An actor model for rule-based systems

Our research originates from a study of the possibilities of integrating rules and objects in knowledge-based systems. In the present work, we are interested in the interactionist perspective of an object. The stepwise reasoning of a diagnostic expert system, possibly involving subgoaling and interactions with the environment, can be easily codified by means of production rules over proposition literals. This set of rules can be graphically represented in a network manner denoting the relations between the rules. The individual nodes in the network can be expressed by means of autonomous objects and their relations, interpreted as possible communications between them. The objects are given a structure and a proper behaviour and cooperate for performing logical reasoning by means of forward and backward chaining inference processes. Therefore, designing this system implies addressing several basic issues such as inter-object communications and their synchronization. The problem here is not necessarily to develop a great intelligence locally but to develop strong networks of good communicators. This approach belongs to the interactionist representation current, where objects are called actors. In principle, the actors may carry out computation in parallel and provide a conceptual foundation for massively concurrent object-oriented paradigms. From this point of view, a system allowing for the simultaneous investigation of several rules and premises in the forward or the backward chaining would be significantly more efficient.     

Integrating database technology, rule-based systems and temporal reasoning for effective information systems: the TEMPORA paradigm

Abstract. Recent years have witnessed a growing realization that the development of large data-intensive, transaction-oriented information systems is becoming increasingly more difficult as user requirements become broader and more sophisticated. Contemporary approaches have been criticized for producing systems which are difficult to maintain and which provide little assistance in organizational developments. This paper introduces the TEMPORA paradigm, which is currently under development and which advocates a closer alignment between organizational policy and information system functionality. This viewpoint impacts on a number of critical issues related to the development process of information systems most notably in the nature of conceptual models, the discipline adopted for the development, the type of support provided by CASE tools and the run-time environment. The paper introduces the philosophy and architecture of the TEMPORA paradigm and describes the conceptual models, tools and run-time environment which render such an approach a feasible undertaking.     

A framework for validation of rule-based systems

We describe a complete methodology for the validation of rule-based expert systems. This methodology is presented as a five-step process that has two central themes: 1) to create a minimal set of test inputs that adequately cover the domain represented in the knowledge base; and 2) a Turing Test-like methodology that evaluates the system's responses to the test inputs and compares them to the responses of human experts. The development of minimal set of test inputs takes into consideration various criteria, both user-defined, and domain-specific. These criteria are used to reduce the potentially very large set of test inputs to one that is practical, keeping in mind the nature and purpose of the developed system. The Turing Test-like evaluation methodology makes use of only one panel of experts to both evaluate each set of test cases and compare the results with those of the expert system, as well as with those of the other experts. The hypothesis being presented is that much can be learned about the experts themselves by having them anonymously evaluate each other's responses to the same test inputs. Thus, we are better able to determine the validity of an expert system. Depending on its purpose, we introduce various ways to express validity as well as a technique to use the validity assessment for the refinement of the rule base. Lastly, we describe a partial implementation of the test input minimalization process on a small but nontrivial expert system. The effectiveness of the technique was evaluated by seeding errors into the expert system, generating the appropriate set of test inputs and determining whether the errors could be detected by the suggested methodology.     

Evolutionary rule-based systems for imbalanced data sets

This paper investigates the capabilities of evolutionary on-line rule-based systems, also called learning classifier systems (LCSs), for extracting knowledge from imbalanced data. While some learners may suffer from class imbalances and instances sparsely distributed around the feature space, we show that LCSs are flexible methods that can be adapted to detect such cases and find suitable models. Results on artificial data sets specifically designed for testing the capabilities of LCSs in imbalanced data show that LCSs are able to extract knowledge from highly imbalanced domains. When LCSs are used with real-world problems, they demonstrate to be one of the most robust methods compared with instance-based learners, decision trees, and support vector machines. Moreover, all the learners benefit from re-sampling techniques. Although there is not a re-sampling technique that performs best in all data sets and for all learners, those based in over-sampling seem to perform better on average. The paper adapts and analyzes LCSs for challenging imbalanced data sets and establishes the bases for further studying the combination of re-sampling technique and learner best suited to a specific kind of problem.     

