具有两级不确定性的推理模型

本文描述了断言信度偏差,规则强度偏差,断言可靠度,规则可靠度,二级不确定性等概念,给出了处理信度偏差与规则强度偏差的推理方法和具有两级不确定性(可信度,可靠度)的推理模型,本文还对参与协作的诸ES所输出的结论的评价方法进行了讨论。     

模糊推理在空调系统故障诊断中的应用

针对空调系统的故障诊断问题及其特点，提出了一种基于诊断知识的模糊描述和模糊推理方法，阐述了空调系统故障诊断专家系统中前向推理、后向推理以及正反向混台推理模糊断言可信度的计算方法，并且给出了相应的实例。     

分布式多视点软件开发环境中的不一致性检测与管理

针对基于分布式多视点环境的软件开发中存在的不一致性问题,提出了一个不一致性管理模型ＩＭＭ,建立了水平不一致性和垂直不一致性的新概念。讨论了不一致性的检测方法和处理策略。     

基于领域自然语言理解的知识库管理系统

文中根据数学领域自然语言理解的特点,主要实现了一个基于领域自然语言理解的知识库管理系统,给出了概念知识库异常检测的基本内容、流程和方法.重点分析丁知识库的冗余性、不一致性等方面的内容,给出了相应的检测方法,应用于知识库的具体组织与实现,在一定程度上解决了知识库的冗余和一致性问题.目的是向中学数学智能辅导系统提供自然语言接口,建立领域知识库作为理解建模系统的输入.该系统在教学领域智能辅导系统中得到了较好的应用.     

基于线性误差断言的推理方法

误差在系统中是普遍存在的。在安全关键系统中,对误差的定量分析是必要的,而以往的推理验证方法较少考虑误差。误差通常用区间数来刻画,从而推广了线性断言,并给出了线性误差断言的概念。此外,结合凸集的性质,提出了求解线性误差断言顶点的具体方法,并验证了该方法的正确性。通过分析相关概念及定理,将判断线性误差断言之间的蕴含关系的问题转化为前驱断言的顶点是否被包含在后驱断言的零点集的判断问题,从而给出了判断线性误差断言的蕴含关系的具体方法步骤,且该方法易于在计算机上编程实现。最后,给出该方法在火车加速状态上的应用,并且用大量随机实例测试了该方法的正确性。与不含误差语义的推理方法相比,该方法在含误差参数的系统的推理验证领域是有优势的。     

一种简化证据理论模型的研究

文中试图从几个方面来改进简化证据理论模型ＵＲＭ１的性能,针对ＵＲＭ１,文中给出了两个基本支持函数（相当于证据理论的Ｍａｓｓ函数）间的距离函数,进一步讨论了基本支持函数间的组合条件;给出了基于向最高权威看齐策略的组合带权威性因子基本支持函数的方法。提出了一种对证据理论的证据区间进行决策的方法,一方面,它比ＵＲＭ１中的证据区间决策方法更具一般性,另一方面,它更符合人类的直觉,同时,它对多个异质型专家系     

吝啬语义信任协商

自动信任协商足通过数字证书的交互披露在陌生实体间建立信任的过程,现有的自动信任协商框架要求主体出示完整属性证书来证明自己满足对方资源披露策略中的身份断言约束条件,导致了属性证书中身份信息过度披露问题.该文中作者提出一种通过交换DL-TNL语义身份断言来建立信任关系的吝啬语义信任协商框架,避免了完整属性证书的直接出示,减少信任建立过程中身份信息被披露的程度,并提出一种正确、完备且有效的吝啬语义信任协商策略.在该策略下,交易双方的协商引擎可以快速有效地从由身份断言权威签发的包含多个DL-TNL语义身份断言的身份断占证书中,自动计算出批露最少信息且符合对方策略的身份断言集进行出示,以最大限度地减少信任建立过程中被披露的身份信息,并保证理论上存在成功可能性时,使用该策略必然可以有效地帮助主体最终获得网络资源的访问权限.有关该策略完备性和正确性的证明以及策略实施中所使用的符合性检测算法的实现和分析均在文中给出.     

一种基于时滞弱化策略的通用信度重估模型

为提高信度评估的可靠性，从改善信度评估时效性角度出发，提出一种改进的通用信度重估模型GMRR。该模型借助原始信度、历史信度、信度波动率和信度波动趋势4种基础数据，实现了信度评估时效性机制与现有信度重估机制的有效集成，从多个维度对原始信度展开重估。给出信度评估时效性机制采用的时滞弱化策略、GMRR模型描述及分量计算算法，对模型的信度重估效果进行实验评估，结果表明，与现有信度重估模型相比，该模型获得的重估信度接近于目标实体的真实行为，同时能抑制恶意实体的策略波动行为。     

