An adaptive genetic-based signature learning system for intrusion detection

Rule-based intrusion detection systems generally rely on hand crafted signatures developed by domain experts. This could lead to a delay in updating the signature bases and potentially compromising the security of protected systems. In this paper, we present a biologically-inspired computational approach to dynamically and adaptively learn signatures for network intrusion detection using a supervised learning classifier system. The classifier is an online and incremental parallel production rule-based system.A signature extraction system is developed that adaptively extracts signatures to the knowledge base as they are discovered by the classifier. The signature extraction algorithm is augmented by introducing new generalisation operators that minimise overlap and conflict between signatures. Mechanisms are provided to adapt main algorithm parameters to deal with online noisy and imbalanced class data. Our approach is hybrid in that signatures for both intrusive and normal behaviours are learnt.The performance of the developed systems is evaluated with a publicly available intrusion detection dataset and results are presented that show the effectiveness of the proposed system.     

基于Web Service的虚拟研究中心KMS构建

针对虚拟研究中心知识管理敏捷化、柔性化和分布式的需求特性,结合Web Service技术搭建了虚拟研究中心知识管理系统网络体系架构,提出一种基于Web Service的虚拟研究中心知识管理系统模型。依据模型框架,构建雅砻江水电开发虚拟研究中心知识管理系统。研究表明,依据模型构建的虚拟研究中心知识管理系统能有效地实现基于Web的知识共享、交流和创新,模型具有较好的可靠性和可行性。     

Online collaborative learning in dyads: Effects of knowledge distribution and awareness

Research on collaborative learning traditionally assumes a certain degree of symmetry between the learning partners in terms of both their learning-relevant traits and their individual learning outcomes. However, if one collaborative partner is clearly more able, skilled, or knowledgeable than the other partner, then it remains unclear who profits more from the collaboration. The present study aimed to explore this issue by manipulating symmetry in prior knowledge within small groups of online learners (dyads) and measuring their problem-solving efficiency and incidental learning gain on an individual and dyad level. Awareness of this symmetry/asymmetry was manipulated, too, to discern it as a potential moderator. Dyads with symmetrical and asymmetrical prior knowledge performed equally well on most measures. Moreover, on average, the more and the less knowledgeable partners in the asymmetrical conditions had equal learning gains. However, while in dyads with symmetrical knowledge learning gains were correlated between the partners, in the asymmetrical dyads they were not. Awareness of symmetry/asymmetry did not act as a moderator, but, overall, dyads with awareness of each other's knowledge learned more from each other than dyads without such awareness. The benefit of awareness was, however, specific to the learning content exposed via awareness. We conclude that researchers and practitioners should be careful when choosing or assigning collaborative partners to each other, as only for partners with symmetrical prior knowledge can a symmetrical increase in knowledge be expected. We further discuss the implications of these findings for research on knowledge awareness and collaboration.     

The role of creation mode and social networking mode in knowledge creation performance: Mediation effect of creation process

This study proposes and empirically examines a research model that incorporates the knowledge creation mode and social networking mode to describe knowledge creation performance. The mediation effect of the knowledge creation process is explored in terms of socialization, externalization, combination, and internalization (SECI). The data collected from the manufacturing and service industries in Taiwan were analyzed. The goal-driven mode and web topology are found to be significantly associated with product or service creation primarily because of the creation efficacy aspect. The SECI with web topology has a mediation function when the goal-driven mode is adopted. Implications and suggestions are also provided.     

A collaborative system for capturing and reusing in-context design knowledge with an integrated representation model

Current research on design knowledge capture and reuse has predominantly focused on either the codification view of knowledge or the personalisation view of knowledge, resulting in a failure to address designers’ knowledge needs caused by a lack of context of information and insufficient computational support. Precisely motivated by this gap, this work aims to address the integration of these two views into a complete, contextual and trustworthy knowledge management scheme enabled by the emerging collaborative technologies. Specifically, a knowledge model is developed to represent an integrated knowledge space, which can combine geometric model, knowledge-based analysis codes and problem-solving strategies and processes. On this basis, a smart collaborative system is also designed and developed to streamline the design process as well as to facilitate knowledge capture, retrieval and reuse as users with different roles are working on various tasks within this process. An engineering case study is undertaken to demonstrate the idea of collaborative knowledge creation and sharing and evaluate the effectiveness of the knowledge representation model and the collaborative technologies employed. As evidenced in the development and evaluation, the methods proposed are effective for capturing an integrated knowledge space and the collaborative knowledge management system not only facilitates problem-solving using knowledge-based analysis but also supplies in-context tacit knowledge captured from the communications between users throughout the design process.     