A software engineering methodology for rule-based systems

Current expert systems are typically difficult to change once they are built. The authors introduce a method for developing more easily maintainable rule-based expert systems, which is based on dividing the rules into groups and focusing attention on those facts that carry information between rules in different groups. They describe a new algorithm for grouping the rules of a knowledge base automatically and a notation set of software tools for the proposed method. The approach is supported by a study of the connectivity of rules and facts in rule-based systems; it is found that they indeed have the latent structure necessary for the programming methodology. Recent experimental results also support the approach. In contrast to the homogeneous way in which the facts of a rule-based system are usually viewed, this approach shows that certain facts are more important than others with regard to future modifications of the rules     

Reliability testing of rule-based systems

Rule-based software systems are becoming more common in industrial settings, particularly to monitor and control large, real-time systems. The authors describe an algorithm for reliability testing of rule-based systems and their experience using it to test an industrial network surveillance system     

Verification of rule-based expert systems

We are investigating the problem of establishing computational rather than syntactic properties of forward-chaining rule-based expert systems. We model an expert system as a computation on working memory, define its execution semantics, and present proof techniques suitable for those semantics. Specifically, we model execution as a Dijkstra guarded-do construct, and use Dijkstra's Invariance Theorem and weakest precondition predicate transformers to establish invariants (safety properties) and postconditions (liveness properties). Our approach is an application of well-developed methods developed by Dijkstra and others for the verification of procedural programs. This paper introduces the approach, reports some initial results, and discusses future work.     

Optimizing real-time equational rule-based systems

Analyzing and reducing the execution-time upper bound of real-time rule-based expert systems is a very important task because of the stringent timing constraints imposed on this class of systems. We present a new runtime optimization to reduce the execution-time upper bound of real-time rule-based expert systems. In order to determine rules to be evaluated at runtime, a predicate dependency list, which consists of a predicate, its active rule set and corresponding inactive rule set, is created for each predicate in a real-time rule-based program. Based on the predicate dependency list and the current value of each variable, the new runtime optimization dynamically selects rules to be evaluated at runtime. For the timing analysis of the proposed algorithm, we introduce a predicate-based rule dependency graph, a predicate-based enable-rule graph, and their construction algorithm. We also discuss the bounded time of the equational logic rule-based program using the predicate-based rule dependency graph as well as the predicate-based enable-rule graph. The implementation and performance evaluation of the proposed algorithm using both synthetic and practical real-time rule-base programs are also presented. The performance evaluation shows that the runtime optimizer reduces the number of rule evaluations and predicate evaluations as well as the response time upper bound significantly, and the new algorithm yields better execution-time upper bound compared to other optimization methods.     

Missing data imputation for fuzzy rule-based classification systems

Fuzzy rule-based classification systems (FRBCSs) are known due to their ability to treat with low quality data and obtain good results in these scenarios. However, their application in problems with missing data are uncommon while in real-life data, information is frequently incomplete in data mining, caused by the presence of missing values in attributes. Several schemes have been studied to overcome the drawbacks produced by missing values in data mining tasks; one of the most well known is based on preprocessing, formerly known as imputation. In this work, we focus on FRBCSs considering 14 different approaches to missing attribute values treatment that are presented and analyzed. The analysis involves three different methods, in which we distinguish between Mamdani and TSK models. From the obtained results, the convenience of using imputation methods for FRBCSs with missing values is stated. The analysis suggests that each type behaves differently while the use of determined missing values imputation methods could improve the accuracy obtained for these methods. Thus, the use of particular imputation methods conditioned to the type of FRBCSs is required.     

