Knowledge Acquisition Strategies Research Articles

Using shared representations to improve coordination and intent inference

In groupware, users must communicate about their intentions and aintain common knowledge via communication channels that are explicitly designed into the system. Depending upon the task, generic communication tools like chat or a shared whiteboard may not be sufficient to support effective coordination. We have previously reported on a methodology that helps the designer develop task specific communication tools, called coordinating representations, for groupware systems. Coordinating representations lend structure and persistence to coordinating information. We have shown that coordinating representations are readily adopted by a user population, reduce coordination errors, and improve performance in a domain task. As we show in this article, coordinating representations present a unique opportunity to acquire user information in collaborative, user-adapted systems. Because coordinating representations support the exchange of coordinating information, they offer a window onto task and coordination-specific knowledge that is shared by users. Because they add structure to communication, the information that passes through them can be easily exploited by adaptive technology. This approach provides a simple technique for acquiring user knowledge in collaborative, user-adapted systems. We document our application of this approach to an existing groupware system. Several empirical results are provided. First, we show how information that is made available by a coordinating representation can be used to infer user intentions. We also show how this information can be used to mine free text chat for intent information, and show that this information further enhances intent inference. Empirical data shows that an automatic plan generation component, which is driven by information from a coordinating representation, reduces coordination errors and cognitive effort for its users. Finally, our methodology is summarized, and we present a framework for comparing our approach to other strategies for user knowledge acquisition in adaptive systems.

Effects of Knowledge Management Implementation in Hospitals

From the knowledge management point of view, the fundamental mission of hospital management is the delivery of quality medical services, utilizing highly specialized knowledge to solve healthcare problems within various resource constraints. Similar to other knowledge-intensive industries operating in highly challenging business environments, hospitals of all sizes must view the creation, organization, distribution, and application of knowledge as a critical aspect of their management activities. Knowledge management represents a viable strategy as hospitals strive simultaneously to provide quality medical services, improve operational efficiency, and conform to the government’s documentation and reporting regulations. This study examines the correlation as well as the causal relationships among knowledge characteristics, knowledge acquisition strategy, implementation measures, and performance of knowledge management implementations in the context of hospital management. Using primary data collected in Taiwan, our analyses show that the characteristics of knowledge affect the ways in which knowledge management is implemented, and the implementation measures, in turn, have a significant impact on the results of knowledge management implementations.

Integrated design of the intelligent web-based Chinese Medical Diagnostic System (CMDS) – Systematic development for digestive health

Abstract Chinese Medicine (CM) has been recognized as the popular complementary medicine in the world and the abundance of healing knowledge has been transmitted and enriched from generation to generation. The well-recorded archive of Chinese medical knowledge has grown and developed into very detailed theory and philosophy together with a comprehensive and subtle system of diagnosis. Recently, the Internet has been quite often used in a variety of applications for medical purpose, i.e. to provide real-time and platform-independent access which can build and share knowledge about clinical diagnosis experiences, knacks, diseases treatments, and even prescriptions. In this respect, there is a real need of an intelligent system for Chinese medical physician or education because Chinese medicine has evolved complex methods of diagnosis and treatment tailored to the individual’s subtle patterns of disharmony. This allows it not only to treat fully manifest diseases, but also to assist in maintaining health and balance to prevent illnesses from occurring. Ontologies become an important mechanism to build knowledge-based systems. The role of ontologies is to capture domain knowledge and provide a commonly agreed upon understanding of a domain. The advantages include the sharing and re-use of knowledge, and the better engineering of knowledge-based system with respect to acquisition, verification, and maintenance. In this paper, we propose a unifying framework for intelligent disease diagnosis system named CMDS – Chinese Medical Diagnostic System where the medical ontology has been integrated for the system development, and the methodologies of its implementation for digestive health. CMDS uses web interface and expert system technology to act as human expertise and diagnose a number of digestive system diseases. Besides, to efficiently elicit the knowledge of digestive system from domain experts and construct the medical ontology, a hybrid knowledge acquisition strategy is proposed. CMDS provides a truly precise analysis for digestive system disease and the prototype system can diagnose up to 50 types of diseases amongst 10 species of primary digestive system, and just uses over 500 rules and 600 images for various diseases. The satisfactory performance of the system has proven that it could serve the educational purpose, act as a consultant role and the functionality can be extended to the whole body health system.

