On automatic knowledge validation for Bayesian knowledge bases

Santos, Eugene; Dinh, Hang

doi:10.1016/j.datak.2007.07.004

Cited by 10 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, we want to strength our evaluation results on internal accuracy of our user model (Nguyen & Santos, 2007a) as well as the use of prior knowledge in our user modeling approach (Nguyen & Santos, 2007b). Finally, we are mapping our current approach to modeling context into Bayesian knowledge-bases (Santos & Santos, 1998;Santos & Dinh, 2008) to better utilize probabilistic semantics for uncertainty like its use in our preference network.…”

Section: Discussionmentioning

confidence: 99%

“…Uncertainty is one of the key challenges in modeling a user for IR, as mentioned earlier. While there are some other approaches to modeling uncertainty such as Dempster-Shafer theory (Shafer, 1976), we selected Bayesian networks (Pearl, 1988) since it provides a mathematically sound model of uncertainty and we have expertise in efficiently building and reasoning over them (Santos et al, 2003c;Santos & Dinh, 2008). The novelty of our approach lies with the fine-grained representation of a user model, the ability to learn user knowledge incrementally and dynamically, and the evaluation framework to assess the effectiveness of a user model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Modeling Users for Adaptive Information Retrieval by Capturing User Intent

Santos

Nguyen

2009

Collaborative and Social Information Retrieval and Access

View full text Add to dashboard Cite

In this chapter, we study and present our results on the problem of employing a cognitive user model for Information Retrieval (IR) in which a user’s intent is captured and used for improving his/her effectiveness in an information seeking task. The user intent is captured by analyzing the commonality of the retrieved relevant documents. The effectiveness of our user model is evaluated with regards to retrieval performance using an evaluation methodology which allows us to compare with the existing approaches from the information retrieval community while assessing the new features offered by our user model. We compare our approach with the Ide dec-hi approach using term frequency inverted document frequency weighting which is considered to be the best traditional approach to relevance feedback. We use CRANFIELD, CACM and MEDLINE collections which are very popular collections from the information retrieval community to evaluate relevance feedback techniques. The results show that our approach performs better in the initial runs and works competitively with Ide dec-hi in the feedback runs. Additionally, we evaluate the effects of our user modeling approach with human analysts. The results show that our approach retrieves more relevant documents to a specific analyst compared to keyword-based information retrieval application called Verity Query Language.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Modeling Users for Adaptive Information Retrieval by Capturing User Intent

Santos

Nguyen

2009

Collaborative and Social Information Retrieval and Access

View full text Add to dashboard Cite

show abstract

“…Also, validity is only tested on a boolean level, and no sequential test knowledge can be represented. Santos and Dinh [46] introduce a validation framework for Bayesian knowledge bases. Since the exact comparison of solutions' probabilities is not reasonable in the general case, the validity of the expected and derived solutions is defined by ordering the solutions' ratings, i.e., a test case is valid if there exists no incorrect solution with a higher probability than a correct solution.…”

Section: Related Workmentioning

confidence: 99%

Advanced empirical testing

Baumeister

2011

Knowledge-Based Systems

View full text Add to dashboard Cite

In today's industrial applications, we see that knowledge systems are successfully implemented. However, critical domains require the elaborate and thoughtful validation of the knowledge bases before the deployment. Empirical testing, also known as regression testing, denotes the most popular validation technique, where predefined test cases are used to simulate and review the correct behavior of the system. In this paper, we motivate that the classic notions of a test case and the corresponding measures are not sufficient in many application scenarios. We present enhanced notions generalizing the standard test case, and we show appropriate extensions of the measures precision and recall, that work on these test case notions. Furthermore, the effective inspection of test runs is important whenever test cases fail. We introduce a novel visualization technique that allows for the effective and intuitive analysis of test cases and test run outcomes. The new visualization is useful for debugging a knowledge base and test case, respectively, but it also provides an intuitive overview of the status of the entire test suite. A case study reports on the (repeated) validation of a medical decision-support system and demonstrates the practical relevance of the presented work.

show abstract

Automated knowledge base management: A survey

Martinez‐Gil

2015

Computer Science Review

View full text Add to dashboard Cite

A fundamental challenge in the intersection of Artificial Intelligence and Databases consists of developing methods to automatically manage Knowledge Bases which can serve as a knowledge source for computer systems trying to replicate the decision-making ability of human experts. Despite of most of tasks involved in the building, exploitation and maintenance of KBs are far from being trivial, significant progress has been made during the last years. However, there are still a number of challenges that remain open. In fact, there are some issues to be addressed in order to empirically prove the technology for systems of this kind to be mature and reliable.

show abstract

On automatic knowledge validation for Bayesian knowledge bases

Cited by 10 publications

References 31 publications

Modeling Users for Adaptive Information Retrieval by Capturing User Intent

Modeling Users for Adaptive Information Retrieval by Capturing User Intent

Advanced empirical testing

Automated knowledge base management: A survey

Contact Info

Product

Resources

About