从C++的语言特征看面向对象的知识表示在专家系统中的应用

本文在分析了C++的语言特性以及知识表示复杂性的基础上,针对AI领域中ES实用化这一重要环节,结合知识工程的基本原理,提出一种面向对象的知识表示方法(OOKR),讨论了用C++的实现OOKR方面的技术策略,并给出了一个面向对象的图形知识库应用框架(OOCAD)的实例。     

一种多ES集成式决策支持系统的总体结构设计探讨

决策支持系统(DSS)是管理决策科学领域里的一个新的研究分支,专家系统(ES)则是人工智能领域里的一个最热门的研究子域,DSS同ES的集成研究已开始引起引起人们的关注。本文阐述了DSS同多ES集成的必要性,给出了多ES集成式决策支持系统(MESIDSS)的基本概念和特征,描述了MESIDSS所需的总体结构,讨论了MESIDSS中各组成部分所应具备的功能,最后指出了MESIDSS今后有待进一步研究的课题。     

Conflict-resolution in multi-ES cooperation system

In the so called “Group Problem”^[4](we would like to term it “Group-Consulting Type Cooperation”),ESs of the same cooperating group may have different views towards the same problem,and the differences may sometimes be very great.For the circumstance of those ESw with relatively closer views,we present in this paper three methods of consistency treatment and a method for calculating the maximal daviation of the confidence value about the cooperating group‘s common view base on the consistency degree of theose different views.For the circumstance of those views being not very close to each other and also not very different from each other,we designed two synthesis functions for synthesizing multiple views.These two circumstances are formally defined in this paper.A comparison between these two functions and the combining functon in [4] is also demonstrated.In the last section of this paper,we present a discussion about some further research topics of conflict-resolution in the group problem.     

Cooperative Multiple Pursuers against a Single Evader

This paper considers a pursuit-evasion game for non-holonomic systems where a group of pursuers attempts to capture an evader in a bounded connected domain. The problem is challenging because all vehicles have the same maneuvering capability in terms of speed and turn radius constraint. The paper initially discusses a simple approach for holonomic systems that is based on the minimization of the safe-reachable area (the area containing the set of points to where an evader can travel without being caught). This idea is then extended to develop a pursuit-evasion strategy for non-holonomic systems. However, solving such a problem is computationally intractable. Therefore, we propose a computationally efficient algorithm to obtain approximate solutions. This paper also proposes an alternative approach to obtain a simple yet effective solution to the cooperative pursuit problem that is based on missile guidance laws. As there is no analytical proof of capture, we empirically evaluate the performance of the algorithms and perform a comparative study using solutions obtained from umpteen simulations. A total of four different cooperative pursuit strategies and three different evader strategies are taken into account for the comparative study. In the process, an evader strategy which is superior to that based on the optimization of safe-reachable area is also identified.     

Diversified recommendation method combining topic model and random walk

As one of the most widely used algorithms in recommendation field, collaborative filtering (CF) predicts the unknown rating of items based on similar neighbors. Although many CF-based recommendation methods have been proposed, there still be room for improvement. Firstly, the data sparsity problem still remains a big challenge for CF algorithms to find similar neighbors. Secondly, there are many redundant similar items in the recommendation list generated by traditional CF algorithms, which cannot meet the user wide interest. Therefore, we propose a diversified recommendation method combining topic model and random walk. A weighted random walk model is presented to find all direct and indirect similar neighbors on the sparse data, improving the accuracy of rating prediction. By taking both users’ behavior data and items’ lags into account, we give a diversity measurement method based on the topic distribution of items discovered by Linked-LDA model. Furthermore, a diversified ranking algorithm is developed to balance the accuracy and diversity of recommendation results. We compare our method with six other recommendation methods on a real-world dataset. Experimental results show that our method outperforms the other methods and achieves the best personalized recommendation effect.     

网络能耗最小化的射频能量源布置与发射功率设置

射频能量捕获是应对无线网络节点能量受限的有效方法之一.射频能量源（energy source,简称ES）的布置位置和发送功率决定了各个节点的能量捕获功率.现有的研究工作大部分考虑的是没有给定侯选位置的场景.然而,在实际应用场景中,网络区域往往存在很多不可布置能量源的区域,使得能量源只能在一些合理的候选位置中布置.目前仅有少量相关工作研究如何在ES的候选布置位置中选择合适布置位置.已知节点位置、节点的能量捕获功率需求值、ES的个数以及ES的候选布置位置.研究并设计了最小化ES总供能的ES布置与发送功率设置方案.首先将该问题建模为混合整数规划问题;然后分别提出了一种具有较低复杂度的启发式算法和一种能够达到更小总供能的基于遗传算法的算法.仿真结果表明,与布置位置随机挑选法相比,这两种算法的网络总功耗降低了约90%,而遗传算法可达到比启发式算法高约35%的节能效果.因此,基于遗传算法的布置算法可用于中小规模的ES布置场景,而启发式算法可用于大规模的ES布置场景.     

