Monitoring Online Publications about Public Authorities Activity by means of Ontological Semantics. / Dobrov, A.V; Dobrova, A.E.; Soms, N.L.
EGOSE '15 November 24-25, 2015, St.Petersburg, Russian Federation. Association for Computing Machinery, 2015.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research
}
TY - GEN
T1 - Monitoring Online Publications about Public Authorities Activity by means of Ontological Semantics
AU - Dobrov, A.V
AU - Dobrova, A.E.
AU - Soms, N.L.
PY - 2015
Y1 - 2015
N2 - In this paper, an attempt is made to describe an on-going research, aimed at creating a tool of content-analysis and opinion-mining, based on monitoring online publications about Russian federal pubic authorities activity by means of ontological semantics. This research continues a series of studies focused on the ‘agenda’ that is formed in the media on topics related to the development of e-government and online services. The problem of linguistic ambiguity is partially solved by semantical restrictions imposed by conceptual relations specified in the ontology, thus increasing precision of the analysis. Recall is also increased by means of conceptual hierarchies and synonymy. These methods have allowed to perform automatic monitoring of online publications that refer anyhow to specific public authorities, to choose a set of publications that contain evaluations of their activity, and to make a marked-up corpus from this collection. Nevertheless, the currently developed corpus shows that, in most cases, the e
AB - In this paper, an attempt is made to describe an on-going research, aimed at creating a tool of content-analysis and opinion-mining, based on monitoring online publications about Russian federal pubic authorities activity by means of ontological semantics. This research continues a series of studies focused on the ‘agenda’ that is formed in the media on topics related to the development of e-government and online services. The problem of linguistic ambiguity is partially solved by semantical restrictions imposed by conceptual relations specified in the ontology, thus increasing precision of the analysis. Recall is also increased by means of conceptual hierarchies and synonymy. These methods have allowed to perform automatic monitoring of online publications that refer anyhow to specific public authorities, to choose a set of publications that contain evaluations of their activity, and to make a marked-up corpus from this collection. Nevertheless, the currently developed corpus shows that, in most cases, the e
KW - Ontological semantics
KW - NLU
KW - conceptual relations
KW - lexical disambiguation.
M3 - статья в сборнике материалов конференции
SN - 978-1-4503-4070-0/15/11
BT - EGOSE '15 November 24-25, 2015, St.Petersburg, Russian Federation
PB - Association for Computing Machinery
ER -
ID: 4772683