TY - GEN

T1 - Automated detection of non-relevant posts on the russian imageboard “2ch”

T2 - 6th International Conference on Analysis of Images, Social Networks and Texts, AIST 2017

AU - Bakarov, Amir

AU - Gureenkova, Olga

PY - 2018/1/1

Y1 - 2018/1/1

N2 - This study considers the problem of automated detection of non-relevant posts on Web forums and discusses the approach of resolving this problem by approximation it with the task of detection of semantic relatedness between the given post and the opening post of the forum discussion thread. The approximated task could be resolved through learning the supervised classifier with a composed word embeddings of two posts. Considering that the success in this task could be quite sensitive to the choice of word representations, we propose a comparison of the performance of different word embedding models. We train 7 models (Word2Vec, Glove, Word2Vec-f, Wang2Vec, AdaGram, FastText, Swivel), evaluate embeddings produced by them on dataset of human judgements and compare their performance on the task of non-relevant posts detection. To make the comparison, we propose a dataset of semantic relatedness with posts from one of the most popular Russian Web forums, imageboard “2ch”, which has challenging lexical and grammatical features.

AB - This study considers the problem of automated detection of non-relevant posts on Web forums and discusses the approach of resolving this problem by approximation it with the task of detection of semantic relatedness between the given post and the opening post of the forum discussion thread. The approximated task could be resolved through learning the supervised classifier with a composed word embeddings of two posts. Considering that the success in this task could be quite sensitive to the choice of word representations, we propose a comparison of the performance of different word embedding models. We train 7 models (Word2Vec, Glove, Word2Vec-f, Wang2Vec, AdaGram, FastText, Swivel), evaluate embeddings produced by them on dataset of human judgements and compare their performance on the task of non-relevant posts detection. To make the comparison, we propose a dataset of semantic relatedness with posts from one of the most popular Russian Web forums, imageboard “2ch”, which has challenging lexical and grammatical features.

KW - 2ch

KW - Compositional semantics

KW - Distributional semantics

KW - Imageboard

KW - Semantic relatedness

KW - Word embeddings

KW - Word similarity

KW - Word similarity Word embeddings

KW - Compositional semantics 2ch

UR - http://www.scopus.com/inward/record.url?scp=85039438003&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-73013-4_2

DO - 10.1007/978-3-319-73013-4_2

M3 - Conference contribution

AN - SCOPUS:85039438003

SN - 9783319730127

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 16

EP - 21

BT - Analysis of Images, Social Networks and Texts - 6th International Conference, AIST 2017, Revised Selected Papers

A2 - VanDerAalst, WMP

A2 - Ignatov, DI

A2 - Khachay, M

A2 - Kuznetsov, SO

A2 - Lempitsky, null

A2 - Lomazova, IA

A2 - Loukachevitch, N

A2 - Napoli, A

A2 - Panchenko, A

A2 - Pardalos, PM

A2 - Savchenko, AV

A2 - Wasserman, S

PB - Springer-Verlag GmbH and Co. KG

Y2 - 27 July 2017 through 29 July 2017

ER -