Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | 2021 IEEE International Conference on Big Knowledge |
| Untertitel | (ICBK) |
| Herausgeber/-innen | Zhiguo Gong, Xue Li, Sule Gunduz Oguducu, Lei Chen, Baltasar Fernandez Manjon, Xindong Wu |
| Herausgeber (Verlag) | Institute of Electrical and Electronics Engineers Inc. |
| Seiten | 315-322 |
| Seitenumfang | 8 |
| ISBN (elektronisch) | 9781665438582 |
| Publikationsstatus | Veröffentlicht - 2021 |
| Veranstaltung | 12th IEEE International Conference on Big Knowledge, ICBK 2021 - Virtual, Auckland, Neuseeland Dauer: 7 Dez. 2021 → 8 Dez. 2021 |
Abstract
Decision-making tasks usually follow five steps: identifying the problem, collecting data, extracting evidence, iden-tifying arguments, and making the decision. This paper focuses on two steps of decision-making: extracting evidence by building knowledge graphs (KGs) of specialized topics and identifying sentences' arguments through sentence-level argument mining. We present a hybrid model that combines topic modeling using latent Dirichlet allocation (LDA) and word embeddings to obtain external knowledge from structured and unstructured data. We use a topic model to extract topic- and sentence-specific evidence from the structured knowledge base Wikidata. A knowledge graph is constructed based on the cosine similarity between the entity word vectors of Wikidata and the vector of the given sentence. A second graph based on topic-specific articles found via Google supplements the general incompleteness of the structured knowledge base. Combining these graphs, we obtain a graph-based model that, as our evaluation shows, successfully capitalizes on both structured and unstructured data.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Artificial intelligence
- Informatik (insg.)
- Angewandte Informatik
- Informatik (insg.)
- Information systems
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
2021 IEEE International Conference on Big Knowledge: (ICBK). Hrsg. / Zhiguo Gong; Xue Li; Sule Gunduz Oguducu; Lei Chen; Baltasar Fernandez Manjon; Xindong Wu. Institute of Electrical and Electronics Engineers Inc., 2021. S. 315-322.
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - Topic-Guided Knowledge Graph Construction for Argument Mining
AU - Li, Weichen
AU - Abels, Patrick
AU - Ahmadi, Zahra
AU - Burkhardt, Sophie
AU - Schiller, Benjamin
AU - Gurevych, Iryna
AU - Kramer, Stefan
PY - 2021
Y1 - 2021
N2 - Decision-making tasks usually follow five steps: identifying the problem, collecting data, extracting evidence, iden-tifying arguments, and making the decision. This paper focuses on two steps of decision-making: extracting evidence by building knowledge graphs (KGs) of specialized topics and identifying sentences' arguments through sentence-level argument mining. We present a hybrid model that combines topic modeling using latent Dirichlet allocation (LDA) and word embeddings to obtain external knowledge from structured and unstructured data. We use a topic model to extract topic- and sentence-specific evidence from the structured knowledge base Wikidata. A knowledge graph is constructed based on the cosine similarity between the entity word vectors of Wikidata and the vector of the given sentence. A second graph based on topic-specific articles found via Google supplements the general incompleteness of the structured knowledge base. Combining these graphs, we obtain a graph-based model that, as our evaluation shows, successfully capitalizes on both structured and unstructured data.
AB - Decision-making tasks usually follow five steps: identifying the problem, collecting data, extracting evidence, iden-tifying arguments, and making the decision. This paper focuses on two steps of decision-making: extracting evidence by building knowledge graphs (KGs) of specialized topics and identifying sentences' arguments through sentence-level argument mining. We present a hybrid model that combines topic modeling using latent Dirichlet allocation (LDA) and word embeddings to obtain external knowledge from structured and unstructured data. We use a topic model to extract topic- and sentence-specific evidence from the structured knowledge base Wikidata. A knowledge graph is constructed based on the cosine similarity between the entity word vectors of Wikidata and the vector of the given sentence. A second graph based on topic-specific articles found via Google supplements the general incompleteness of the structured knowledge base. Combining these graphs, we obtain a graph-based model that, as our evaluation shows, successfully capitalizes on both structured and unstructured data.
KW - Argument mining
KW - Knowledge graph
KW - Topic model
UR - http://www.scopus.com/inward/record.url?scp=85125065684&partnerID=8YFLogxK
U2 - 10.1109/ICKG52313.2021.00049
DO - 10.1109/ICKG52313.2021.00049
M3 - Conference contribution
AN - SCOPUS:85125065684
SP - 315
EP - 322
BT - 2021 IEEE International Conference on Big Knowledge
A2 - Gong, Zhiguo
A2 - Li, Xue
A2 - Oguducu, Sule Gunduz
A2 - Chen, Lei
A2 - Manjon, Baltasar Fernandez
A2 - Wu, Xindong
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 12th IEEE International Conference on Big Knowledge, ICBK 2021
Y2 - 7 December 2021 through 8 December 2021
ER -