Abstract | ||
---|---|---|
Agent applications are often viewed as unduly expensive to develop and maintain in commercial contexts. Organizations often settle for less sophisticated and more traditional software in place of agent technology because of (often misplaced) fears about the development and maintenance costs of agent technology, and the often mistaken perception that traditional software offers better returns on investment. This paper aims to redress this by developing a plan recognition framework for agent program learning, where behavior logs of legacy applications (or even manually executed processes) are mined to extract a 'draft' version of agent code that could eventually replace these applications or processes. We develop and implement techniques for inferring agent plans, specifically inferring agent goals. We propose two ways to infer goals for plans without and with a goal library respectively. Besides, a preferred goal is considered when a goal library is provided, using the notions of consistency, maximal entailment and minimality. The complexity of the plan recognition framework is analyzed and the experimental results show that the average runtime for generating Belief-Desire-Intention (BDI) plans relying on the number of expansion nodes, choice branching factor and parallel branching factor in workflow nets (WF-nets), and that the plan recognition framework is feasible and computable. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1109/INCoS.2013.72 | INCoS |
Keywords | Field | DocType |
expansion nodes,bdi goal recognition,plan recognition framework,planning (artificial intelligence),petri nets,agent goal inference,belief-desire-intention,wf-nets,learning (artificial intelligence),goal library,multi-agent systems,inferring agent plan,(bdi) goal recognition,agent technology,(bdi) plan recognition,inferring agent goal,preferred goal,agent program learning,traditional software,agent code,workflow nets,choice branching factor,legacy applications,agent application,(bdi) context recognition,learning artificial intelligence,multi agent systems | Logical consequence,Petri net,Computer science,Redress,Multi-agent system,Software,Artificial intelligence,Legacy system,Distributed computing,Software engineering,Plan recognition,Perception,Machine learning | Conference |
Citations | PageRank | References |
0 | 0.34 | 11 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hongyun Xu | 1 | 48 | 3.10 |
Youqun Shi | 2 | 11 | 5.04 |
Qiying Cao | 3 | 172 | 13.39 |