CS Colloquium Series @ UCY

Department of Computer Science - University of Cyprus

The Department of Computer Science at the University of Cyprus holds research colloquiums and social hours approximately once weekly. All university students, faculty, and staff are invited to attend. Notifications about new and upcoming events are automatically disseminated to a variety of institutional lists.
If you don't receive these notifications, but want to get informed about upcoming colloquium announcements, you can do the following:
mail List rss RSS Directions Directions

Colloquium Coordinator: Demetris Zeinalipour

Colloquium: Toward Interpretable Recommender Systems, Prof. Michalis Vlachos (University of Lausanne, Switzerland), Thursday, January 16, 2020, 11:00-12:00 EET.


The Department of Computer Science at the University of Cyprus cordially invites you to the Colloquium entitled:

Toward Interpretable Recommender Systems

 

Speaker: Prof. Michalis Vlachos
Affiliation: University of Lausanne, Switzerland
Category: Colloquium
Location: Room 148, Faculty of Pure and Applied Sciences (FST-01), 1 University Avenue, 2109 Nicosia, Cyprus (directions)
Date: Thursday, January 16, 2020
Time: 11:00-12:00 EET
Host: Dr. Demetris Zeinalipour (dzeina-AT-cs.ucy.ac.cy)
URL: https://www.cs.ucy.ac.cy/colloquium/index.php?speaker=cs.ucy.2020.vlachos

Abstract:
With the advent of Big Data, we can learn more about our clients, and we can create 360-degree views which help better understand their needs and their pain points. In this talk, I will describe the building of a business-to-business (B2B) recommendation engine that predicts the interest of clients in products using both proprietary and public data sources. For such a system, interpretability of the predicted action is particularly important, because generated recommendations will be furnished to the salesperson responsible for the client account. We address interpretability by transforming the recommendation problem into an instance of an overlapping co-clustering problem. With the use of GPUs, we can also significantly reduce the training of the recommendation engine from hours to only a few seconds, converting time-consuming analytics processes into totally interactive sessions for the Data Scientists. Finally, I will present how the results of the recommendation engine can be coupled with a Natural Language interface to provide an easy-to-use search platform.

Short Bio:
Michalis Vlachos is a Professor with the Faculty of Business and Economics (HEC) at the University of Lausanne, Switzerland. Previously, he worked at IBM Research in Zurich., and at IBM T.J. Watson Research Center, NY. He has also visited Microsoft Research. His research interests span the areas of business analytics, data mining and recommender systems. He holds a Ph.D. in Computer Science from the University of California, Riverside, and an MBA from University of Illinois, Urbana-Champaign. Dr. Vlachos has authored 100+ publications and holds over 20 granted or filed patents. For his work at IBM, he has received seven Technical Accomplishment Awards and five Invention Plateau Awards. He has been a recipient of two best-paper awards, a Fulbright Scholarship and a Marie-Curie International Reintegration Grant. He has been awarded an ERC grant on the topic "Exact Mining from InExact Data." He is an ACM Distinguished Speaker and a Senior Member of the IEEE.

Note:
Video Recording: This presentation will be recorded and available after the presentation through the "Multimedia file" URL below.

 Recording: https://www.youtube.com/watch?v=jYV_M1rgFK0

  Web: https://www.cs.ucy.ac.cy/colloquium/
  Mailing List: https://listserv.cs.ucy.ac.cy/mailman/listinfo/cs-colloquium
  RSS: https://www.cs.ucy.ac.cy/colloquium/rss.xml
  Calendar: https://www.cs.ucy.ac.cy/colloquium/schedule/cs.ucy.2020.vlachos.ics