[Forum SIS] Avviso di seminario

[Francesca De Battisti]

Giovedì12 ottobre alle ore 10.30, pressol’aula seminari del Dipartimento di Economia, Management e Metodi Quantitatividell’Università degli Studi di Milano (II piano, entrata da via Conservatorio,n. 7), MarkCarman ( Monash University, Australia) terrà unseminario dal titolo: 

  

“InvestigatingPerformance and Scalability for Rank Learning with Regression Tree Ensembles” 

  

Abstract   

When ranking Web pages against userqueries, a large number of signals can be leveraged to estimate the relevanceof each document, including query similarity, user-profile similarity,PageRank, etc. Rank learning algorithms provide a coherent framework forcombining these signals in order to maximize retrieval performance. As such,they have become a crucial component of current Information Retrievalinfrastructure. State-of-the-art rank learning techniques discover non-linearcombinations of features and are mostly based on ensembles of regression trees,using either bagged or boosted ensembles. With an interest in both performanceand scalability of these algorithms, we investigate the importance of differentaspects, such as the number of negative examples used, the size of subsamplesfrom which trees are learnt, and most importantly the type of objectivefunction used for recursively partition the feature space. 

  

Brief Bio 

Mark Carman is a Senior Lecturer atMonash University, a top-100 rated university in Melbourne, Australia. Hejoined Monash in 2010 after doing a postdoc at the University of Lugano. Hereceived his PhD from the University of Trento in 2006 having spent his PhDtenure at both the Fondazione Bruno Kessler (FBK) and the Information SciencesInstitute (ISI) of USC. Mark's research lies in Data Science with a particularfocus on problems in Information Retrieval. He has worked on techniques forlearning Web Search rankings, for scaling learning algorithms to large dataquantities, for robust clustering in high dimensions, for improvingquality-control in crowd-sourcing, and for personalising search results and recommendingcontent. Other applications of his work include speeding up digital forensicinvestigations, detecting sentiment and sarcasm in text, correcting errors inOCR output, and estimating user expertise in social media. Mark has authored alarge number of publications in prestigious venues, including full papers atSIGIR, KDD, IJCAI, CIKM, ECIR, WSDM, HT, CoNLL, EACL, HCOMP and ICDAR, andarticles in TOIS, IR, JMLR, ML, PR, JAIR, CS&L, JASIST, DI and CSUR.Moreover, he has served on the program committees of many IR/DM/AI conferences,including SIGIR, WSDM, CIKM, ECIR, KDD, WWW, EMNLP, ACML, IJCAI and AAAI and iscurrently an Associate Editor for the journal TOIS. 

  


 

Tutti gli interessati sono invitati a partecipare.  







***********************************************************
Francesca De Battisti
Dipartimento di Economia, Management e Metodi Quantitativi 
(III piano, studio n. 29)
Via Conservatorio 7
20122 - Milano
Tel: 02.50321464
Fax: 02.50321505
***********************************************************


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20171005/ea4e6c54/attachment.html>