No abstract

When working in an unfamiliar online environment, it can be helpful to have an observer that can intervene and guide a user toward a desirable outcome while avoiding undesirable outcomes or frustration. The Intervention Problem is deciding when to intervene in order to help a user. The Intervention Problem is similar to, but distinct from, Plan Recognition because the observer must not only recognize the intended goals of a user but also when to intervene to help the user when necessary. We formalize a family of Intervention Problems and show that how these problems can be solved using a combination of Plan Recognition methods and classification algorithms to decide whether to intervene. For our benchmarks, the classification algorithms dominate three recent Plan Recognition approaches. We then generalize these results to Human-Aware Intervention, where the observer must decide in real time whether to intervene human users solving a cognitively engaging puzzle. Using a revised feature set more appropriate to human behavior, we produce a learned model to recognize when a human user is about to trigger an undesirable outcome. We perform a human-subject study to evaluate the Human-Aware Intervention. We find that the revised model also dominates existing Plan Recognition algorithms in predicting Human-Aware Intervention.

Service oriented architecture (SOA) is a widely used model for enterprise application integration, mainly due to the presence of accepted standards and tools required to implement web services. While the service oriented architecture provides an elegant model for interoperability, it does not define a computational model for building the endpoints of complex distributed services it exposes in a typical enterprise application. On the other hand, software agent technology provides powerful models such as the Belief-Desire-Intention (BDI) model that allows developers to build distributed systems with enhanced reasoning capabilities. We propose a web services based communication model for interagent communication allowing heterogeneous agents in a multi agent system to freely collaborate. We specifically focus on goal and beliefset sharing in BDI based agents. By providing a web services based interface, we allow agents as well as non-agent based services in a distributed system to yield the benefits of agent technology. Applicability of this proposition is demonstrated by designing and implementing a prototype multi-agent system for flood forecasting.

This paper proposes a taxonomy of experimental usecase scenarios to facilitate research into trustworthy autonomous systems (TAS). Unable to identify an open-access repository of usecases to support our research, the project team embarked on development of an online library where fellow researchers would be able to find, share and recommend usecases to other practitioners in the field. To organise the library's content, we needed a taxonomy and, informed by a commitment to the conduct of responsible research and innovation (RRI), we prioritised stakeholder involvement to shape its development. Conflict arose, however, between the project team's objective-a rigorous taxonomy focused on surfacing genuine "benchmarks" that can be used to test a multiplicity of variables in a range of domains under differing experimental conditions-and stakeholder expectation that the library would provide details of particular studies and results. How then can we deliver a product that meets project requirements while ensuring that it is genuinely useful to-and respects the needs and preferences of-a diverse range of stakeholders? A practical solution has to be found. CCS CONCEPTS• Information systems → Digital libraries and archives; Collaborative and social computing systems and tools.

No abstract