Multiple Classification Ripple Down Rules Research Articles

Intelligent conversation system using multiple classification ripple down rules and conversational context

We introduce an extension to Multiple Classification Ripple Down Rules (MCRDR), called Contextual MCRDR (C-MCRDR). We apply C-MCRDR knowledge-base systems (KBS) to the Textual Question Answering (TQA) and Natural Language Interface to Databases (NLIDB) paradigms in restricted domains as a type of spoken dialog system (SDS) or conversational agent (CA). C-MCRDR implicitly maintains topical conversational context, and intra-dialog context is retained allowing explicit referencing in KB rule conditions and classifications. To facilitate NLIDB, post-inference C-MCRDR classifications can include generic query referencing – query specificity is achieved by the binding of pre-identified context. In contrast to other scripted, or syntactically complex systems, the KB of the live system can easily be maintained courtesy of the RDR knowledge engineering approach. For evaluation, we applied this system to a pedagogical domain that uses a production database for the generation of offline course-related documents. Our system complemented the domain by providing a spoken or textual question-answering alternative for undergraduates based on the same production database. The developed system incorporates a speech-enabled chatbot interface via Automatic Speech Recognition (ASR) and experimental results from a live, integrated feedback rating system showed significant user acceptance, indicating the approach is promising, feasible and further work is warranted. Evaluation of the prototype’s viability found the system responded appropriately for 80.3% of participant requests in the tested domain, and it responded inappropriately for 19.7% of requests due to incorrect dialog classifications (4.4%) or out of scope requests (15.3%). Although the semantic range of the evaluated domain was relatively shallow, we conjecture that the developed system is readily adoptable as a CA NLIDB tool in other more semantically-rich domains and it shows promise in single or multi-domain environments.

Exploring a role for MCRDR in enhancing telehealth diagnostics

In-home telehealth devices are becoming increasingly popular when it comes to supporting the health management of home-based patients. With the devices capable of highly active monitoring, using sensors which produce large amounts of data, the deployment of telehealth devices into the home highlights the need for improved ways to collate, classify and dynamically interpret data safely and effectively. For clinicians working at a distance, the amounts of data generated by all in-home patient telematics devices poses questions on how best to intelligently filter, analyze and interpret this data to make diagnoses and respond to changes in patient conditions. In order to manage this issue, expert systems, applied for decades in other health fields, might play a role. In this paper, we explore how one type of expert system, Multiple Classification Ripple Down Rules (MCRDR), might address the issues. This paper begins by reviewing the capabilities of expert systems. Specifically, MCRDR is reviewed and its integration with an example telehealth device, MediStation, is explored. The range of potential benefits which might accrue when MCRDR and the MediStation are linked is identified as are some research and development challenges. Moving forwards, a simple simulator is introduced as one approach which is shown to be effective at exploring this exciting area of research. This paper takes the first steps towards introducing expert systems into the uHealth field and presents a simulator for this purpose.

An investigation into drug-related problems identifiable by commercial

Accredited pharmacists conduct home medicines reviews (HMRs) to detect and resolve potential drug-related problems (DRPs). A commercial expert system, Medscope Review Mentor (MRM), has been developed to assist pharmacists in the detection and resolution of potential DRPs. This study compares types of DRPs identified with the commercial system which uses multiple classification ripple down rules (MCRDR) with the findings of pharmacists. HMR data from 570 reviews collected from accredited pharmacists was entered into MRM and the DRPs were identified. A list of themes describing the main concept of each DRP identified by MRM was developed to allow comparison with pharmacists. Theme types, frequencies, similarity and dissimilarity were explored. The expert system was capable of detecting a wide range of potential DRPs: 2854 themes; compared to pharmacists: 1680 themes. The system identified the same problems as pharmacists in many patient cases. Ninety of 119 types of themes identifiable by pharmacists were also identifiable by software. MRM could identify the same problems in the same patients as pharmacists for 389 problems, resulting in a low overlap of similarity with an averaged Jaccard Index of 0.09. MRM found significantly more potential DRPs than pharmacists. MRM identified a wide scope of DRPs approaching the range of DRPs that were identified by pharmacists. Differences may be associated with system consistency and perhaps human oversight or human selective prioritisation. DRPs identified by the system were still considered relevant even though the system identified a larger number of problems.

