, “Concept to Code: Neural Networks for Sequence Learning” at The European Conference on Information Retrieval (ECIR) to be held from 14-18 April 2019 in Germany. Prof. Patra has been working meticulously on this arena of Recommender Systems and Sequential Learning and with immense expertise will be presenting a tutorial talk at ERIC, one of the most sought after conferences of the world. For the same, Prof. Patra has been provided with sponsorship of Rs. 2 Lakhs INR from Flipkart. Prof. Patra has teamed up with Omprakash Sonie, Muthusamy Chelliah and Surender Kumar, all of them being from Flipkart.

The tutorial as stated in the Abstract focuses on conceptual understanding of Embedding methods and Recurrent Neural Networks which are currently applied in Information Retrieval. The Recurrent Neural Networks are efficient in modelling sequential data (ex: clicks, add to cart, purchase data) which is generated by users in a session and across sessions.

Talking of the upcoming ECIR conference Prof. Bidyut Kumar Patra said,

Being accepted to deliver a talk at ECIR 2019, one of the top conferences in the world for Data Science, is a great thing for us and very few Indian Institutes could make it there. Initially, we submitted the proposal to deliver a 3-hour talk but this being an important and broad topic we are asked to deliver a full day 6 hours talk.

He further added

Opportunities are available for students out there. Dual degree students are better to collaborate with to produce good research outcomes as they have an extra year to work with us. We do not have dual degree students here now. Nevertheless, students should take the initiative and should start early. If the work produced is good it would be accepted for top conferences and funds would eventually turn up. For this conference too, we were clueless of where would we get funds from yet we submitted the proposal. It was accepted and funds did turn up. So, all that matter is quality.

Team MM wishes Prof. Bidyut Kumar Patra all the luck for the conference.

DISCLAIMER: The content, opinions or views expressed on the Monday Morning's website and its social media platforms, including, but not limited to Facebook, Instagram and Twitter pages, are strictly the property of Monday Morning and represent the extensive research and work of the working team of respective academic year of Monday Morning and not those of the institute. The reports and statements published are consolidated from the collected background research and interviews. The institute's official statements can be found in the press releases published by the institute or via an RTI application.

No article or any statements by Monday Morning is to be reproduced, presented or distributed in part or whole without prior permission of the Executive Body of Monday Morning for any purposes, including, but not limited to print and electronic form.


    Leave a comment

    Login to comment.
    Ask a Question Forum