BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:Eventi DIAG
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20151025T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20160327T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.7196.field_data.0@www.ugovricerca.uniroma1.it
DTSTAMP:20260405T023903Z
CREATED:20151120T095922Z
DESCRIPTION:In many data-mining applications\, the input consists of a coll
 ection of entities (e.g.\, reviews about a product\, experts that declare 
 certain skills\,network nodes or edges) and the goal is to identify a subs
 et of important entities (e.g.\, useful reviews\, competent experts\, infl
 uential nodes respectively). Existing work identifies important entities e
 ither by entity ranking or by entity selection. Entity-ranking methods ass
 ociate a score with every entity. The main drawback of these approaches is
  that they ignore the redundancy between the highly scored entities. Entit
 y-selection methods try to overcome this drawback by evaluating the goodne
 ss of a group of entities collectively. These methods identify the best se
 t of entities\, implying that all entities not in the group are unimportan
 t. Such dichotomy of entities conceals the fact that there may be other su
 bsets of entities with equally-good (or almost as good) goodness scores. I
 n this talk\, we will discuss how the drawbacks of the above methods can b
 e overcome by integrating the entity-ranking and entity-selection paradigm
 s. That is\, we will introduce entity-ranking mechanisms that are based on
  entity selection and entity-selection mechanisms that are based on entity
  ranking. In this framework\, the importance scores of individual entities
  are determined by how many good groups of entities they participate in. C
 onsequently\, a good group of entities consists of entities with high impo
 rtance scores. The main challenge we will discuss is how to explore the so
 lution space of combinatorial problems in order to identify many entities 
 that participate in many good solutions. In the talk\, we will describe ho
 w our methods can be applied to applications related to expert management 
 systems\, management of online product reviews\, and network analysis (inc
 luding physical and social networks).Bio: Evimaria Terzi is an associate p
 rofessor at the Computer Science Department at Boston University. Before j
 oining BU in 2009\, she was a research scientist at IBM Almaden Research C
 enter. Evimaria has received her Ph.D. from University of Helsinki\, Finla
 nd and her MSc from Purdue University. Evimaria is a recipient of the Micr
 osoft Faculty Fellowship (2010)\, NSF CAREER award (2013) and multiple oth
 er NSF awards. Her research interests span a wide range of data-mining top
 ics including problems arising in online social networks and social media.
DTSTART;TZID=Europe/Paris:20151127T140000
DTEND;TZID=Europe/Paris:20151127T140000
LAST-MODIFIED:20151126T140335Z
LOCATION:Via Ariosto 25\, Room A2
SUMMARY:Evimaria Terzi: Entity Selection and Ranking for Data Mining applic
 ations - Prof. Evimaria Terzi
URL;TYPE=URI:http://www.ugovricerca.uniroma1.it/node/7196
END:VEVENT
END:VCALENDAR