Online knowledge validation with prudence analysis in a document management application

Prudence analysis (PA) is a relatively new, practical and highly innovative approach to solving the problem of brittleness in knowledge based system (KBS) development. PA is essentially an online validation approach where as each situation or case is presented to the KBS for inferencing the result is simultaneously validated. Therefore, instead of the system simply providing a conclusion, it also provides a warning when the validation fails. Previous studies have shown that a modification to multiple classification ripple-down rules (MCRDR) referred to as rated MCRDR (RM) has been able to achieve strong and flexible results in simulated domains with artificial data sets. This paper presents a study into the effectiveness of RM in an eHealth document monitoring and classification domain using human expertise. Additionally, this paper also investigates what affect PA has when the KBS developer relied entirely on the warnings for maintenance. Results indicate that the system is surprisingly robust even when warning accuracy is allowed to drop quite low. This study of a previously little touched area provides a strong indication of the potential for future knowledge based system development.

An incremental knowledge acquisition-based system for critical domains

In some real life situations, humans have to make decisions in critical environments. When a human analyzes a situation, (s)he has to decide whether there is a risky situation and what actions have to be performed. It is always desirable to detect these situations in advance, because the solution could be easier and the expert would have more time to make the best decision. An intelligent system may analyze the information, extract conclusions, format and order the causes leading to the severe condition, so becoming the decision-making process less dramatic. Multiple Classification Ripple Down Rules (MCRDR) are a successful intelligent classification technique which has proven its efficiency in several application domains, but it has some limitations to define complex situations. In this work, an extension to MCRDR to cover with complex domains is proposed. The validation of this methodological extension has been done through the development of a prototype for complex medical domains and this is also presented in this paper.

Development of an intelligent decision support system for medication review

The aim was to develop and evaluate a pilot version of a knowledge-based system that can identify existing and potential medication-related problems from patient information. This intelligent system could directly support pharmacists and other health professionals providing medication reviews. Rather than being based on static rules to trigger alerts, this system utilizes a multiple classification ripple-down rules approach, which allows the user to build rules incrementally and improve the accuracy of the knowledge base in identifying medication-related problems while the system is in use, with no outside assistance or training. The system contextualizes the potential drug therapy problems by taking into consideration the patient's demographics, and other medical condition and drugs. The system is capable of both being instructed in the domain of medication review through its routine use by an expert, and acting similarly to the expert when analysing genuine medication review cases. The system was handed over to an experienced clinical pharmacist (expert), with no knowledge or conclusions preloaded into the system. The expert was then able to add the case details and generate the rules required for 126 actual medication review cases. Over 250 rules were generated from the review cases, incorporating demographics, medical history, symptoms, medications and pathology results from these cases. At the completion of the cases, more than 80% of the potential medication-related problems identified by the expert were also detected by the system. The false positive rate, or number of incorrect medication-related problems identified by the system, was <10% overall and was zero for the last 15 cases analysed. The system found significantly more potential medication-related problems than the expert, with the system consistently remaining at least one finding ahead. There was a high incidence of missed potential medication-related problems by the expert, which were automatically repaired by the system. The knowledge-based system has already demonstrated that the technique employed is well suited to a domain of this nature and has furthermore demonstrated that it is capable of improving the quality of service that the medication reviewer can provide. The system will be further enhanced and tested prior to use in the field. It should help pharmacists in the provision of medication reviews, improving their clinical and time management capabilities, and enhancing their ability to contribute to the quality use of medications.

