AGENTS VIC Meeting
Friday, 29 July 2011 @ 9:30AM.
Coffee, tea and refreshments from 9:15.
University of Melbourne, Rm 5.08, ICT Building (Dept of CSSE), 111 Barry St.
Jan Richter, Swinburne University of Technology
Establishing Composite SLAs through Concurrent QoS Negotiation with Surplus Redistribution
The end-to-end QoS negotiation for SLA establishment for composite services involves compound multi-party negotiations in which the composite service provider concurrently negotiates with multiple candidates for each atomic service, selecting the one that best satisfies the atomic service QoS preferences while ensuring that the end-to-end QoS requirements are also fulfilled. In order to be able to negotiate with potential candidates, it is necessary to derive the atomic utility boundaries from the global utility boundary. Additionally, there has to be a mechanism for updating these boundaries in subsequent negotiation rounds based upon the individual negotiation outcomes. In this paper, we propose an algorithm for the decomposition of global utility boundary into atomic service utility boundaries, and the surplus redistribution from successful negotiation outcomes among the remaining negotiations. The proposed mechanism is a practical approach to efficiently coordinate concurrent service negotiations within complex workflows, enabling the iterative and interactive adjustment of the negotiation boundaries for each atomic service in a composition based on the performance of other atomic negotiations. We demonstrate the feasibility of our approach by evaluating it with some popular negotiation strategies using the Specialised Property Search Scenario.
Daniel Macias, RMIT University
Building Modular Knowledge Bases for Conversational Agents
The process of constructing domain-specific ontologies presents challenges in the time and human effort required. Although some efforts have been made to automate this process using hierarchical relations, relatively little has been done on incorporating other types of semantic associations between concepts. We describe MKBUILD, a tool that follows a methodology to create domain-specific ontologies containing related concepts, drawing from existing large-scale resources such as WordNet and Wikipedia/DBPedia. The context for this work is to provide Modular Knowledge Bases (MKBs) for a conversational agent designed to operate on a mobile platform with a small computational footprint. The MKBs that we generate will be utilised by the conversational agent when processing speech fragments and for generating coherent narrative structure for free-flowing conversations. In order to obtain semantic associations between concepts in the ontology that we generate, we use hierarchical relations and word senses which are obtained from WordNet and semantic associations between concepts which are obtained from Wikipedia. As an initial evaluation, we ask human participants to rate the relevance of concepts in constructed domain-specific ontologies to the associated domain, using sample ontologies created using our technique. We obtain promising results, as participants consider above 68% of the concepts of sample ontologies to be relevant to the domain.
