H. Dong, W. Wang, und F. Coenen. PRICAI 2018: Trends in Artificial Intelligence, Seite 29--41. Cham, Springer International Publishing, (2018)
Zusammenfassung
An interesting research direction is to discover structured knowledge from user generated data. Our work aims to find relations among social tags and organise them into hierarchies so as to better support discovery and search for online users. We cast relation discovery in this context to a binary classification problem in supervised learning. This approach takes as input features of two tags extracted using probabilistic topic modelling, and predicts whether a broader-narrower relation holds between them. Experiments were conducted using two large, real-world datasets, the Bibsonomy dataset which is used to extract tags and their features, and the DBpedia dataset which is used as the ground truth. Three sets of features were designed and extracted based on topic distributions, similarity and probabilistic associations. Evaluation results with respect to the ground truth demonstrate that our method outperforms existing ones based on various features and heuristics. Future studies are suggested to study the Knowledge Base Enrichment from folksonomies and deep neural network approaches to process tagging data.
Beschreibung
Learning Relations from Social Tagging Data | SpringerLink
%0 Conference Paper
%1 10.1007/978-3-319-97304-3_3
%A Dong, Hang
%A Wang, Wei
%A Coenen, Frans
%B PRICAI 2018: Trends in Artificial Intelligence
%C Cham
%D 2018
%E Geng, Xin
%E Kang, Byeong-Ho
%I Springer International Publishing
%K broader_narrower folksonomy hyponymy knowledge_graph myown ontology_learning probabilistic_association probabilistic_topic_model semantic_relations semantics social_tagging
%P 29--41
%T Learning Relations from Social Tagging Data
%X An interesting research direction is to discover structured knowledge from user generated data. Our work aims to find relations among social tags and organise them into hierarchies so as to better support discovery and search for online users. We cast relation discovery in this context to a binary classification problem in supervised learning. This approach takes as input features of two tags extracted using probabilistic topic modelling, and predicts whether a broader-narrower relation holds between them. Experiments were conducted using two large, real-world datasets, the Bibsonomy dataset which is used to extract tags and their features, and the DBpedia dataset which is used as the ground truth. Three sets of features were designed and extracted based on topic distributions, similarity and probabilistic associations. Evaluation results with respect to the ground truth demonstrate that our method outperforms existing ones based on various features and heuristics. Future studies are suggested to study the Knowledge Base Enrichment from folksonomies and deep neural network approaches to process tagging data.
%@ 978-3-319-97304-3
@inproceedings{10.1007/978-3-319-97304-3_3,
abstract = {An interesting research direction is to discover structured knowledge from user generated data. Our work aims to find relations among social tags and organise them into hierarchies so as to better support discovery and search for online users. We cast relation discovery in this context to a binary classification problem in supervised learning. This approach takes as input features of two tags extracted using probabilistic topic modelling, and predicts whether a broader-narrower relation holds between them. Experiments were conducted using two large, real-world datasets, the Bibsonomy dataset which is used to extract tags and their features, and the DBpedia dataset which is used as the ground truth. Three sets of features were designed and extracted based on topic distributions, similarity and probabilistic associations. Evaluation results with respect to the ground truth demonstrate that our method outperforms existing ones based on various features and heuristics. Future studies are suggested to study the Knowledge Base Enrichment from folksonomies and deep neural network approaches to process tagging data.},
added-at = {2018-07-31T02:17:40.000+0200},
address = {Cham},
author = {Dong, Hang and Wang, Wei and Coenen, Frans},
biburl = {https://www.bibsonomy.org/bibtex/2d3e903f52fb23507e5910d0dbe48d29b/hangdong},
booktitle = {PRICAI 2018: Trends in Artificial Intelligence},
description = {Learning Relations from Social Tagging Data | SpringerLink},
editor = {Geng, Xin and Kang, Byeong-Ho},
interhash = {08deb3faa8bfc052b6b04e73d6cbe577},
intrahash = {d3e903f52fb23507e5910d0dbe48d29b},
isbn = {978-3-319-97304-3},
keywords = {broader_narrower folksonomy hyponymy knowledge_graph myown ontology_learning probabilistic_association probabilistic_topic_model semantic_relations semantics social_tagging},
pages = {29--41},
publisher = {Springer International Publishing},
timestamp = {2018-07-31T02:18:30.000+0200},
title = {Learning Relations from Social Tagging Data},
year = 2018
}