Decision-theoretic planning with generalized first-order decision diagrams

Many tasks in AI require representation and manipulation of complex functions. First-Order Decision Diagrams (FODD) are a compact knowledge representation expressing functions over relational structures. They represent numerical functions that, when constrained to the Boolean range, use only existential quantification. Previous work has developed a set of operations for composition and for removing redundancies in FODDs, thus keeping them compact, and showed how to successfully employ FODDs for solving large-scale stochastic planning problems through the formalism of relational Markov decision processes (RMDP). In this paper, we introduce several new ideas enhancing the applicability of FODDs. More specifically, we first introduce Generalized FODDs (GFODD) and composition operations for them, generalizing FODDs to arbitrary quantification. Second, we develop a novel approach for reducing (G)FODDs using model checking. This yields – for the first time – a reduction that maximally reduces the diagram for the FODD case and provides a sound reduction procedure for GFODDs. Finally we show how GFODDs can be used in principle to solve RMDPs with arbitrary quantification, and develop a complete solution for the case where the reward function is specified using an arbitrary number of existential quantifiers followed by an arbitrary number of universal quantifiers.     

Dealing with logical omniscience: Expressiveness and pragmatics

We examine four approaches for dealing with the logical omniscience problem and their potential applicability: the syntactic approach, awareness, algorithmic knowledge, and impossible possible worlds. Although in some settings these approaches are equi-expressive and can capture all epistemic states, in other settings of interest (especially with probability in the picture), we show that they are not equi-expressive. We then consider the pragmatics of dealing with logical omniscience—how to choose an approach and construct an appropriate model.     

Options analysis and knowledge management: Implications for theory and practice

Real options can be a powerful tool for quantifying the value of strategic and operational flexibility associated with uncertain IT investments. They also constitute a new way of thinking as to how knowledge management (KM) can be implemented and managed to maximize the upside potential while minimizing any downside risk. This paper explored how practitioners can incorporate options analysis into contemporary knowledge management. Options analysis is recognizing real options and then determines how they add value. The main issue here is to manage knowledge so that the theoretical option value is realized in practice. This paper developed a new metric, knowledge management performance index (KMPI), for evaluating the performance of a firm in its KM at any given point in time as follows: knowledge creation, knowledge conversion, knowledge circulation and knowledge completion. The higher the efficiency of the KM process, the higher the KMPI, thereby is enabling firms to become increasingly knowledgeable. To prove the contribution of the KMPI, a questionnaire survey was conducted among 121 firms listed on the Taiwan Stock Exchange Corporation (TSEC). This paper makes five important contributions: (1) it shows that if the KM process improves, the KMPI is enhanced. This finding is based on a survey of the questionnaires. As a result, the KMPI can effectively and efficiently represent the KM process; (2) it shows that when the KMPI increases, a company’s performance will be enhanced in five different areas This indicates a significant relationship between the KMPI and the company’s performance; (3) it also shows that the higher the KMPI, the higher the performance of the organization. This is especially valuable to any organization where knowledge is not being used optimally; (4) it provides an option analysis that may be helpful to managers for making the right decision in an uncertain environment; and (5) it presents the first application of the Black-Scholes model to use an actual business situation involving KM as its test bed. The results proved that the option pricing model can act as a measurement guideline for the entire knowledge of the whole company.     

RFuzzy: Syntax, semantics and implementation details of a simple and expressive fuzzy tool over Prolog

We present the RFuzzy framework, a Prolog-based tool for representing and reasoning with fuzzy information. The advantages of our framework in comparison to previous tools along this line of research are its easy, user-friendly syntax, and its expressivity through the availability of default values and types.In this approach we describe the formal syntax, the operational semantics and the declarative semantics of RFuzzy (based on a lattice). A least model semantics, a least fixpoint semantics and an operational semantics are introduced and their equivalence is proven. We provide a real implementation that is free and available. (It can be downloaded from http://babel.ls.fi.upm.es/software/rfuzzy/.) Besides implementation details, we also discuss some actual applications using RFuzzy.     

Comparing three methods for modeling the uncertainty in knowledge discovery from area-class soil maps

Knowledge discovery has been demonstrated as an effective approach to extracting knowledge from existing data sources for soil classification and mapping. Soils are spatial entities with fuzzy boundaries. Our study focuses on the uncertainty associated with class assignments when classifying such entities. We first present a framework of knowledge representation for categorizing spatial entities with fuzzy boundaries. Three knowledge discovery methods are discussed next for extracting knowledge from data sources. The methods were designed to maintain information for modeling the uncertainties associated with class assignments when using the extracted knowledge for classification. In a case study of knowledge discovery from an area-class soil map, all three methods were able to extract knowledge embedded in the map to classify soils at accuracies comparable to that of the original map. The methods were also able to capture membership gradations and helped to identify transitional zones and areas of potential problems on the source map when measures of uncertainties were mapped. Among the three methods compared, a fuzzy decision tree approach demonstrated the best performance in modeling the transitions between soil prototypes.