Monika Schubert

An efficient diagnosis algorithm for inconsistent constraint sets

Abstract Constraint sets can become inconsistent in different contexts. For example, during a configuration session the set of customer requirements can become inconsistent with the configuration knowledge base. Another example is the engineering phase of a configuration knowledge base where the underlying constraints can become inconsistent with a set of test cases. In such situations we are in the need of techniques that support the identification of minimal sets of faulty constraints that have to be deleted in order to restore consistency. In this paper we introduce a divide and conquer-based diagnosis algorithm (FastDiag) that identifies minimal sets of faulty constraints in an overconstrained problem. This algorithm is specifically applicable in scenarios where the efficient identification of leading (preferred) diagnoses is crucial. We compare the performance of FastDiag with the conflict-directed calculation of hitting sets and present an in-depth performance analysis that shows the advantages of our approach.

Consumer decision making in knowledge-based recommendation

In contrast to customers of bricks and mortar stores, users of online selling environments are not supported by human sales experts. In such situations recommender applications help to identify the products and/or services that fit the user’s wishes and needs. In order to successfully apply recommendation technologies we have to develop an in-depth understanding of decision strategies of users. These decision strategies are explained in different models of human decision making. In this paper we provide an overview of selected models and discuss their importance for recommender system development. Furthermore, we provide an outlook on future research issues.

Recommendation and decision technologies for requirements engineering

Requirements engineering (RE) is considered as one of the most critical phases in the software life-cycle, and poorly implemented RE processes are among the major risks for project failure. Stakeholders are often faced with the challenge that the complexity of information outstrips their capability to survey it and to decide about which requirements should be taken into account. Additionally, preferences regarding a set of requirements are typically not known beforehand but constructed within the scope of a decision making process. In this paper we introduce a simple application scenario and discuss recommendation and decision technologies which can be exploited for proactively supporting stakeholders in their decision making.

Personalized user interfaces for product configuration

Configuration technologies are well established as a foundation of mass customization which is a production paradigm that supports the manufacturing of highly-variant products under pricing conditions similar to mass production. A side-effect of the high diversity of products offered by a configurator is that the complexity of the alternatives may outstrip a users capability to explore them and make a buying decision. In order to improve the quality of configuration processes, we combine knowledge-based configuration with collaborative and content-based recommendation algorithms. In this paper we present configuration techniques that recommend personalized default values to users. Results of an empirical study show improvements in terms of, for example, user satisfaction or the quality of the configuration process.

Personalized diagnoses for inconsistent user requirements

Abstract Knowledge-based configurators are supporting configuration tasks for complex products such as telecommunication systems, computers, or financial services. Product configurations have to fulfill the requirements articulated by the user and the constraints contained in the configuration knowledge base. If the user requirements are inconsistent with the constraints in the configuration knowledge base, users have to be supported in finding out a way from the no solution could be found dilemma. In this paper we introduce a new algorithm (PersDiag) that allows the determination of personalized diagnoses for inconsistent user requirements in knowledge-based configuration scenarios. We present the results of an empirical study that show the advantages of our approach in terms of prediction quality and efficiency.

FastXplain: conflict detection for constraint-based recommendation problems

Constraint-based recommender systems support users in the identification of interesting items from large and potentially complex assortments. Within the scope of such a preference construction process, users are repeatedly defining and revising their requirements. As a consequence situations occur where none of the items completely fulfills the set of requirements and the question has to be answered which is the minimal set of requirements that has to be changed in order to be able to find a recommendation. The identification of such minimal sets relies heavily on the identification of minimal conflict sets. Existing conflict detection algorithms are not exploiting the basic structural properties of constraint-based recommendation problems. In this paper we introduce the FastXplain conflict detection algorithm which shows a significantly better performance compared to existing conflict detection algorithms. In order to demonstrate the applicability of our algorithm we report the results of a corresponding performance evaluation.

Empirical knowledge engineering: cognitive aspects in the development of constraint-based recommenders

Constraint-based recommender applications provide valuable support in item selection processes related to complex products and services. This type of recommender operates on a knowledge base that contains a deep model of the product assortment as well as constraints representing the companys marketing and sales rules. Due to changes in the product assortment as well as in marketing and sales rules, such knowledge bases have to be adapted very quickly and frequently. In this paper we focus on a specific but very important aspect of recommender knowledge base development: we analyze the impact of different constraint representations on the cognitive effort of a knowledge engineer to successfully complete certain knowledge acquisition tasks. In this context, we report results of an initial empirical study and provide first basic recommendations regarding the design of recommender knowledge bases.

BFX: DIAGNOSING CONFLICTING REQUIREMENTS IN CONSTRAINT-BASED RECOMMENDATION

When interacting with constraint-based recommender applications, users describe their preferences with the goal of identifying the products that fit their wishes and needs. In such a scenario, users are repeatedly adapting and changing their requirements. As a consequence, situations occur where none of the products completely fulfils the given set of requirements and users need a support in terms of an indicator of minimal sets of requirements that need to be changed in order to be able to find a recommendation. The identification of such minimal sets relies heavily on the existence of (minimal) conflict sets. In this paper we introduce BFX (Boosted FastXplain), a conflict detection algorithm which exploits the basic structural properties of constraint-based recommendation problems. BFX shows a significantly better performance compared to existing conflict detection algorithms. In order to demonstrate the performance of BFX, we report the results of a comparative performance evaluation.

Adaptive utility-ased recommendation

Knowledge-based recommenders support customers in preference construction processes related to complex products and services. In this context, utility constraints (scoring rules) play an important role. They determine the order in which items (products and services) are presented to customers. In many cases utility constraints are faulty, i.e., calculate rankings which are not expected and accepted by marketing and sales experts. The adaptation of these constraints is extremely time-consuming and often an error-prone process. In this paper we present an approach which effectively supports the automated adaptation of utility constraint sets based on solutions for corresponding nonlinear optimization problems. This approach significantly increases the applicability of knowledge-based recommendation by allowing the automated reproduction of item rankings specified by marketing and sales experts.

Solving Over-constrained Problems Using Network Analysis

Requirements for which no recommendation can be calculated are unsatisfactory for the user. The detection and resolution of conflicts between those requirements and the product assortment is an important functionality to successfully guide the user to a solution. In this paper we introduce a new approach how to identify minimal conflict sets in over constrained problems through network analysis. Conflict sets offer the information which constraints (requirements) need to be changed to retrieve a solution. Random constrained problems are used to evaluate our approach and compare it to existing conflict detection algorithms. A major result of this evaluation is that our approach is superior in settings typical for knowledge-based recommendation problems.