Generalized vector space model: Difference between revisions

Revision as of 08:35, 6 August 2013

Latest revision as of 19:28, 29 January 2023

(20 intermediate revisions by 18 users not shown)

Line 1:

{{short description|Generalization of the vector space model used in information retrieval}}

I LIKE TO EAT SHRIMP AND READ BOOKS LOL :D LOLOL

{{Confusing|date=January 2010}}

The '''Generalized vector space model''' is a generalization of the [[vector space model]] used in [[information retrieval]]. Wong ''et al.''<ref name="wong">{{citation | chapter=Generalized vector spaces model in information retrieval | first1=S. K. M. | title=Proceedings of the 8th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '85 | pages=18–25 | last1=Wong | first2=Wojciech|last2=Ziarko|first3=Patrick C. N.|last3=Wong | publisher=[[Association for Computing Machinery|SIGIR ACM]] | date=1985-06-05| doi=10.1145/253495.253506 | isbn=0897911598 | doi-access=free }}</ref> presented an analysis of the problems that the pairwise orthogonality assumption of the [[vector space model]] (VSM) creates. From here they extended the VSM to the generalized vector space model (GVSM).

==Definitions==

GVSM introduces term to term correlations, which deprecate the pairwise orthogonality assumption. More specifically, the factor considered a new space, where each term vector ''t_i'' was expressed as a linear combination of ''2ⁿ'' vectors ''m_r'' where ''r = 1...2ⁿ''.

For a document ''d_k'' and a query ''q'' the similarity function now becomes:

:<math>sim(d_k,q) = \frac{\sum _{j=1}^n \sum _{i=1}^n w_{i,k}*w_{j,q}*t_i \cdot t_j }{\sqrt{\sum _{i=1}^n w_{i,k}^2}*\sqrt{\sum _{i=1}^n w_{i,q}^2}}</math>

where ''t_i'' and ''t_j'' are now vectors of a ''2ⁿ'' dimensional space.

Term correlation <math>t_i \cdot t_j</math> can be implemented in several ways. For an example, Wong et al. uses the term occurrence frequency matrix obtained from automatic indexing as input to their algorithm. The term occurrence and the output is the term correlation between any pair of index terms.

==Semantic information on GVSM==

There are at least two basic directions for embedding term to term relatedness, other than exact keyword matching, into a retrieval model:

# compute semantic correlations between terms

# compute frequency co-occurrence statistics from large corpora

Recently Tsatsaronis<ref>{{citation | title=A Generalized Vector Space Model for Text Retrieval Based on Semantic Relatedness | url=http://www.aclweb.org/anthology/E/E09/E09-3009.pdf | last1= Tsatsaronis | first1=George | first2=Vicky | last2=Panagiotopoulou | publisher=[[Association for Computing Machinery|EACL ACM]] |date=2009-04-02}}</ref> focused on the first approach.

They measure semantic relatedness (''SR'') using a thesaurus (''O'') like [[WordNet]]. It considers the path length, captured by compactness (''SCM''), and the path depth, captured by semantic path elaboration (''SPE'').

They estimate the <math>t_i \cdot t_j</math> inner product by:

<math>t_i \cdot t_j = SR((t_i, t_j), (s_i, s_j), O)</math>

where ''s_i'' and ''s_j'' are senses of terms ''t_i'' and ''t_j'' respectively, maximizing <math>SCM \cdot SPE</math>.

Building also on the first approach, Waitelonis et al.<ref>{{citation | title=Linked Data enabled Generalized Vector Space Model to improve document retrieval | url=http://ceur-ws.org/Vol-1581/paper4.pdf | last1= Waitelonis | first1=Jörg | first2=Claudia | last2=Exeler | last3=Sack| first3=Harald | publisher=ISWC 2015, CEUR-WS 1581 |date=2015-09-11}}</ref> have computed semantic relatedness from [[Linked data|Linked Open Data]] resources including [[DBpedia]] as well as the [[YAGO (database)|YAGO taxonomy]].

Thereby they exploits taxonomic relationships among semantic entities in documents and queries after [[Entity linking|named entity linking]].

== References ==

{{reflist}}

[[Category:Vector space model]]