Learning from Ontology Streams with Semantic Concept Drift

Jiaoyan Chen, Freddy Lécué, Jeff Z. Pan, Huajun Chen*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

Data stream learning has been largely studied for extracting knowledge structures from continuous and rapid data records. In the semantic Web, data is interpreted in ontologies and its ordered sequence is represented as an ontology stream. Our work exploits the semantics of such streams to tackle the problem of concept drift i.e., unexpected changes in data distribution, causing most of models to be less accurate as time passes. To this end we revisited (i) semantic inference in the context of supervised stream learning, and (ii) models with semantic embeddings. The experiments show accurate prediction with data from Dublin and Beijing.

Original languageEnglish
Title of host publication26th International Joint Conference on Artificial Intelligence, IJCAI 2017
EditorsCarles Sierra
PublisherAAAI Press / International Joint Conferences on Artificial Intelligence
Pages957-963
Number of pages7
ISBN (Electronic)9780999241103
DOIs
Publication statusPublished - 2018
Event26th International Joint Conference on Artificial Intelligence, IJCAI 2017 - Melbourne, Australia
Duration: 19 Aug 201725 Aug 2017

Conference

Conference26th International Joint Conference on Artificial Intelligence, IJCAI 2017
CountryAustralia
CityMelbourne
Period19/08/1725/08/17

Fingerprint

Ontology
Semantics
Semantic Web
Experiments

Keywords

  • knowledge representation
  • reasoning and logic
  • decription logics and ontologies
  • machine learning
  • time series/data streams

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Chen, J., Lécué, F., Pan, J. Z., & Chen, H. (2018). Learning from Ontology Streams with Semantic Concept Drift. In C. Sierra (Ed.), 26th International Joint Conference on Artificial Intelligence, IJCAI 2017 (pp. 957-963). AAAI Press / International Joint Conferences on Artificial Intelligence. https://doi.org/10.24963/ijcai.2017/133

Learning from Ontology Streams with Semantic Concept Drift. / Chen, Jiaoyan; Lécué, Freddy; Pan, Jeff Z.; Chen, Huajun.

26th International Joint Conference on Artificial Intelligence, IJCAI 2017. ed. / Carles Sierra. AAAI Press / International Joint Conferences on Artificial Intelligence, 2018. p. 957-963.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chen, J, Lécué, F, Pan, JZ & Chen, H 2018, Learning from Ontology Streams with Semantic Concept Drift. in C Sierra (ed.), 26th International Joint Conference on Artificial Intelligence, IJCAI 2017. AAAI Press / International Joint Conferences on Artificial Intelligence, pp. 957-963, 26th International Joint Conference on Artificial Intelligence, IJCAI 2017, Melbourne, Australia, 19/08/17. https://doi.org/10.24963/ijcai.2017/133
Chen J, Lécué F, Pan JZ, Chen H. Learning from Ontology Streams with Semantic Concept Drift. In Sierra C, editor, 26th International Joint Conference on Artificial Intelligence, IJCAI 2017. AAAI Press / International Joint Conferences on Artificial Intelligence. 2018. p. 957-963 https://doi.org/10.24963/ijcai.2017/133
Chen, Jiaoyan ; Lécué, Freddy ; Pan, Jeff Z. ; Chen, Huajun. / Learning from Ontology Streams with Semantic Concept Drift. 26th International Joint Conference on Artificial Intelligence, IJCAI 2017. editor / Carles Sierra. AAAI Press / International Joint Conferences on Artificial Intelligence, 2018. pp. 957-963
@inproceedings{ec09d836c4b44076a430329e288c5e25,
title = "Learning from Ontology Streams with Semantic Concept Drift",
abstract = "Data stream learning has been largely studied for extracting knowledge structures from continuous and rapid data records. In the semantic Web, data is interpreted in ontologies and its ordered sequence is represented as an ontology stream. Our work exploits the semantics of such streams to tackle the problem of concept drift i.e., unexpected changes in data distribution, causing most of models to be less accurate as time passes. To this end we revisited (i) semantic inference in the context of supervised stream learning, and (ii) models with semantic embeddings. The experiments show accurate prediction with data from Dublin and Beijing.",
keywords = "knowledge representation, reasoning and logic, decription logics and ontologies, machine learning, time series/data streams",
author = "Jiaoyan Chen and Freddy L{\'e}cu{\'e} and Pan, {Jeff Z.} and Huajun Chen",
note = "This work is funded by NSFC 61473260/61673338/61672393, the Alibaba-ZJU joint project on e-Business Knowledge Graph, and the EU Marie Curie IAPP K-Drive project (286348).",
year = "2018",
doi = "10.24963/ijcai.2017/133",
language = "English",
pages = "957--963",
editor = "Carles Sierra",
booktitle = "26th International Joint Conference on Artificial Intelligence, IJCAI 2017",
publisher = "AAAI Press / International Joint Conferences on Artificial Intelligence",

}

TY - GEN

T1 - Learning from Ontology Streams with Semantic Concept Drift

AU - Chen, Jiaoyan

AU - Lécué, Freddy

AU - Pan, Jeff Z.

AU - Chen, Huajun

N1 - This work is funded by NSFC 61473260/61673338/61672393, the Alibaba-ZJU joint project on e-Business Knowledge Graph, and the EU Marie Curie IAPP K-Drive project (286348).

PY - 2018

Y1 - 2018

N2 - Data stream learning has been largely studied for extracting knowledge structures from continuous and rapid data records. In the semantic Web, data is interpreted in ontologies and its ordered sequence is represented as an ontology stream. Our work exploits the semantics of such streams to tackle the problem of concept drift i.e., unexpected changes in data distribution, causing most of models to be less accurate as time passes. To this end we revisited (i) semantic inference in the context of supervised stream learning, and (ii) models with semantic embeddings. The experiments show accurate prediction with data from Dublin and Beijing.

AB - Data stream learning has been largely studied for extracting knowledge structures from continuous and rapid data records. In the semantic Web, data is interpreted in ontologies and its ordered sequence is represented as an ontology stream. Our work exploits the semantics of such streams to tackle the problem of concept drift i.e., unexpected changes in data distribution, causing most of models to be less accurate as time passes. To this end we revisited (i) semantic inference in the context of supervised stream learning, and (ii) models with semantic embeddings. The experiments show accurate prediction with data from Dublin and Beijing.

KW - knowledge representation

KW - reasoning and logic

KW - decription logics and ontologies

KW - machine learning

KW - time series/data streams

UR - http://www.scopus.com/inward/record.url?scp=85031931222&partnerID=8YFLogxK

U2 - 10.24963/ijcai.2017/133

DO - 10.24963/ijcai.2017/133

M3 - Conference contribution

SP - 957

EP - 963

BT - 26th International Joint Conference on Artificial Intelligence, IJCAI 2017

A2 - Sierra, Carles

PB - AAAI Press / International Joint Conferences on Artificial Intelligence

ER -