Agent-based game-theoretic model for collaborative web services: Decision making analysis

In agent and (web) service computing, collaboration takes place when distributed entities have limited knowledge and capabilities, so they cannot perform required tasks without interacting and helping each other. For instance, web services, which are loosely-coupled business applications, are called to cooperate in distributed settings for the sake of efficiency. In this context, agents that abstract and act on behalf of web services could act in cooperative groups that gather a number of agents sharing some common goals. Enabling those agent-based web services to decide about their strategies in terms of joining and acting within groups, inviting other agents to join, and leaving a group to act alone is an open issue that we address in this work. In this paper, we propose a framework where agent-based web services select strategies that maximize their outcomes. These strategies could be categorized into cooperative strategies involving other agents and strategies that highlight the single operative attitude. Although cooperation seems to bring better utility to cooperative agents, we highlight that web services in some environments obtain better outcome while they act individually (i.e., outside the group). This means that the cost of cooperation (in some particular cases) might negatively influence the outcome and obtained utility. As solution, we propose in this paper (1) an agent-based model that formalizes web services decision making where different parameters are considered; and (2) a game-theoretic framework that analyzes the web services strategies allowing them to maximize their acting performance where non-zero-sum games are being used. The paper presents theoretical results, which are also confirmed through extensive simulations.     

A new method selection approach for fuzzy group multicriteria decision making

Fuzzy multicriteria decision making (MCDM) has been widely used in ranking a finite number of decision alternatives characterized by fuzzy assessments with respect to multiple criteria. In group decision settings, different fuzzy group MCDM methods often produce inconsistent ranking outcomes for the same problem. To address the ranking inconsistency problem in fuzzy group MCDM, this paper develops a new method selection approach for selecting a fuzzy group MCDM method that produces the most preferred group ranking outcome for a given problem. Based on two group averaging methods, three aggregation procedures and three defuzzification methods, 18 fuzzy group MCDM methods are developed as an illustration to solve the general fuzzy MCDM problem that requires cardinal ranking of the decision alternatives. The approach selects the group ranking outcome of a fuzzy MCDM method which has the highest consistency degree with its corresponding ranking outcomes of individual decision makers. An empirical study on the green bus fuel technology selection problem is used to illustrate how the approach works. The approach is applicable to large-scale group multicriteria decision problems where inconsistent ranking outcomes often exist between different fuzzy MCDM methods.     

面向常识的时间推理

常识和时间推理是人工智能研究的两个主要课题.Allen等人提出的时间推理缺少时间点、时区和时距的统一表示;过分考虑计算,缺少规则推理;求解算法难以应用于多Agent环境并且没有考虑常识不一致性.该文提出一种时间信息表示网络,分析了约束之间的推导规则,给出了常识时间问题的多Agent合作满足弱路径一致性的求解方法.该文的工作改进了Meiri,Wetprasit和Sattar等人的工作,为时间推理结合常识特性和适应多Agent合作环境提供了可行的方案.     

Rollout strategies for sequential fault diagnosis

Test sequencing is a binary identification problem wherein one needs to develop a minimal expected cost testing procedure to determine which one of a finite number of possible failure sources, if any, is present. The problem can be solved optimally using dynamic programming or AND/OR graph search methods (AO/sup */, CF, and HS). However, for large systems, the associated computation with dynamic programming or AND/OR graph search methods is substantial, due to the rapidly increasing number of OR nodes (denoting ambiguity states) and AND nodes (denoting tests) in the search graph. In order to overcome the computational explosion, the one-step or multistep lookahead heuristic algorithms have been developed to solve the test sequencing problem. In this paper, we propose to apply rollout strategies, which can be combined with the one-step or multistep lookahead heuristic algorithms, in a computationally more efficient manner than the optimal strategies, to obtain solutions superior to those using the one-step or multistep lookahead heuristic algorithms. The rollout strategies are illustrated and tested using a range of real-world systems. We show computational results, which suggest that the information-heuristic based rollout policies are significantly better than other rollout policies based on Huffman coding and entropy.     

Typicality‐based collaborative filtering for book recommendation

Nowadays, personalized recommender system placed an important role to predict the customer needs, interest about particular product in various application domains, which is identified according to the product ratings. During this process, collaborative filtering (CF) has been utilized because it is one of familiar techniques in recommender systems. The conventional CF methods analyse historical interactions of user‐item pairs based on known ratings and then use these interactions to produce recommendations. The major challenge in CF is that it needs to calculate the similarity of each pair of users or items by observing the ratings of users on same item, whereas the typicality‐based CF determines the neighbours from user groups based on their typicality degree. Typicality‐based CF can predict the ratings of users with improved accuracy. However, to eliminate the cold start problem in the proposed recommender system, the demographic filtering method has been employed in addition to the typicality‐based CF. A weighted average scheme has been applied on the combined recommendation results of both typicality‐based CF and demographic‐based CF to produce the best recommendation result for the user. Thereby, the proposed system has been able to achieve a coverage ratio of more than 95%, which indicates that the system is able to provide better recommendation for the user from the available lot of products.