Plans, goals and selection rules in the comprehension of computer programs

The notion of the programming plan has been proposed as a mechanism through which one can explain the nature of expertise in programming. Soloway and Ehrlich (1984) suggest that such expertise is characterized by the existence and use of programming plans. However, studies in other complex problem-solving domains, notably text editing, suggest that expertise is characterized not only by the possession of plan-related structures but also by the development of appropriate selection rules which govern the implementation of plans in appropriate situations (Card et al. 1980, Kay and Black 1984, 1986). This paper presents an experimental study which examines the role of programming plans in the context of skill development in programming. The results of this study suggest that plan-based structures cannot be used in isolation to explain novice/expert differences. Indeed, such structures appear to prevail at intermediate levels of skill. The major characteristic of expertise in programming would appear to be strongly related to the development of appropriate selection rules and to so-called program discourse rules. This in turn suggests that current views on the role of plan-based structures in expert programming performance are too limited in their conception to provide an adequate basis for a thorough analysis of the problem-solving activity in the programming domain.     

基于派生谓词的STRIPS领域知识提取策略

领域知识的获取是智能规划研究中的重要内容之一.派生规则是一种基于逻辑推理的领域知识表示方法.在对动作模型和派生规则综合分析的基础上提出了基于派生谓词的STRIPS领域知识提取策略,并给出了该提取策略的算法描述.在规划求解过程中,利用提取所得的领域规则可减少派生规则的逻辑推导,从而提高规划效率.对任意一个规划领域,利用提...     

Thread-based probabilistic models for expert finding in enterprise Microblogs

Within the enterprise important conversations are shifting to Social Media Microblogs thus providing a new source for the acquisition of tacit knowledge useful for intelligent agents. While automated retrieval of candidate experts from Web documents has been well-researched, little exists that leverages social aspects of Microblogs for this purpose. We show that the experts and their related expertise can also be identified in the enterprise corpus. Analysis reveals the existence of problem-solving threads where the requestor seeking help often elicits responses from experts which consequently records tacit (experiential) knowledge. Here we present rules to automate the acquisition of this tacit knowledge. The rules are based on probabilistic models enhanced with linguistics to exploit the role patterns in threads. Heuristics also strengthen local evidence about associations between candidates, documents and (expertise) terms. Applying this, we demonstrate that the Enhanced Models can ameliorate the negative impact of Microblogs (such as sparse data, short content, and implicit associations). The experiments show that the candidate models significantly outperform the document models in the Microblog environment. This is different from previous research in Web environments. Examples also illustrate underlying insights and emphasize the ‘additive’ nature of expertise found in Social Media.     

A quantitative analysis of pattern production and its relationship to expert performance

Abstract. The modelling and measurement of expertise is a relatively new research area in artificial intelligence and cognitive science. Many domains do not have a formal method for evaluating expertise. When formal methods exist, they are frequently inefficient. Using an extension to the IAM program, a pattern recognition and acquisition method for evaluating the level of expertise for the domain of chess is developed. Chess players, as well as experts in other domains, use cognitive chunks of perceptual patterns to gain a cognitive economy that enables them to evaluate complex domain situations faster and more accurately than novices. The IAM program acquires a representative collection of the perceptual patterns demonstrated by a domain expert and uses those patterns to analyse skill level. A longitudinal study of a developing player and a comparison of the developing player to an established expert demonstrates the utility of the developed method for evaluating expertise.     

On the learning of rule uncertainties and their integration into probabilistic knowledge bases

We present a natural and realistic knowledge acquisition and processing scenario. In the first phase a domain expert identifies deduction rules that he thinks are good indicators of whether a specific target concept is likely to occur. In a second knowledge acquisition phase, a learning algorithm automatically adjusts, corrects and optimizes the deterministic rule hypothesis given by the domain expert by selecting an appropriate subset of the rule hypothesis and by attaching uncertainties to them. Then, in the running phase of the knowledge base we can arbitrarily combine the learned uncertainties of the rules with uncertain factual information.Formally, we introduce the natural class of disjunctive probabilistic concepts and prove that this class is efficiently distribution-free learnable. The distribution-free learning model of probabilistic concepts was introduced by Kearns and Schapire and generalizes Valiant's probably approximately correct learning model. We show how to simulate the learned concepts in probabilistic knowledge bases which satisfy the laws of axiomatic probability theory. Finally, we combine the rule uncertainties with uncertain facts and prove the correctness of the combination under an independence assumption.     

