Automatic Discovery and Exploitation of Domain Knowledge in Planning
نویسنده
چکیده
Recent advances in AI planning technology have drastically improved the capabilities of modern planners. However, these advances have left research in domain specialization behind; many older specialization techniques are no longer applicable to modern planners. This paper explores how to automate domain specialization in a modern planning as constraint satisfaction engine using existing analysis and reformulation techniques. In particular, the paper examines how TIM, PABLO, and Crawford et al.’s symmetry-breaking predicates aid in automating the CPlan planning as constraint satisfaction system. ∗Thanks to Jeremy Buhler, Craig Kaplan, Rachel Pottinger, and Denise Pinnel for their help. Thanks also to my generals committee for their effort and ease of scheduling. This work was funded in part by the ARCS Foundation Tom and Barbara Cable Fellowship and by a National Science Foundation Graduate Fellowship.
منابع مشابه
The Relationship between Knowledge Management and the Process of Entrepreneurship in Sport Organizations
In the current competitive world, organizations can reach competitive advantage which support entrepreneurship by providing the required tools. One of the most important tools for developing entrepreneurship which was neglected in previous studies is organizational knowledge management. The present paper aims to shed light on the role and importance of knowledge management in sport entreprene...
متن کاملDesigning an Ontology for Knowledge Discovery in Iran’s Vaccine
Ontology is a requirement engineering product and the key to knowledge discovery. It includes the terminology to describe a set of facts, assumptions, and relations with which the detailed meanings of vocabularies among communities can be determined. This is a qualitative content analysis research. This study has made use of ontology for the first time to discover the knowledge of vaccine in Ir...
متن کاملExploiting ontologies and higher order knowledge in relational data mining Doctoral Thesis
Present day knowledge discovery tasks require mining heterogeneous and structured data and knowledge sources. The key enabling factors for performing these tasks include efficient exploitation of knowledge about the domain of discovery and utilizing meta knowledge about the data mining process, which facilitates the construction of complex workflows consisting of highly specialized algorithms. ...
متن کاملAutomatic Discovery of Technology Networks for Industrial-Scale R&D IT Projects via Data Mining
Industrial-Scale R&D IT Projects depend on many sub-technologies which need to be understood and have their risks analysed before the project can begin for their success. When planning such an industrial-scale project, the list of technologies and the associations of these technologies with each other is often complex and form a network. Discovery of this network of technologies is time consumi...
متن کاملExpert Discovery: A web mining approach
Expert discovery is a quest in search of finding an answer to a question: “Who is the best expert of a specific subject in a particular domain within peculiar array of parameters?” Expert with domain knowledge in any field is crucial for consulting in industry, academia and scientific community. Aim of this study is to address the issues for expert-finding task in real-world community. Collabor...
متن کامل