See one, do one, teach one––exploring the core teaching beliefs of medical school faculty

This paper explores the core teaching beliefs of medical school faculty and establishes whether these beliefs differ among basic science, clinical, and instruction specialist faculty. One hundred and twenty-five medical school teachers who were members of professional organizations dedicated to the improvement of medical school teaching completed a Q-sort of 56 statements reflecting their core teaching beliefs. The statements described beliefs about motivation, knowledge and skill acquisition, retention, feedback, transfer, teacher characteristics, and teaching strategies. Q-sorts were completed by 37 basic scientists (30% of respondents), 59 clinicians (47%) and 29 instruction specialists (23%) working in medical schools. Fifty-two participants were classroom teachers (42%), 66 were classroom and clinical teachers (53%), and seven reported that they do not teach (6%). The Q-sort results indicate how medical school faculty members differ in their core beliefs about teaching and learning. Thirty-two respondents (26%) focused on the student as a person first. Eight (6%) were content oriented. Thirty-four (27%) were performance oriented; their focus was on having students learn and apply knowledge and skills to accomplish clinical tasks. Fifty-one respondents (41%) were found to have a blend of these viewpoints. Respondents’ type of training or type of teaching did not provide a reliable indication of core teaching beliefs classification.

Distant networking: The knowledge acquisition strategies of 'out-cluster' biotechnology firms

This paper addresses the conditions that enable new biotechnology firms (NBF) operating outside major biotechnology clusters, to obtain the resources and competences necessary to emerge and grow. Drawing on in-depth research on the structure, rationale and evolution of scientific and technological relationships of a group of Portuguese NBFs, the paper discusses the knowledge sourcing strategies devised by these firms and the type of factors that favour their adoption. NBFs are found to draw creatively from a diverse combination of local and distant technological relationships and to profit from a number of mechanisms that can reduce the impact of geographical distance on access to partners and on the transmission of knowledge. This behaviour is favoured by factors associated with the strength of the local science base in relevant fields and to the level of international mobility and exposure achieved by local scientists.

The influence of concept mapping on achievement, self-regulation, and self-efficacy in students of English as a second language

This study investigated the effectiveness of concept mapping used as a learning strategy with students in English as a Second Language classrooms. Seventy-nine ESL students participated in the study. Variables of interest were students’ achievement when learning from English-language text, students’ reported use of self-regulation strategies (self-monitoring and knowledge acquisition strategies), and students’ self-efficacy for learning from English-language text. A randomized pre-test–post-test control group design was employed. The findings showed a statistically significant interaction of time, method of instruction, and level of English proficiency for self-monitoring, self-efficacy, and achievement. For all four outcome variables, the concept mapping group showed significantly greater gains from pre-test to post-test than the individual study group. The findings have implications for both practice and research.

The role of information technology in the explication and crystallization of tacit healthcare knowledge

As a major contributor and user of healthcare knowledge, the healthcare enterprise requires intelligent knowledge management strategies to effectively exploit expert-quality healthcare knowledge. In this paper, we discuss a novel knowledge acquisition and representation strategy using healthcare scenarios. This strategy is complemented with an ontology- or thesaurus-based input standardization mechanism to limit the diverse nature of the healthcare experts’ input. We also present in detail an algorithm for healthcare knowledge crystallization that, in addition to simulating the process of annealing, employs the notion of nucleation and growth on healthcare knowledge. The crystallized knowledge potentially undergoes a repair process that employs analogical reasoning to identify less useful knowledge and transform them, with the aid of more established knowledge, into useful ones. We conclude by asserting that the strategies presented do provide an all-rounded approach in our efforts to bring quality healthcare knowledge and services to the masses.

Internal knowledge generation: the research laboratory and innovative productivity in the pharmaceutical industry

Abstract As firms face increasing pressures associated with new forms of knowledge-based competition, researchers have begun to focus on knowledge acquisition strategies to the point of ignoring knowledge creation strategies. This paper investigates the impact of research centers in the pharmaceutical industry upon this neglected form of knowledge creation–innovation productivity. We find that firms with separate research facilities are more innovative than firms without such facilities. The results diverge when locational distance of the laboratory is considered for patent versus drug innovations, and vary across diversification types. These findings suggest that managers and researchers need to consider the relationship between the research center and corporate headquarters — and the role of the corporate headquarters — when developing knowledge generation strategies.

Knowledge acquisition using VRML and the unified modelling language: the case of the RGB computer room

This paper describes the development of knowledge acquisition strategies for computing systems that aid designers in the construction of intelligent, heterogeneous, dynamic computational models of the built environment. The project uses a structure–behaviour–function model of the environment. The modelling strategy uses Virtual Reality Modelling Language (VRML) as a prototyping tool for constructing the built environment models. Unified Modelling Language (UML), a visual object oriented modelling language, is used in a reverse engineering mode as part of the knowledge acquisition strategy.

A strategy for developing practice guidelines for the ICU using automated knowledge acquisition techniques.

