This shows you the differences between two versions of the page.

Link to this comparison view

vlado:pub:sreda:1213 [2012/06/04 13:26]
vlado created
vlado:pub:sreda:1213 [2012/06/04 13:36] (current)
Line 1: Line 1:
-====== 1211. sredin seminar, 16maj 2012 ====== +====== 1213. sredin seminar, 13junij 2012 ====== 
-Vladimir Batagelj <vladimir.batagelj@fmf.uni-lj.si>+Klemen Simonič <klemen.simonic@ijs.si>
-**Bibliometrična omrežja**+**Linked data: Predicting missing properties**
 +The volume of available structured data is increasing, particularly
 +in the form of Linked Data, where relationships between individual
 +pieces of data are encoded by a graph-like structure. The structure 
 +is a multigraph where: concepts define nodes and properties and 
 +relationships are represented by edges. Despite increasing scales 
 +of the data, the use and applicability of these resources is 
 +currently limited by mistakes and omissions in the linking data.
-Na seminarju bom predstavil nekaj novih zamisli pri analizi +In this talk, we look at the missing property recommendation  
-bibliometričnih omrežij+problemGiven a specific query node in our multigraph dataset,  
- +can we correctly rank possibly omitted properties by likelihood.  
-Najprej si bomo ogledali omrežja sodelovanjZ Moniko Cerinšek sva +We propose a general method based on leveraging properties from  
-v analizi omrežja Zentralblatt vpeljala indeks sodelovalnosti +similar nodes in our dataset. To determine the likelihoodwe use 
-posameznega avtorja pri predpostavkida vsi soavtorji prispevajo +weighted averages of property counts to determine the most likely 
-enak delež k posameznemu delu. Na seminarju bom predstavil posplošitev +missing property. To determine similar nodeswe use a variety of  
-in matematično ozadje pristopa za primerko so ti deleži lahko tudi +local, global and external measures of similarityFinally, we  
-različni+present a comprehensive evaluation of the performance of our  
- +approach for a number of different averaging schemes and similarity 
-V drugem delu si bomo ogledali odpravljanje ciklov v omrežjih +measures on three very large-scale datasetstwo based on DBpedia 
-sklicevanj z uporabo transformacije "preprint". Pri tej potrebujemo +and one based on Freebase.
-sezname vseh poti med pari točk. Predstavil bom preprost polkolobar+
-ki omogoča njih izračun. +
 [[vlado:pub:sreda|Sreda/wiki]] [[vlado:pub:sreda|Sreda/wiki]]
vlado/pub/sreda/1213.txt · Last modified: 2012/06/04 13:36 by vlado
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Noncommercial-Share Alike 3.0 Unported
Recent changes RSS feed Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki