ArticlePDF Available

Examining a case-based reasoning system using mental models as a framework

Authors:
  • Simmons University

Abstract

Research examining how people use case-based reasoning retrieval systems has been limited. It has not been clear how mental models can be used as a framework to examine the usability of case-based reasoning retrieval systems. In this paper, we report our findings on how the presence of a mental model affects the experiences of users of the knowledge innovation for technology in education (KITE) system, the first developed CBR retrieval system in the educational technology integration community.
... The CBR module was created by using a CBR generator developed by the University of Missouri ( Shyu, 2003). The CBR generator uses a near-neighbor algorithm (He et al, 2010) to match a query-case with existing cases in the database. ...
... ). CBR is especially suitable for cost estimation in complex domains(Bisio & Malabocchia, 1995;Duan, Street & Xu, 2011;He, Erdelez & Wang, 2010;Moore, Erdelez & He, 2006;Prietula, Vicinanza & Mukhopadhyay, 1996; ...
Article
Currently, a work breakdown structure (WBS) approach is used as the most common cost estimation approach for online course production projects. To improve the practice of cost estimation, this paper proposes a novel framework to estimate the cost for online course production projects using a case-based reasoning (CBR) technique and a WBS. A prototype was developed to illustrate the effectiveness of the framework. Preliminary results suggest that the combined approach of using CBR and WBS is preferred and enables project managers to generate reasonable estimates for online course production projects more efficiently.
... As an artificial intelligence (AI) approach, case-based reasoning (CBR) has been successfully used to build a wide range of knowledge-based systems [33,53,23,4,15,35,54]. In particular, many CBR systems (with hundreds to tens of thousands of cases) have been developed to support problem solving, decision making and business intelligence in various fields such as health sciences, engineering, law, business and education [11,29,20,4,10,18,21,30,38,49]. ...
Article
Full-text available
The design and implementation of case-based reasoning (CBR) applications is time-consuming. To facilitate the development of CBR applications in various problem domains, the CBR community has created a number of CBR shells and software frameworks in the past twenty years. This paper provides a review of the state-of-the-art of CBR shells and software frameworks, highlights why the integration of Web 2.0 and CBR development tools is useful, and gives an example as to how we implement such integration. We use this example to illustrate how Web 2.0 features such as blogging functions can be integrated in a CBR system. Design recommendations and insights for implementing a Web 2.0-based CBR shell are also provided.
... Many case-based reasoning (CBR) systems have been developed in the past two decades. Recently, there has been an increased interest on the application of CBR to the health sciences, business and educational fields (Begum, Ahmed, Funk, Xiong, & Folke, 2011;Bridge, & Healy, 2012;He, Erdelez, & Wang, 2010;Huang, Chen, & Lee, 2007;Li, & Sun, 2008;Wang & Yang, 2012). While there is a growing body of literature devoted to CBR, much of this literature focuses on implementation of functional capabilities and the development of new techniques, models and algorithms (Ahn & Kim, 2009;Begum, Ahmed, Funk, Xiong, & Sch eele, 2009;Begum et al., 2011;Chang, 2005;Li, & Ho, 2009). ...
Article
Many CBR systems have been developed in the past. However, currently many CBR systems are facing a sustainability issue such as outdated cases and stagnant case growth. Some CBR systems have fallen into disuse due to the lack of new cases, case update, user participation and user engagement. To encourage the use of CBR systems and give users better experience, CBR system developers need to come up with new ways to add new features and values to the CBR systems. The author proposes a framework to use text mining and Web 2.0 technologies to improve and enhance CBR systems for providing better user experience. Two case studies were conducted to evaluate the usefulness of text mining techniques and Web 2.0 technologies for enhancing a large scale CBR system. The results suggest that text mining and Web 2.0 are promising ways to bring additional values to CBR and they should be incorporated into the CBR design and development process for the benefit of CBR users.
Article