Support vector learning for fuzzy rule-based classification systems

To design a fuzzy rule-based classification system (fuzzy classifier) with good generalization ability in a high dimensional feature space has been an active research topic for a long time. As a powerful machine learning approach for pattern recognition problems, the support vector machine (SVM) is known to have good generalization ability. More importantly, an SVM can work very well on a high- (or even infinite) dimensional feature space. This paper investigates the connection between fuzzy classifiers and kernel machines, establishes a link between fuzzy rules and kernels, and proposes a learning algorithm for fuzzy classifiers. We first show that a fuzzy classifier implicitly defines a translation invariant kernel under the assumption that all membership functions associated with the same input variable are generated from location transformation of a reference function. Fuzzy inference on the IF-part of a fuzzy rule can be viewed as evaluating the kernel function. The kernel function is then proven to be a Mercer kernel if the reference functions meet a certain spectral requirement. The corresponding fuzzy classifier is named positive definite fuzzy classifier (PDFC). A PDFC can be built from the given training samples based on a support vector learning approach with the IF-part fuzzy rules given by the support vectors. Since the learning process minimizes an upper bound on the expected risk (expected prediction error) instead of the empirical risk (training error), the resulting PDFC usually has good generalization. Moreover, because of the sparsity properties of the SVMs, the number of fuzzy rules is irrelevant to the dimension of input space. In this sense, we avoid the "curse of dimensionality." Finally, PDFCs with different reference functions are constructed using the support vector learning approach. The performance of the PDFCs is illustrated by extensive experimental results. Comparisons with other methods are also provided.     

Framework for building rule-based machine diagnostic expert systems

In the endeavour to build an expert system called XBAK using Personal Consultant Plus for the diagnosis of sophisticated equipment used in microchip manufacturing, a rule-based machine diagnostic expert system architecture was developed. The approach, features and technical implementation of this application-independent problem-solving structure are described. The architecture can be used as a framework for solving similar problems in the area of machine diagnostics.     

Modeling control in rule-based systems

The Comex modeling tool (control knowledge modeling and execution tool) focuses on modeling a problem-solving strategy and representing control knowledge, the knowledge of how to select among several problem-solving actions. In addition, Comex has facilities that let you execute early versions of the model to simulate the intended system's behavior and continuously execute the evolving model, until it becomes the real system. Comex addresses the rule-based paradigm's biggest shortcoming by adding control structures to executable models. Beginning with the specification phase, you can simulate a system's behavior, then map tasks to a rule base. The result is a structured, rule-based system built according to accepted software-engineering principles     

Adaptive fuzzy rule-based classification systems

This paper proposes an adaptive method to construct a fuzzy rule-based classification system with high performance for pattern classification problems. The proposed method consists of two procedures: an error correction-based learning procedure, and an additional learning procedure. The error correction-based learning procedure adjusts the grade of certainty of each fuzzy rule by its classification performance. That is, when a pattern is misclassified by a particular fuzzy rule, the grade of certainty of that rule is decreased. On the contrary, when a pattern is correctly classified, the grade of certainty is increased. Because the error correction-based learning procedure is not meaningful after all the given patterns are correctly classified, we cannot adjust a classification boundary in such a case. To acquire a more intuitively acceptable boundary, we propose an additional learning procedure. We also propose a method for selecting significant fuzzy rules by pruning unnecessary fuzzy rules, which consists of the error correction-based learning procedure and the concept of forgetting. We can construct a compact fuzzy rule-based classification system with high performance     

An ensemble method for fuzzy rule-based classification systems

Fuzzy rule-based classification systems are very useful tools in the field of machine learning as they are able to build linguistic comprehensible models. However, these systems suffer from exponential rule explosion when the number of variables increases, degrading, therefore, the accuracy of these systems as well as their interpretability. In this article, we propose to improve the comprehensibility through a supervised learning method by automatic generation of fuzzy classification rules, designated SIFCO–PAF. Our method reduces the complexity by decreasing the number of rules and of antecedent conditions, making it thus adapted to the representation and the prediction of rather high-dimensional pattern classification problems. We perform, firstly, an ensemble methodology by combining a set of simple classification models. Subsequently, each model uses a subset of the initial attributes: In this case, we propose to regroup the attributes using linear correlation search among the training set elements. Secondly, we implement an optimal fuzzy partition thanks to supervised discretization followed by an automatic membership functions construction. The SIFCO–PAF method, analyzed experimentally on various data sets, guarantees an important reduction in the number of rules and of antecedents without deteriorating the classification rates, on the contrary accuracy is even improved.     

