Filaments of Meaning in Word Space

@incollection{citeulike:2760043, abstract = {Word space models, in the sense of vector space models built on distributional data taken from texts, are used to model semantic relations between words. We argue that the high dimensionality of typical vector space models lead to unintuitive effects on modeling likeness of meaning and that the local structure of word spaces is where interesting semantic relations reside. We show that the local structure of word spaces has substantially different dimensionality and character than the global space and that this structure shows potential to be exploited for further semantic analysis using methods for local analysis of vector space structure rather than globally scoped methods typically in use today such as singular value decomposition or principal component analysis.}, author = {Karlgren, Jussi and Holst, Anders and Sahlgren, Magnus}, citeulike-article-id = {2760043}, doi = {10.1007/978-3-540-78646-7_52}, journal = {Advances in Information Retrieval}, keywords = {hal, projection, random, semantic, space, spaces, term}, pages = {531--538}, posted-at = {2008-06-23 15:02:44}, priority = {0}, title = {Filaments of Meaning in Word Space}, url = {http://dx.doi.org/10.1007/978-3-540-78646-7_52}, year = {2008} }

TY - CHAP ID - citeulike:2760043 L3 - citeulike-article-id:2760043 N2 - Word space models, in the sense of vector space models built on distributional data taken from texts, are used to model semantic relations between words. We argue that the high dimensionality of typical vector space models lead to unintuitive effects on modeling likeness of meaning and that the local structure of word spaces is where interesting semantic relations reside. We show that the local structure of word spaces has substantially different dimensionality and character than the global space and that this structure shows potential to be exploited for further semantic analysis using methods for local analysis of vector space structure rather than globally scoped methods typically in use today such as singular value decomposition or principal component analysis. JF - Advances in Information Retrieval EP - 538 TI - Filaments of Meaning in Word Space SP - 531 KW - hal KW - projection KW - random KW - semantic KW - space KW - spaces KW - term AU - Karlgren, Jussi AU - Holst, Anders AU - Sahlgren, Magnus PY - 2008/// UR - http://dx.doi.org/10.1007/978-3-540-78646-7_52 ER -