On the potential of domain literature for clustering and Bayesian network learning
Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining
Artificial Intelligence in Medicine
Using literature and data to learn Bayesian networks as clinical models of ovarian tumors
Artificial Intelligence in Medicine
| Hi-index | 0.00 |
The increasing mount and variety of domain knowledge and the v ilability ofincreasingly l rge quantities of electronic liter ture requires new types of support for thedevelopment of complex knowledge models.In previous publications we proposed theapplication of so c lled Annotated Bayesian Networks (ABN),textually enriched probabilisticdomain models,which help knowledge engineers and medical experts to find and organize theinformation necess ry in model building.In this paper we describe n information retriev llanguage in which the formalized domain knowledge nd the attached textual information c nbe accessed in n integrated fashion and can be used to define various retrieval schemes andrelevance measures.This language,on one hand,provides maximum flexibility for knowledgeengineers to exploit the v ilable annotated domain model s contextual inform tion.On theother hand,it allows the definition of complex,high-level queries,in which the contextual useof the annotated domain model can be optimized for clinical situations.We compare theperformance of the standard and the proposed query language in the ovarian c ncer domain.