Matrix formulation of fuzzy rule-based systems

In this paper, a matrix formulation of fuzzy rule based systems is introduced. A gradient descent training algorithm for the determination of the unknown parameters can also be expressed in a matrix form for various adaptive fuzzy networks. When converting a rule-based system to the proposed matrix formulation, only three sets of linear/nonlinear equations are required instead of set of rules and an inference mechanism. There are a number of advantages which the matrix formulation has compared with the linguistic approach. Firstly, it obviates the differences among the various architectures; and secondly, it is much easier to organize data in the implementation or simulation of the fuzzy system. The formulation will be illustrated by a number of examples.     

基于规则面向用户的动态视图的实现

PostgreSQL是一种对象关系型数据库管理系统.利用PostgreSQL的规则系统,实现了一种面向用户的动态视图建立的方法.该方法利用PostgreSQL的系统表、系统函数,通过创建视图的插入、更新、删除规则,动态的创建会话用户的视图.用户通过访问接口LIBPQ连接到数据库,授权访问和操作自己的视图.在数据库一级对用户数据进行了隔离.     

A token-flow paradigm for verification of rule-based expert systems

This paper presents a novel approach to the verification of rule-based systems (RBSs). A graph structure, called the rule-dependency graph (RDG), is introduced to describe the dependency relationship among the rules of an RBS, in which each type of improper knowledge forms a specific topological structure. Knowledge verification is then performed by searching for such topological structures through a token-flow paradigm. An algorithm is provided, which automatically generates a minimally sufficient set of literals as test tokens in the detection procedure. The proposed scheme can be applied to rules of non-Horn clause form in both propositional and first-order logic, and restrictions imposed by other graph-based approaches can be avoided. Furthermore, explicit and potential anomalies of RBSs can be correctly found, and efficient run-time validation is made possible.     

Parallel implementation of rule-based expert systems for interactive applications

A novel approach to parallel implementation of rule-based expert systems is presented in this paper. This approach is tailored for expert systems in interactive applications. Rule-based expert systems are modeled by state space and AND/OR graphs. The interdependences among rules are analyzed to guide rule-base partitioning and memory bank assignment. Parallel execution of rule-based expert systems is investigated in the environment of closely coupled multiprocessors. Algorithms are developed for the parallel execution of rules and for the allocation of data to memory banks in interactive applications.     

Design of rule-based models through information granulation

In this study, we explore the combination of two well defined topics in fuzzy systems research: fuzzy rule based systems, and information granulation. Rule based systems are a powerful and well-studied form of knowledge representation, due to their approximation abilities and interpretability. In recent years, these types of systems have become increasingly powerful with regards to modeling accuracy; however, many of these improvements come at the cost of model interpretability. This recent direction of research has left an unexplored avenue towards the generation of increasingly interpretable fuzzy rule based models, which we intend to explore. Information granulation is a relatively new, yet very promising area of research in human centric systems. As a form of knowledge representation, information granulation is very well suited to fuzzy rule based systems, where rules represent linguistic quantities in a, intuitively understandable format. It is notable that the combination of these two concepts has been left largely unstudied. We aim to explore this union by defining a methodology for the construction of a partially granular fuzzy rule based model. The aim of this novel model format is to provide a first step in the improvement of fuzzy model interpretability, through the use of information granulation. We are additionally interested in studying new ways of generating fuzzy rules; hence, we will also look at the use of hierarchical clustering as a potential alternative to the tried and tested Fuzzy C Means clustering algorithm. The models created using hierarchical clustering are then compared with those generated using Fuzzy C Means to evaluate the effectiveness of this algorithm. As a result of these experiments, we demonstrate that partially granular fuzzy rules are capable of providing a significant improvement to fuzzy rule interpretability, and we believe that granular fuzzy models present an exciting avenue of future research in human centric systems.