基于多层推理机制的机械产品概念设计

将类比推理的方法与基于实例的方法相结合，建立了基于多层推理的机械产品概念设计系统、知识的表示，采用面向应用的规则方法，框架式的知识结构。该系统有自学习的能力，也可以由用户进行知识的更改和完善，用户界面良好。     

基于案例推理的网络舆情辅助决策系统研究

为了实现网络舆情监测与辅助决策,提出一种基于案例推理的网络舆情辅助决策系统框架。首先综述网络舆情研究现状,提出基于案例推理的网络舆情辅助决策系统的意义;其次,研究基于案例推理的网络舆情辅助决策系统的工作流程,并提出基于案例推理的网络舆情辅助决策系统的框架;接着研究基于案例推理的网络舆情辅助决策系统中的关键内容,即案例表示与检索,并给出详细实现方法;最后,在以上研究基础上实现原型系统,结果表明该系统可以有效实现舆情辅助决策。     

AN INTEGRATED CLUSTERING‐BASED APPROACH TO FILTERING UNFAIR MULTI‐NOMINAL TESTIMONIES

Reputation systems have contributed much to the success of electronic marketplaces. However, the problem of unfair testimonies has to be addressed effectively to improve the robustness of reputation systems. Until now, most of the existing approaches focus only on reputation systems using binary testimonies, and thus have limited applicability and effectiveness. In this paper, We propose an i ntegrated CLU stering‐ B ased approach called iCLUB to filter unfair testimonies for reputation systems using multinominal testimonies, in an example application of multiagent‐based e‐commerce. It adopts clustering techniques and considers buyer agents’ local as well as global knowledge about seller agents. Experimental evaluation demonstrates the promising results of our approach in filtering various types of unfair testimonies, its robustness against collusion attacks, and better performance compared to competing models.     

ACTION FAILURE RECOVERY VIA MODEL‐BASED DIAGNOSIS AND CONFORMANT PLANNING

A plan carried on in the real world may be affected by a number of unexpected events, plan threats, which cause significant deviations between the intended behavior of the plan executor (i.e., the agent) and the observed one. These deviations are typically considered as action failures. This paper addresses the problem of recovering from action failures caused by a specific class of plan threats: faults in the functionalities of the agent. The problem is approached by exploiting techniques of the Model‐Based Diagnosis (MBD) for detecting failures (plan execution monitoring) and for explaining these failures in terms of faulty functionalities (agent diagnosis). The recovery process is modeled as a replanning problem aimed at fixing the faulty components identified by the agent diagnosis. However, since the diagnosis is in general ambiguous (a failure may be explained by alternative faults), the recovery has to deal with such an uncertainty. The paper advocates the adoption of a conformant planner, which guarantees that the recovery plan, if it exists, is executable no matter what the actual cause of the failure. The paper focuses on a single agent performing its own plan, however the proposed methodology takes also into account that agents are typically situated into a multiagent scenario and that commitments between agents may exist. The repair strategy is therefore conceived to overcome the causes of a failure while assuring the commitments an agent has agreed with other team members.     

A NEW E‐LEARNING ACHIEVEMENT EVALUATION MODEL BASED ON ROUGH SET AND SIMILARITY FILTER

The trend of utilizing information and Internet technologies as teaching and learning tools is rapidly expanding into education. E‐learning is one of the most popular learning environments in the information era. The Internet enables students to learn without limitations of space and time. Furthermore, the learners can repeatedly review the context of a course without the barrier of distance. Recently, student‐centered instruction has become the primary trend in education, and the e‐learning system, which is considered with regard to of personalization and adaptability, is more and more popular. By means of e‐learning systems, teachers can adjust the learning schedule instantly for each learner according to a student's achievements and build more adaptive learning environments. Sometimes, teachers give biased assessments of students’ achievements under uncontrollable conditions (i.e., tiredness, preference) and are in dire need of overcoming this predicament. To solve the drawback mentioned, a new model to evaluate learning achievements based on rough set and similarity filter is proposed. The proposed model includes four facets: (1) select important features (attributes) to enhance classification performance by feature selection methods; (2) utilize minimal entropy principle approach (MEPA) to fuzzify the quantitative data; (3) select linguistic values for each feature and delete inconsistent data using the similarity threshold (similarity filter); and (4) generate rules based on rough set theory (RST). The practical e‐learning achievement data sets are collected by an e‐learning online examination system from a university in Taiwan. To verify our model, the performances of the proposed model are compared with the listing models. Results of this study demonstrate that the proposed model outperforms the listing models.     

An Automatic Building Approach To Special Takagi‐Sugeno Fuzzy Network For Unknown Plant Modeling And Stable Control

In previous studies, several stable controller design methods for plants represented by a special Takagi‐Sugeno fuzzy network (STSFN) have been proposed. In these studies, the STSFN is, however, derived directly from the mathematical function of the controlled plant. For an unknown plant, there is a problem if STSFN cannot model the plant successfully. In order to address this problem, we have derived a learning algorithm for the construction of STSFN from input‐output training data. Based upon the constructed STSFN, existing stable controller design methods can then be applied to an unknown plant. To verify this, stable fuzzy controller design by parallel distributed compensation (PDC) method is adopted. In PDC method, the precondition parts of the designed fuzzy controllers share the same fuzzy rule numbers and fuzzy sets as the STSFN. To reduce the controller rule number, the precondition part of the constructed STSFN is partitioned in a flexible way. Also, similarity measure together with merging operation between each neighboring fuzzy set are performed in each input dimension to eliminate the redundant fuzzy sets. The consequent parts in STSFN are designed by correlation measure to select only the significant input terms to participate in each rule's consequence and reduce the network parameters. Simulation results in the cart‐pole balancing system have shown that with the proposed STSFN building approach, we are able to model the controlled plant with high accuracy and, in addition, can design a stable fuzzy controller with small parameter number.     

Evolution@home: observations on participant choice,work unit variation and low‐effort global computing

Many simulation efforts in ecology and evolutionary biology employ individual‐based models that are well suited for including many biological details. These models often pose serious computational challenges if all biologically interesting parameter combinations are to be explored. The challenges are even greater for biologists who often lack supercomputing facilities and the manpower for implementing complex global computing systems such as SETI@home. Under such limiting conditions, evolution@home started as a one‐man effort to distribute simulations of Muller's ratchet to Internet‐connected computers of participants from the general public. This paper addresses experiences in low‐effort global computing made with evolution@home over more than four years. It shows how allowing participants to choose the class of computational complexity they want to contribute to can help to deal with the bewildering variety of computational complexities that easily result from individual‐based models. Results suggest that, as a first rough approximation, participants' complexity choices are distributed surprisingly even over all reasonable classes of CPU‐time and RAM requirements. More often than not, participants tend to finish the simulations they start, if they are committed enough to submit any results at all. Potential uses of intermediate simulation results are discussed and the error of magnitude is introduced to help to deal with imprecise CPU‐time predictions. Experiences with the choices of over 300 users who have contributed more than 100 000 simulations with a total of over 80 years CPU time are reviewed. Copyright © 2007 John Wiley & Sons, Ltd.