To implement practice guideline entry tools in a reminder system in order to provide decision support to health care workers in clinical care and emergency care environments. To design a knowledge acquisition environment that enables physicians to formulate, update, and verify guidelines without the assistance of a knowledge engineer. We developed a knowledge acquisition environment for the Intensive Care Unit (ICU) consisting of 1) a graphical knowledge acquisition tool, 2) tools that perform logical and semantic tests on proposed guidelines, 3) a Patient Data Management System (PDMS) containing clinical patient data, and 4) an expert system that reminds ICU health care workers of inconsistencies between a treatment plan and implemented guidelines. Physicians enter the guidelines using the knowledge acquisition tool, after which consistency and correctness tests are performed on the guidelines. The guidelines are then transferred to the knowledge base of the reminder system and validated by applying the new guidelines to a large stored data set of previous patients. If the new guidelines are approved, they are exported to the reminder system that is used in daily practice. ICU physicians used the knowledge acquisition tool to enter 58 guidelines into the reminder system's knowledge base. These guidelines were tested on a data set consisting of 803 previously admitted patients. As a result, 27 guidelines fired at least once, generating 406 reminders in total. Of the 406 generated reminders, 356 (88%) were issued correctly and 50 (12%) were false alarms. The reminders that were issued correctly involved 3 situations: 1) the database contained inconsistent or incomplete information, 2) the actions or decisions of the health care workers were not the most appropriate ones, and 3) there was a potential risk involved. All false alarms were caused by the fact that the corresponding guidelines were not specific enough to handle certain exceptions. As a result of this analysis, the guidelines could be improved in such a way as to eliminate all false alarms. These first results demonstrate that this bottom-up knowledge acquisition strategy, implemented by the automated knowledge acquisition tools, enables medical specialists to improve the quality of computer support in an ICU without assistance of a knowledge engineer.

Strategies for Knowledge Acquisition from Cartographic Maps by Blind and Visually Impaired Adults

Two experiment are reported both of which employed a research design originally used by Thorndyke and Stasz (1980) to elicit the spontaneous strategies of map readers, and to relate these to their skill in map reading. In Experiment 1, blind, visually impaired and sighted children explored a tactile or a print map while “thinking aloud”. The protocols obtained were analysed according to a set of strategy types. The children were also asked to produce a copy of the map to test knowledge of it. These were according to accuracy. Experiment 2 used the same basic copy procedure with blind, visually impaired and sighted adults, who explored a more complex town map and plan of a building. Sighted participants produced more accurate copies of the map, and used strategies that focused on global spatial relations on the map, while blind and visually impaired participants tended to focus on individual localised elements of the map. However, those blind and visually impaired participants who produced more accurate copies of the maps, tended also to adopt a more global focus.

Strategies for Knowledge Acquisition from Cartographic Maps by Blind and Visually Impaired Adults

Two experiment are reported both of which employed a research design originally used by Thorndyke and Stasz (1980) to elicit the spontaneous strategies of map readers, and to relate these to their skill in map reading. In Experiment 1, blind, visually impaired and sighted children explored a tactile or a print map while “thinking aloud”. The protocols obtained were analysed according to a set of strategy types. The children were also asked to produce a copy of the map to test knowledge of it. These were according to accuracy. Experiment 2 used the same basic copy procedure with blind, visually impaired and sighted adults, who explored a more complex town map and plan of a building. Sighted participants produced more accurate copies of the map, and used strategies that focused on global spatial relations on the map, while blind and visually impaired participants tended to focus on individual localised elements of the map. However, those blind and visually impaired participants who produced more accurate copies of the maps, tended also to adopt a more global focus.

SWXPert: A Prototype Expert System for Serviceware Selection

Solid waste management has become a major concern from the perspective of consumers, policy makers, and the food service industry. The purpose of this study was to develop and evaluate a prototype expert system that could aid food service professionals in the serviceware decision. Initial knowledge acquisition involved sending a questionnaire to 327 school food service professionals. The responses provided by 172 (52.8%) food service professionals were factor analyzed to identify which variables were most significant and the relative importance these variables had in their decision making process. Based on the results of this analysis, a prototype rule-based expert system (SWXpert) was created using VP-Expert.Thirty school district food service directors evaluated the decision reached by the prototype expert system (SWXpert) on 30 sample school districts. Overall, a 75.8% agreement rate was found between SWXpert and the school food service directors. This paper presents an alternative knowledge acquisition strategy that can be used for the development of the initial prototype expert system.

