EUVIMEDEUVIMED entdecken
Open Access

Weakly Supervised Concept Map Generation Through Task-Guided Graph Translation

Jiaying Lu, Xiangjue Dong, Carl Yang

IEEE Transactions on Knowledge and Data Engineering · 01.10.2023 · 35 (10) · 10871-10883

DOI: 10.1109/tkde.2023.3252588 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Knowledge and Data Engineering, erschienen am 01.10.2023. Verfasst von Jiaying Lu und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, Emory Univeristy, Atlanta GA, 30322. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIDDK NIH HHS, internal fund, Computer Science Department of Emory University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent years have witnessed the rapid development of concept map generation techniques due to their advantages in providing well-structured summarization of knowledge from free texts. Traditional unsupervised methods do not generate task-oriented concept maps, whereas deep generative models require large amounts of training data. In this work, we present GT-D2G (Graph Translation-based Document To Graph), an automatic concept map generation framework that leverages generalized NLP pipelines to derive semantic-rich initial graphs, and translates them into more concise structures under the weak supervision of downstream task labels. The concept maps generated by GT-D2G can provide interpretable summarization of structured knowledge for the input texts, which are demonstrated through human evaluation and case studies on three real-world corpora. Further experiments on the downstream task of document classification show that GT-D2G beats other concept map generation methods. Moreover, we specifically validate the labeling efficiency of GT-D2G in the label-efficient learning setting and the flexibility of generated graph sizes in controlled hyper-parameter studies.

Schlagwörter

Stichwörter der Autor:innen

Concept Map GenerationDocument ClassificationDocument SummarizationGraph TranslationWeak Supervision

Autor:innen und Institutionen

  • Department of Computer Science, Emory Univeristy, Atlanta GA, 30322.
  • Department of Computer Science and Engineering, Texas A&M University, College Station, Texas, 77843.

Förderung

  • NIDDK NIH HHS
  • internal fund
  • Computer Science Department of Emory University

Zitieren

APA

Lu, J., Dong, X., & Yang, C. (2023). Weakly Supervised Concept Map Generation Through Task-Guided Graph Translation. IEEE Transactions on Knowledge and Data Engineering, 35(10), 10871-10883. https://doi.org/10.1109/tkde.2023.3252588

Vancouver

Lu J, Dong X, Yang C. Weakly Supervised Concept Map Generation Through Task-Guided Graph Translation. IEEE Transactions on Knowledge and Data Engineering. 2023;35(10):10871-10883. doi:10.1109/tkde.2023.3252588

BibTeX
@article{Lu2023,
  author = {Lu, Jiaying and Dong, Xiangjue and Yang, Carl},
  title = {Weakly Supervised Concept Map Generation Through Task-Guided Graph Translation},
  journal = {IEEE Transactions on Knowledge and Data Engineering},
  year = {2023},
  volume = {35},
  number = {10},
  pages = {10871--10883},
  doi = {10.1109/tkde.2023.3252588}
}
Kostenlos starten →