This article reports on a longitudinal analysis of query logs of a web-based case library system during an 8-year period (from 2005 to 2012). The analysis studies 3 different information-seeking approaches: keyword searching, browsing, and case-based reasoning (CBR) searching provided by the system by examining the query logs that stretch over 8 years. The longitudinal dimension of this study offers unique possibilities to see how users used the 3 different approaches over time. Various user information-seeking patterns and trends are identified through the query usage pattern analysis and session analysis. The study identified different user groups and found that a majority of the users tend to stick to their favorite information-seeking approach to meet their immediate information needs and do not seem to care whether alternative search options will offer greater benefits. The study also found that return users used CBR searching much more frequently than 1-time users and tend to use more query terms to look for information than 1-time users.
Article
Full-text available
Molecular electronics describes a field that seeks to implement electronic components made of molecular building blocks. To date, few studies have used conjugated polymers in molecular junctions despite the fact that they potentially transport charge more efficiently than the extensively investigated small-molecular systems. Here we report a novel type of molecular tunnelling junction exploring the use of conjugated polymers, which are self-assembled into ultrathin films in a distinguishable 'planar' manner from the traditional vertically oriented small-molecule monolayers. Electrical measurements on the junctions reveal molecular-specific characteristics of the polymeric molecules in comparison with less conjugated small molecules. More significantly, we decorate redox-active functionality into polymeric backbones, demonstrating a key role of redox centre in the modulation of charge transport behaviour via energy level engineering and external stimuli, and implying the potential of employing tailor-made polymeric components as alternatives to small molecules for future molecular-scale electronics.
Article
Full-text available
We report the use of bilayer graphene as an atomically smooth contact for nanoscale devices. A two-terminal bucky-ball (C60)-based molecular memory is fabricated with bilayer graphene as a contact on the polycrystalline nickel electrode. Graphene provides an atomically smooth covering over an otherwise rough metal surface. The use of graphene additionally prohibits the electromigration of nickel into the C60 layer. The devices exhibit a low-resistance state in the first sweep cycle and irreversibly switch to a high-resistance state at 0.8 to 1.2 V bias. In the subsequent cycles, the devices retain the high-resistance state, thus making it write-once read-many memory.
Article
Case-based reasoning (CBR), as a promising technology for problem solving and decision support, has drawn considerable attention during the last 20years. As CBR systems become more frequently deployed in real-world situations and as large-scale case-bases become more commonly used in practice, the development and maintenance of the case-base becomes critical to CBR practitioners. In reality, adding cases to a case-base and updating cases in a case-base can be troublesome and time-consuming processes. It has become increasingly important for CBR practitioners to be able to implement an efficient way to develop and maintain the case base. However, techniques for case-base development and maintenance (such as adding cases and updating cases) have not received enough attention and are often neglected by CBR researchers. This paper discusses Wikis and XML (specifically, the Office Open XML format) and proposes an integrated approach to facilitate case-base development and maintenance in adding cases and in updating cases.
Article
Full-text available
This paper reportes a study that investigated the effects of conceptual description and search practice on users’ mental models and information seeking in a case-based reasoning retrieval (CBR) system with a best match search mechanism. This study also found examined how the presence of a mental model affects the users’ search performance and satisfaction in this system. The results of this study revealed that the conceptual description and search practice treatments do not have significantly different effects on the types of user’s mental models, search correctness, and search satisfaction. However, the search practice group spent significantly less time than the conceptual description group in finding the results. Qualitative analysis for the subjects’ post mental models revealed that subjects in the conceptual description group seem to have more complete mental models of the best match system than those in the search practice group. This study also that subjects with the best match mental models have significantly higher search correctness and search result satisfaction than subjects without the best match mental models. However, the best match mental models do not guarantee less search time in finding the results. This study did not find a significant correlation among search time, search correctness and search satisfaction. The study concludes with suggestions for future research and implications for system developers who are interested in CBR retrieval systems.