RULES-IT: incremental transfer learning with RULES family

In today’s world of excessive development in technologies, sustainability and adaptability of computer applications is a challenge, and future prediction became significant. Therefore, strong artificial intelligence (AI) became important and, thus, statistical machine learning (ML) methods were applied to serve it. These methods are very difficult to understand, and they predict the future without showing how. However, understanding of how machines make their decision is also important, especially in information system domain. Consequently, incremental covering algorithms (CA) can be used to produce simple rules to make difficult decisions. Nevertheless, even though using simple CA as the base of strong AI agent would be a novel idea but doing so with themethods available in CA is not possible. It was found that having to accurately update the discovered rules based on new information in CA is a challenge and needs extra attention. In specific, incomplete data with missing classes is inappropriately considered, whereby the speed and data size was also a concern, and future none existing classes were neglected. Consequently, this paper will introduce a novel algorithm called RULES-IT, in order to solve the problems of incremental CA and introduce it into strong AI. This algorithm is the first incremental algorithm in its family, and CA as a whole, that transfer rules of different domains to improve the performance, generalize the induction, take advantage of past experience in different domain, and make the learner more intelligent. It is also the first to introduce intelligent aspects into incremental CA, including consciousness, subjective emotions, awareness, and adjustment. Furthermore, all decisions made can be understood due to the simple representation of repository as rules. Finally, RULES-IT performance will be benchmarked with six different methods and compared with its predecessors to see the effect of transferring rules in the learning process, and to prove how RULES-IT actually solved the shortcoming of current incremental CA in addition to its improvement in the total performance.     

Automated knowledge acquisition for design and manufacturing: The case of micromachined atomizer

In order to remain competitive in the global market, original equipment manufacturers (OEMs) are developing a process-based, knowledge-driven product development environment with emphasis on the acquisition, storing, and utilization of manufacturing knowledge. This is usually achieved by using the symbolic artificial intelligence (AI) approach. Specifically, knowledge-based expert systems are developed to capture human expertise, mostly in terms of IF–THEN production rules. It has been recognized that the development of symbolic knowledge-based expert systems suffers from the so-called knowledge acquisition bottleneck. Knowledge acquisition is the process of collecting domain knowledge and transforming the knowledge into a computerized representation. It is a challenging and time-consuming process due to the difficulties involved in eliciting knowledge from human experts. This paper presents an automated approach for knowledge acquisition by integrating neural networks learning ability and fuzzy logics structured knowledge representation. Using this approach, knowledge is automatically acquired from data and represented using humanly intelligible fuzzy rules. The approach is applied to a case study of the design and manufacturing of micromachined atomizers for gas turbine engine. The influence of geometric features on the performance of the atomizers is investigated. The results are then compared with those obtained using traditional regression analysis approach (abstract mathematical models). It was found that the automated approach provides an efficient means for knowledge acquisition. Since the fuzzy rules extracted are easy to understand, they can be used to allow more clear specification of manufacturing processes and to shorten learning curves for novice manufacturing engineers.     

SEDAF: AN INTELLIGENT EDUCATIONAL SYSTEM FOR MATHEMATICS

The field of artificial intelligence and education, in which AI techniques and methodologies are used to build sophisticated intelligent educational systems, is developing rapidly. In this paper we present an intelligent educational system for teaching high school and college students how to analyze and draw graphs of mathematical functions. The system, named SEDAF, has been developed in a knowledge engineering environment and runs on a Lisp-machine workstation. We illustrate the various modules constituting SEDAF: the user interface; an expert module, capable of solving problems in the subject domain; a diagnosis module, which points out possible reasons for students' errors; a student modeling module, capable of building an explicit representation of the learning status of the student; and a remedial subsystem, called a therapy module, constituted by means-ends tutorial rules that execute teaching actions on the base of the status of the student model. The goal of the presentation is to stress the innovative aspects of the architecture of SEDAF, in particular the use of metalevel knowledge to embed in the system the teaching expertise that allows the system to personalize its behavior to the specific student and to pursue a didactic plan.     

