Shakil M. Khan

Objective and subjective algorithms for grouping association rules

We propose two algorithms for grouping and summarizing association rules. The first algorithm recursively groups rules according to the structure of the rules and generates a tree of clusters as a result. The second algorithm groups the rules according to the semantic distance between the rules by making use of an automatically tagged semantic tree-structured network of items. We provide a case study in which the proposed algorithms are evaluated. The results show that our grouping methods are effective and produce good grouping results.

Behavioral adaptation of information systems through goal models

Customizing software to perfectly fit individual needs is becoming increasingly important in information systems engineering. Users want to be able to customize software behavior through reference to terms familiar to their diverse needs and experience. We present a requirements-driven approach to behavioral customization of software systems. Goal models are constructed to represent alternative behaviors that users can exhibit to achieve their goals. Customization information is then added to restrict the space of possibilities to those that fit specific users, contexts, or situations. Meanwhile, elements of the goal models are mapped to units of source code. This way, customization preferences posed at the requirements level are directly translated into system customizations. Our approach, which we apply to an on-line shopping cart system and an automated teller machine simulator, does not assume adoption of a particular development methodology, platform, or variability implementation technique and keeps the reasoning computation overhead from interfering with the execution of the configured application.

On the semantics of conditional commitment

In this paper, we identify some problems with current formalizations of conditional commitments, i.e. commitments to achieve a goal if some condition becomes true. We present a solution to these problems. We also formalize two types of communicative actions that can be used by an agent to request another agent to achieve a goal or perform an action provided that some condition becomes true. Our account is set within ECASL [9], a framework for modeling communicating agents based on the situation calculus.

ECASL: a model of rational agency for communicating agents

The Cognitive Agent Specification Language (CASL) is a framework for specifying and verifying complex communicating multiagent systems. In this paper, we develop an extended version, ECASL. which incorporates a formal model of means-ends reasoning suitable for a multiagent context. In particular, we define a simple model of cooperative ability, give a definition of rational plans, and show how an agents intentions play a role in determining her next actions. This bridges the gap between intentions to achieve a goal and intentions to act. We also show that in the absence of interference, an agent that is able to achieve a goal, intends to do so, and is acting rationally will eventually achieve it.

Temporal Relevance in Dynamic Decision Networks with Sparse Evidence

Dynamic decision networks have been used in many applications and they are particularly suited for monitoring applications. However, the networks tend to grow very large resulting in significant performance degradation. In this paper, we study the degeneration of relevance of uncertain temporal information and propose an analytical upper bound for the relevance time of information in a restricted class of dynamic decision networks with sparse evidence. An empirical generalization of this analytical result is presented along with a series of experimental results to verify the performance of the empirical upper bound. By discarding irrelevant and weakly relevant evidence, the performance of the network is significantly improved.

Modeling and Reasoning with Decision-Theoretic Goals

Goal models have found important applications in Requirements Engineering as models that relate stakeholder requirements with system or human tasks needed to fulfill them. Often, such task specifications constitute rather idealized plans for requirements fulfillment, where task execution always succeeds. In reality, however, there is always uncertainty as to whether a specification can/will actually be executed as planned. In this paper, we introduce the concept of decision-theoretic goals in order to represent and reason about both uncertainty and preferential utility in goal models. Thus, goal models are extended to express probabilistic effects of actions and also capture the utility of each effect with respect to stakeholder priorities. Further, using a state-of-the-art reasoning tool, analysts can find optimal courses of actions/plans for fulfilling stakeholder goals while investigating the risks of those plans. The technique is applied in a real-world meeting scheduling problem, as well as the London Ambulance Service case study.

Prioritized goals and subgoals in a logical account of goal change – a preliminary report

Most previous logical accounts of goal change do not deal with prioritized goals and do not handle subgoals and their dynamics properly. Many are restricted to achievement goals. In this paper, we develop a logical account of goal change that addresses these deficiencies. In our account, we do not drop lower priority goals permanently when they become inconsistent with other goals and the agents knowledge; rather, we make such goals inactive. We ensure that the agents chosen goals/intentions are consistent with each other and the agents knowledge. When the world changes, the agent recomputes her chosen goals and some inactive goals may become active again. This ensures that our agent maximizes her utility. We also propose an approach for handling subgoals and their dynamics. We prove that the proposed account has some intuitively desirable properties.

Hierarchical grouping of association rules and its application to a real-world domain

One common problem in association rule mining is that often a very large number of rules are generated from the database. The sheer volume of these rules makes it difficult, if not impossible, for human users to analyze and make use of the rules. In this article, we propose two algorithms for grouping and summarizing association rules. The first algorithm recursively groups rules according to the structure of the rules and generates a tree of clusters as a result. The second algorithm groups the rules according to the semantic distance between the rules by making use of a semantic tree-structured network of items. We propose an algorithm for automatically tagging the semantic network so that the rules can be represented as directed line segments in a two-dimensional space and can then be grouped according to the distance between line segments. We also present an application of the two algorithms, in which the proposed algorithms are evaluated. The results show that our grouping methods are effective and produce good grouping results.

A model of rational agency for communicating agents

The Cognitive Agent Specification Language (CASL) is a framework for specifying and verifying complex communicating multiagent systems. In this paper, we extend CASL to incorporate a formal model of means-ends reasoning suitable for a multiagent context. In particular, we define a simple model of cooperative ability, give a definition of rational plans, and show how an agents intentions play a role in determining her next actions. This bridges the gap between intentions to achieve a goal and intentions to act. We also define a notion of subjective plan execution and show that in the absence of interference, an agent that is able to achieve a goal, intends to do so, and is acting rationally will eventually achieve it.

Logical foundations for a rational BDI agent programming language

To provide efficiency, current BDI agent programming languages with declarative goals only support a limited form of rationality --- they ignore other concurrent intentions of the agent when selecting plans, and as a consequence, the selected plans may be inconsistent with these intentions. In this paper, we develop logical foundations for a rational BDI agent programming framework with prioritized declarative goals that addresses this deficiency. We ensure that the agents chosen declarative goals and adopted plans are consistent with each other and with the agents knowledge. We show how agents specified in our language satisfy some key rationality requirements.