Personalised Search and Result Presentation in Semantic Search


June 30, 2017 - 16:00

This Friday, we’ll have two talks followed by drinks.

First the academic talk is given by Faegheh Hasibi, who is a PhD candidate at the Norwegian University of Science and Technology (NTNU) under the supervision of Prof. Krisztian Balog (University of Stavanger).

Then our industrial talk is given by Barrie Kersbergen a senior data scientist at

This edition of SEA will be held in SPUI25.


16:00 – 16:30 Faegheh Hasibi

16:30 – 17:00 Barrie Kersbergen

17:00 – 18:00 Drinks & Snacks

Details of the talks:


Faegheh Hasibi — Semantic Search and Result Presentation with Entity Cards

Modern search engines have moved from the traditional “ten blue links” environment towards understanding searchers’ intent and providing them with the focused responses; a paradigm that is referred to as “semantic search”. Semantic search is an umbrella term that encompasses various techniques, including but not limited to, query understanding, entity retrieval, and result presentation. In this talk, I will give a brief overview on each of these tasks, and further focus on the result presentation aspect. Specifically, I will present methods for generating content for “entity cards”​, the informational panels that are presented at the right column of search engine results pages. I will end the talk by introducing a practical toolkit and dataset that are meant to foster research in this area.

Faegheh Hasibi is a PhD candidate at the Norwegian University of Science and Technology, supervised by Prof. Krisztian Balog (University of Stavanger). She received her M.Sc. from the Chalmers University of Technology and University of Gothenburg. Her research is focused on improving various tasks of semantic search, including query understanding, entity linking, retrieval, and summarization. You can read more about her research interests and activities at her personal homepage.


Barrie Kersbergen — Personalized recommendations for anonymous visitors

This presentation focuses on the development of a system that is used to personalize content for anonymous visitors on the website. The result is a distributed, time critical self learning and predictive system. Insights will be shared in, for example, finding a balance between speed and accuracy of prediction quality.

Barrie Kersbergen is a Senior Data scientist at since 2010. His main focus is on designing and building artificial intelligent, time critical and non time critical systems for further improving customer experience by personalizing online content.