基于CLIPS的卫星任务规划专家系统设计

本文提出了基于CLIPS的卫星任务规划专家系统的设计方法,详细分析了系统的结构和功能,重点讨论了中文产生式系统的BNF范式、基于上下文的推理机制和集合运算符。中文产生式系统的BNF范式基于CLIPS标准BNF范式定义,并依据BNF范式进行规则表示和规则自定义获取;推理机采用上下文限制的规则控制策略,依据不同的上下文加载相关的事实和规则,提高推理机的运行效率;利用规则中的对象逻辑子式进行了集合运算符的设计,并对极值运算符、属性差值运算符和均值运算符等三类集合运算符进行了探讨。该系统解决了卫星任务规划中知识表示和知识获取问题,提高了卫星任务规划推理效率,为卫星任务规划人员提供有效的辅助决策功能。     

Uses of accelerometer data collected from a wearable system

This paper presents work, assessing the use of accelerometers in wearable systems for a number of applications. It discusses and demonstrates how body mounted accelerometers can be used in context aware computing systems and for measuring aspects of human performance, which may be used for teaching and demonstrating skill acquisition, coaching sporting activities, sports and human movement research, and teaching subjects such as physics and physical education. Analysis is restricted to considerations as to how raw data can be used, and how simple calculations of quantities of data in the time domain, can be used. The limitations of the use of such data are discussed.     

基于音视信息融合的桌面机械臂技能获取及控制系统

针对化学实验机械臂控制编程门槛高且技能获取准确率低等发展制约因素,设计一种基于音视信息融合算法的桌面型实验机械臂的控制系统,实验员通过边做边说的示教方式教给机械臂运动技能,进而代替实验员完成一些繁琐、具有危险性的实验工作。系统分为技能获取以及运动控制两部分。其技能获取部分使用改进的双流卷积网络实现动作检测;使用语音AI和正则表达式实现语音提取;再通过音视动作信息融合算法将动作检测和语音部分的识别信息相融合得出高重合度的运动技能,技能获取准确度可达81%以上。运动控制部分使用电机控制和抓取位姿识别,可实现更精细的控制和抓取。系统可用于具有特定流程化学实验的示教控制工作,在代替实验员来完成化学实验工作的同时大大降低了编程门槛,提高了效率。     

Development of an Internet-based intelligent design support system for rolling element bearings

This paper presents a novel approach to developing an intelligent agile design system for rolling bearings based on artificial intelligence (AI), Internet and Web technologies and expertise. The underlying philosophy of the approach is to use AI technology and Web-based design support systems as smart tools from which design customers can rapidly and responsively access the systems' built-in design expertise. The approach is described in detail with a novel AI model and system implementation issues. The major issues in implementing the approach are discussed with particular reference to using AI technologies, network programming, client-server technology and open computing of bearing design and manufacturing requirements.     

The Development of Human Expertise in a Complex Environment

We introduce an innovative technique that quantifies human expertise development in such a way that humans and artificial systems can be directly compared. Using this technique we are able to highlight certain fundamental difficulties associated with the learning of a complex task that humans are still exceptionally better at than their computer counterparts. We demonstrate that expertise goes through significant developmental transitions that have previously been predicted but never explicated. The first signals the onset of a steady increase in global awareness that begins surprisingly late in expertise acquisition. The second transition, reached by only a very few experts in the world, shows a major reorganisation of global contextual knowledge resulting in a relatively minor gain in skill. We are able to show that these empirical findings have consequences for our understanding of the way in which expertise acquisition may be modelled by learning in artificial intelligence systems. This point is emphasised with a novel theoretical result showing explicitly how our findings imply a non-trivial hurdle for learning for suitably complex tasks.     