A knowledge-based approach to assign breast cancer treatments in oncology units

The aim of this work is to present a developed knowledge-based approach for breast cancer treatment. This Decision Support System (DSS) uses an incremental knowledge acquisition technique called Multiple Classification Ripple Down Rules (MCRDR) and a breast cancer treatments knowledge base. This system integrate MCRDR inference engine and use knowledge bases to reach advanced reasoning level during the decision making process. The knowledge base will be built from clinical protocols, and its content will be updated by the doctor using the MCRDR component. This DSS infers a cancer treatment from the (clinical or pathological) input data supplied by the physician and it also allows the maintenance of the knowledge base.

Knowledge-Based System Explanation: The Ripple-Down Rules Alternative

The ability to provide explanations has been seen as a key feature of expert systems (ES) typically not offered by other types of computer systems. ES need to offer explanations because of imprecise domains and the use of heuristics. Verification is not enough. ES need to justify and be accountable. Explanation is seen as an important activity for knowledge-based systems as it satisfies the user's need to decide whether to accept or reject a recommendation. In this paper we review explanation in first-generation and second-generation ES. An alternative is offered to the main approaches which uses multiple classification ripple-down rules and challenges even the goals of explanation. Instead of trying to give explanations which provide a meaningful line of reasoning and which are tailored to suit the individual it may be just as useful to provide the user with sufficient information and browsing tools to develop their own line of reasoning. The type of information that can assist understanding is the context in which the recommendation applies (which is provided through the display of relevant cases and exception rule history) and the ability to explore an abstraction hierarchy of the rules using formal concept analysis. An explanation toolkit aimed at putting the user in control is described and evaluated in this paper.

An easy-maintenance, reusable approach for building knowledge-based systems: application to landscape assessment

Multiple Classification Ripple-Down Rules (MCRDR) is an extended methodology which allows an expert to build and maintain a knowledge-based system for multiple classification without technical assistance. Current MCRDR knowledge bases do not exhibit an explicit model of the relationships for the domain terms used by the expert. This is a strong impediment for both reusing and sharing of MCRDR knowledge bases, as well as for rapid development and maintenance. In this work, we describe how a domain knowledge ontological framework can be integrated with MCRDR, so providing this with explicit reusable knowledge components.

Towards an expert system in ion-exclusion chromatography by means of multiple classification ripple-down rules

We describe the development and maintenance of an expert system to advise on the configuration of systems for ion-exclusion chromatography. The aim of the system is to define appropriate conditions for the separation of desired groups of acids or bases. The system is implemented in a rule-based system, multiple classification ripple-down rules, which offers multiple conclusions from rules based on the attributes of the system. In this case the attributes include physical and chemical properties of the solutes and the availability of instrumentation and accessories. With this information the method conditions can be defined for the detector, mobile-phase, whether suppression is to be used, and other ion-exclusion chromatography method conditions. A unique feature is that some conditions may be filled in by the program or be given by the user. Because of the nature of the “ripple-down rules” approach, in which new knowledge is always added as an amendment to an existing conclusion (and therefore cannot interfere with other conclusions), the expert or user can maintain and alter the system easily according to their own needs without the help of a software engineer. The system was developed and tested using cases from published papers on ion exclusion chromatography. For a set of 83 cases, although the expert system only agreed with the published conditions in 53% of cases, when the predictions were assessed by a recognized ion chromatography expert, 88% were pronounced “workable”.

Help desk system with intelligent interface

Automated help desk systems should retrieve exactly the information required to assist a user as quickly and as easily as possible be it for a lay user who knows little about the domain or for an advanced user who requires more specialized information. Automated help desk systems should also be easily maintainable, as knowledge in domains where help is required often changes very rapidly, for example, help for computer users. The aim of this study was to develop a help desk information retrieval mechanism suitable for a wide range of users and to provide a way of easily maintaining the system. The prototype developed for use over the World Wide Web combines keyword search and case-based reasoning to provide both rapid focusing on a part of the help information and guided interaction when the user is unclear about appropriate keywords. Ease of maintenance is provided by using multiple classification ripple down rules (MCRDR) to maintain the domain knowledge in the system. Further issues that arise include the problem of inappropriate focusing by keywords and maintenance in a distributed environment.