135 A knowledge acquisition and processing strategy based on formal concept analysis

135 A knowledge acquisition and processing strategy based on formal concept analysis

A Knowledge Acquisition and Processing Strategy Based on Formal Concept Analysis

Abstract A new method for knowledge acquisition and processing is developed by using the mathematical theory of formal concept analysis. First it is demonstrated how a decision table for a process control problem (distillation column) is obtained by a dialogue between the expert and the computer. Then the concept analysis approach is briefly outlined. This enables an optimization of conventional expert systems. However, more important for practical real time applications is the direct processing of the knowledge contained in a decision table by combining concept analysis with Shannon's entropy. Thus the resulting knowledge based system is optimal in an information theoretical view. It can be implemented in a procedural language and therefore simply be embedded into existing process control systems.

A knowledge acquisition and processing strategy based on formal concept analysis

A new method for knowledge acquisition and processing is developed by using the mathematical theory of formal concept analysis. First it is demonstrated how a decision table for a process control problem (distillation column) is obtained by a dialogue between the expert and the computer. Then the concept analysis approach is briefly outlined. This enables an optimization of conventional expert systems. However, more important for practical real time applications is the direct processing of the knowledge contained in a decision table by combining concept analysis with Shannon's entropy. Thus the resulting knowledge based system is optimal in an information theoretical view. It can be implemented in a procedural language and therefore simply be embedded into existing process control systems.

Knowledge Acquisition in Fertilization Expert Systems Building Tool

This paper presents an automated knowledge acquisition system which is used in an expert systems building tool to build expert systems for applying fertilizer. A problem-oriented knowledge acquisition strategy and a rule—group knowledge representation approach in the form of rule—skeleton + rule—body are presented. The tool for building fertilization expert systems is introduced briefly. Using this tool, the domain expert that knows little AI techniques can construct an expert system directly without the aids of knowledge engineers.

Selecting knowledge acquisition tools and strategies based on application characteristics

One of the most troublesome and time-consuming activities in constructing a knowledge-based system is the elicitation and modelling of knowledge from the human expert about the problem domain. A major obstacle is that little guidance is available to the domain expert or knowledge engineer to help with (1) classifying the application task and identifying a problem-solving method, and (2) given the application task characteristics, selecting knowledge acquisition tools and strategies to be applied in creating and refining the knowledge base. Our objective is to provide automated assistance to a knowledge engineer or domain expert in analysing the problem domain, classifying the problem tasks and sub-tasks, identifying problem-solving methods, proposing knowledge acquisition tools, and suggesting the use of specific strategies for knowledge acquisition provided in selected tools. We describe such an implementation in the Dialog Manager subsystem of the AQUINAS knowledge acquisition workbench. The Dialog Manager provides advice to potential AQUINAS users as well as continuing guidance to users who select AQUINAS for knowledge base development.

ViewPoint: a troubleshooting-specific knowledge acquisition tool

As knowledge bases grow in size and complexity, tools for extracting and encoding knowledge from the domain experts become critical for the success of an expert system project. Many of the available expert system tools are designed for a broad range of domains and therefore are often inadequate for building and maintaining domain-specific knowledge structures. This results from the fact that some structures may only exist implicitly or have semantics than cannot be interpreted by generic tools. On the other hand, a tool that is specific to just one application in a domain lacks useful generality. A tool should be flexible enough to be widely applicable within the domain for which it was designed. We describe our knowledge acquisition tool, ViewPoint, which achieves this flexibility by being build on top of a generic expert system architecture for the troubleshooting domain. ViewPoint allows (i) the handling of all knowledge structures present in a troubleshooting knowledge base, (ii) the viewing of many levels of detail, from the global overview of the knowledge base to the examination of a single object, and (iii) the ability to cope with the knowledge acquisition strategies used in the domain.

Machine Learning Strategies for Knowledge Acquisition in Autonomous Robot Systems

Machine learning has been studied for more than 20 years. Most of the recent research contributions are basic and come from the areas of Cognitive Science and Artificial Intelligence. The classical domain of machine learning is related to vision and image analysis and image description. Since about five years there has been encreased activity to apply and to extend these general learning methods to really complex systems such as autonomous robots. The purpose of applied machine learning methods is to facilitate and to aid automatic construction and generation of knowledge based systems and to increase the system performance with time. Autonomous robot systems will operate in an unknown real environment and are based on the interaction of the basic system components such as action planner, plan executive and action monitor which need knowledge about the real world and knowledge about how to behave to reach a given goal. A robot system can learn if it is able to make changes and extensions in its behavioural control structure that enables it to better perform a given task. This includes acquisition of declarative procedural knowledge, the developement of manipulative and cognitive skills through instruction and practice as well as the organisation of knowledge into efficient representations. The purpose of automatic self-configurating or self-modifying representations is to make experience based knowledge available for possible future use in the context of increasing robot system performance. This paper reviews some general methods of machine learning application and some first results achieved to date.