Premium
Using syntagmatic relations to examine semantic communities in user generated book reviews
Author(s) -
Yoon Kyunghye,
Bezdicek Adam
Publication year - 2015
Publication title -
proceedings of the association for information science and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.193
H-Index - 14
ISSN - 2373-9231
DOI - 10.1002/pra2.2015.1450520100111
Subject(s) - syntagmatic analysis , computer science , salient , context (archaeology) , meaning (existential) , interpretation (philosophy) , linguistics , core (optical fiber) , information retrieval , world wide web , artificial intelligence , psychology , history , telecommunications , philosophy , archaeology , programming language , psychotherapist
ABSTRACT While several studies have suggested that implicit semantic communities can be identified within user generated content, the majority have focused on identifying these communities via networks of “tags” within folksonomies. This method can be problematic due to the lack of linguistic context inherent in tags as collections of discrete terms. This study attempted a different approach to this problem by analyzing a corpus of user generated reviews for popular science books taken from Amazon. It searched the reviews for patterns of shared core semantic meanings as constituted by syntagmatic relations between “topic” and “comment.” Preliminary results showed that (1) core semantic meanings expressed via “topic” and “comment” relations were salient across reviews for a given book and across books within the genre and (2) specific patterns of core semantic meanings suggested the presence of different user communities constituted through similar manners of commenting on the books. The results provide insight into the social ways that readers accumulate shared meaning around books through repetition, interpretation, annotation.