Goals: a new dimension in data management and analisys for recommendation and retrieval



Dimitra Papadimitriou

Goals: a new dimension in data management and analisys for recommendation and retrieval
Speaker:
Dimitra Papadimitriou
Short Bio:
Dimitra Papadimitriou has recently received her PhD in Information Engineering and Computer Science from ICT doctoral school in University of Trento, Italy. Prior to that, she was with the data engineering group at the Informatics department of Aristotle University, Thessaloniki, Greece where she received her BSc and MSc degree in Computer Science with a specialization in Information Systems. During her PhD, she was a visitor researcher at HP Labs, Palo Alto, USA.
Abstract:
The seminar will present an overview of algorithmic solutions for performing data management and analysis tasks in a goal-aware context. Specifically, we will discuss about the problem of retrieving items of interest in response to explicit user requests or in the form of recommendations from very large item collections by considering the goals for which items are intended. These goals are those for which the items have been generated or produced; and those that may lead the user to (consume) them. We will touch upon different aspects of the design and implementation of goal-aware systems including the identification and extraction of goal-related information in the examined data sources, the modeling of the derived information (graphs, weighting schemes, clusters, indexes), the matching of the user request or previous activity to the goal model elements, and the forming of the system response. Moreover, we will discuss why and how goal-aware techniques can bring into the surface items that would have been neglected otherwise and reduce mismatches giving at the same time a different user perspective of the data set.
Date:
Thursday, May 3rd, 2018 - 16.00
Location:
DIBRIS, Valletta Puggia, Conference Room 3rd - floor 326bis

 

Sei qui: Home Ricerca Seminari di ricerca Goals: a new dimension in data management and analisys for recommendation and retrieval

Questo sito usa i cookie.