Knowledge Fragment Enrichment Using Domain Knowledge Base
Knowledge fragment enrichment aims to complete user input concept fragment by augmenting each concept with rich domain information. This is a widely studied problem in cognitive science, but has not been intensively investigated in computer science. In th
- PDF / 490,067 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 58 Downloads / 239 Views
Renmin University of China, Beijing, China [email protected] 2 University of Illinois at Urbana-Champaign, Champaign, USA {hzhuang3,ysong44,hanj}@illinois.edu 3 Tsinghua University, Beijing, China [email protected], {jietang,lijuanzi}@tsinghua.edu.cn http://www.springer.com/lncs
Abstract. Knowledge fragment enrichment aims to complete user input concept fragment by augmenting each concept with rich domain information. This is a widely studied problem in cognitive science, but has not been intensively investigated in computer science. In this paper, we formally define the problem of knowledge fragment enrichment in domain knowledge base and develop a probabilistic graphical model to tackle the problem. The proposed model is able to model the dependencies among concepts in the input knowledge fragment and also capture the probabilistic relationship between concepts and domain entities. We empirically evaluate the proposed model on two different genres of datasets: PubMed and NSFC. On both datasets, the proposed model significantly improves the accuracy of label prediction task by up to 3–9 % (in terms of MAP) compared with several alternative enrichment methods.
Keywords: Heterogeneous information networks modeling · Semi-supervised labeling
1
·
Hierarchical topic
Introduction
Many human cognitive activities can be viewed as a process of enriching a given knowledge fragment. For example, in the quiz show Jeopardy!1 , contestants are presented with knowledge clues (or knowledge fragment), and must answer related questions. Specifically, contestants need to first enrich the fragment with knowledge in their mind and then match the question with the enriched knowledge fragment. This problem has been intensively studied in cognitive science, such as the theory of knowledge representation—cognitive map proposed by [15] and cognitive model—spreading-activation model proposed by [7]. However, due 1
http://en.wikipedia.org/wiki/Jeopardy!.
c Springer Nature Singapore Pte Ltd. 2016 Y. Li et al. (Eds.): SMP 2016, CCIS 669, pp. 274–286, 2016. DOI: 10.1007/978-981-10-2993-6 24
Knowledge Fragment Enrichment Using Domain Knowledge Base
275
Input Output Computer Science
Computing Applications
Computer Graphics
Knowledge Discovery
Information Security
Computing Science CoRR user Peter J. Denning Commun. ACM large-scale J. Hartmanis J. of Computer Science intelligence John E. Hopcroft
Computing Applications R. Buyya AAAI recognition IJCAI representation Eyal Kushilevitz D. Dobkin WWW description
Computer Graphics Pat Hanrahan SIGGRAPH reality geometry G. S. Owen SCA Eurographic 3D models D. Greenberg
KDD TKDE ICDM
Network k Security MOBICOM network Adrian Perrig Jean-Pierre Hubaux SIGCOMM router INFOCOM protocol Srdjan Capkun
Knowledge Discovery G. Piatetsky-Shapiro data mining Usama M. Fayyad clustering Christos Faloutsos knowledge
Fig. 1. Illustrative example of knowledge fragment enrichment. The left figure shows the input, where a knowledge fragment represented as a concept hierarchy of comput
Data Loading...