Reconceptualizing knowledge networks for enterprise systems implementation: incorporating domain expertise of knowledge sources and knowledge flow intensity

Research on knowledge acquisition through informal social networks during enterprise system implementation has not accounted for the domain expertise of knowledge sources or the quality of knowledge flows. By using data collected from an enterprise resource planning system implementation, this paper reconceptualizes knowledge networks into subnetworks on the basis of the domain expertise of end users and analyzes knowledge acquisition patterns between subnetworks across workgroups having varying performance outcomes. Expertise-based knowledge patterns and their intensities had significant implications for performance outcomes, reiterating their role in the learning process and emphasizing the need to incorporate them into knowledge networking models.     

Trace-driven knowledge acquisition (TDKA) for rule-based real time scheduling systems

Expert scheduling systems, which develop the schedule automatically on a real time basis, are able to respond to the changes of product demand in Flexible Manufacturing Systems (FMS). While developing an expert scheduling system, the most time-consuming and difficult step is knowledge acquisition, the process that elicits the knowledge from experts and transfers it into the knowledge base. A trace-driven knowledge acquisition (TDKA) method is proposed to extract the expertise from the schedules produced by expert schedulers. Three phases are involved in the TDKA process: data collection, data analysis, and rule evaluation. In data collection, the expert schedulers are identified and decisions made during the scheduling process are recorded as a trace. In data analysis, a set of scheduling rules is developed based on the trace. The rules are then evaluated in the last phase. If the resulting rules do not perform as well as the expert schedulers, the process returns to phase two and refines the rules. The whole process stops whenever the resulting rules perform at least as well as the expert schedulers. A circuit board production line is used to demonstrate the feasibility of the TDKA methodology. The scheduling rules perform much better than the expert schedulers from whom the rules are extracted.     

Using conscious reflection,group processes,and AI to facilitate development of expertise

Three experiments investigated procedures derived from research on knowledge acquisition, group processes, and artificial intelligence for facilitating the development of expertise. In each experimental session, subjects learned to control a simulated sugar production factory. Then they formulated written policies for controlling sugar production either alone as individuals or in small groups. An adaptive AI system provided feedback on policy quality. The research also investigated the use of forced reflective practice in which learners attempted to predict what their policy would do while performing the task. The AI system provided feedback about what their policy would have done in each situation and the outcome based on their policy's response or their suggested alternative response. Results indicated that group interaction and feedback from the AI system improved policy quality. However, only when all three procedures were employed, group interaction, AI feedback, and forced reflective practice, was the development of individual expertise on the task enhanced.     

New method for solving reviewer assignment problem using type-2 fuzzy sets and fuzzy functions

Reviewer Assignment Problem (RAP) is one of the cardinal problems in Government Funding agencies where the expertise level of the referee reviewing a proposal needs to be optimised to guarantee the selection of good R&D projects. Although many solutions have been proposed for RAP in the past, none of them deals with the inherent imprecision associated with the problem. For instance, it is not possible to determine the “exact expertise level” of a particular reviewer in a particular domain. In this paper, we propose a novel approach for assigning reviewers to proposals. To calculate the expertise of a reviewer in a particular domain, we create a type-2 fuzzy set by assigning relevant weights to the various factors that affect the expertise of the reviewer in that domain. We also create a fuzzy set of the proposal by selecting three keywords that best represent the proposal. We then use a fuzzy functions based equality operator to compute the equality of the type-2 fuzzy set of experts and the fuzzy set of proposal keywords, which is then subjected to a set of relevant constraints to optimize the solution. We consider the four important aspects: workload balancing of reviewers, avoiding Conflicts of Interest, considering individual preferences by incorporating bidding and mapping multiple keywords of a proposal. As an extension to this approach, we further consider the relative importance of each keyword with respect to the submitted proposal by using representative percentage weights to create the FUZZY sets which represent the keywords. Hence, we propose an integrated solution based on the strong mathematical foundation of fuzzy logic, comprised of all the different aspects of expertise modeling and reviewer assignment. An Expert System has also been developed for the same.