知识发现中的因果关联规则挖掘研究

因果关联规则是知识库中一类重要的知识类型,具有重要的应用价值。首先对因果关系的特殊性质进行了分析,然后基于语言场和广义归纳逻辑因果模型,从表示、挖掘、评价和应用几方面,对因果关联规则的研究进行了详细论述。并在此基础上提出了隐含因果关联规则的概念。通过语言场和推理机制的运用,使因果关联规则这一重要知识形式的挖掘和评价过程具有良好的逻辑性和扩张性。     

Discovering regularities from knowledge bases

Knowledge bases open new horizons for machine learning research. One challenge is to design learning programs to expand the knowledge base using the knowledge that is currently available. This article addresses the problem of discovering regularities in large knowledge bases that contain many assertions in different domains. the article begins with a definition of regularities and gives the motivation for such a definition. It then outlines a framework that attempts to integrate induction with knowledge. Although the implementation of the framework currently uses only a statistical method for confirming hypotheses, its application to a real knowledge base has shown some encouraging and interesting results. © 1992 John Wiley & Sons, Inc.     

Sentimental causal rule discovery from Twitter

Social media, especially Twitter is now one of the most popular platforms where people can freely express their opinion. However, it is difficult to extract important summary information from many millions of tweets sent every hour. In this work we propose a new concept, sentimental causal rules, and techniques for extracting sentimental causal rules from textual data sources such as Twitter which combine sentiment analysis and causal rule discovery. Sentiment analysis refers to the task of extracting public sentiment from textual data. The value in sentiment analysis lies in its ability to reflect popularly voiced perceptions that are stated in natural language. Causal rules on the other hand indicate associations between different concepts in a context where one (or several concepts) cause(s) the other(s). We believe that sentimental causal rules are an effective summarization mechanism that combine causal relations among different aspects extracted from textual data as well as the sentiment embedded in these causal relationships. In order to show the effectiveness of sentimental causal rules, we have conducted experiments on Twitter data collected on the Kurdish political issue in Turkey which has been an ongoing heated public debate for many years. Our experiments on Twitter data show that sentimental causal rule discovery is an effective method to summarize information about important aspects of an issue in Twitter which may further be used by politicians for better policy making.     

Action rule discovery from incomplete data

Action rule is an implication rule that shows the expected change in a decision value of an object as a result of changes made to some of its conditional values. An example of an action rule is ‘credit card holders of young age are expected to keep their cards for an extended period of time if they receive a movie ticket once a year’. In this case, the decision value is the account status, and the condition value is whether the movie ticket is sent to the customer. The type of action that can be taken by the company is to send out movie tickets to young customers. The conventional action rule discovery algorithms build action rules from existing classification rules. This paper discusses an agglomerative strategy that generates the shortest action rules directly from a decision system. In particular, the algorithm can be used to discover rules from an incomplete decision system where attribute values are partially incomplete. As one of the testing domains for our research we take HEPAR system that was built through a collaboration between the Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences and physicians at the Medical Center of Postgraduate Education in Warsaw, Poland. HEPAR was designed for gathering and processing clinical data on patients with liver disorders. Action rules will be used to construct the decision-support module for HEPAR.     

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.     

Customer knowledge discovery from online reviews

The explosive growth of Chinese electronic market has made it possible for companies to better understand consumers?? opinion towards their products in a timely fashion through their online reviews. This study proposes a framework for extracting knowledge from online reviews through text mining and econometric analysis. Specifically, we extract product features, detect topics, and identify determinants of customer satisfaction. An experiment on the online reviews from a Chinese leading B2C (Business-to-Customer) website demonstrated the feasibility of the proposed method. We also present some findings about the characteristics of Chinese reviewers.     

On optimal rule discovery

In machine learning and data mining, heuristic and association rules are two dominant schemes for rule discovery. Heuristic rule discovery usually produces a small set of accurate rules, but fails to find many globally optimal rules. Association rule discovery generates all rules satisfying some constraints, but yields too many rules and is infeasible when the minimum support is small. Here, we present a unified framework for the discovery of a family of optimal rule sets and characterize the relationships with other rule-discovery schemes such as nonredundant association rule discovery. We theoretically and empirically show that optimal rule discovery is significantly more efficient than association rule discovery independent of data structure and implementation. Optimal rule discovery is an efficient alternative to association rule discovery, especially when the minimum support is low.     

Updating knowledge bases

We consider the problem of updating a knowledge base, where a knowledge base is realised as a normal (logic) program. We present procedures for deleting an atom from a normal program and inserting an atom into a normal program, concentrating particularly on the case when negative literals appear in the bodies of program clauses. We also prove various properties of the procedures including their correctness.     

Hybrid knowledge bases

Deductive databases that interact with, and are accessed by, reasoning agents in the real world (such as logic controllers in automated manufacturing, weapons guidance systems, aircraft landing systems, land-vehicle maneuvering systems, and air-traffic control systems) must have the ability to deal with multiple modes of reasoning. Specifically, the types of reasoning we are concerned with include, among others, reasoning about time, reasoning about quantitative relationships that may be expressed in the form of differential equations or optimization problems, and reasoning about numeric modes of uncertainty about the domain which the database seeks to describe. Such databases may need to handle diverse forms of data structures, and frequently they may require use of the assumption-based nonmonotonic representation of knowledge. A hybrid knowledge base is a theoretical framework capturing all the above modes of reasoning. The theory tightly unifies the constraint logic programming scheme of Jaffar and Lassez (1987), the generalized annotated logic programming theory of Kifer and Subrahmanian (1989), and the stable model semantics of Gelfond and Lifschitz (1988). New techniques are introduced which extend both the work on annotated logic programming and the stable model semantics     

Probabilistic knowledge bases

We define a new fixpoint semantics for rule based reasoning in the presence of weighted information. The semantics is illustrated on a real world application requiring such reasoning. Optimizations and approximations of the semantics are shown so as to make the semantics amenable to very large scale real world applications. We finally prove that the semantics is probabilistic and reduces to the usual fixpoint semantics of stratified Datalog if all information is certain. We implemented various knowledge discovery systems which automatically generate such probabilistic decision rules. In collaboration with a bank in Hong Kong we use one such system to forecast currency exchange rates     

Streaming-media knowledge discovery

As the amount of streaming audio and video available to World Wide Web users grows, tools for analyzing and indexing this content will become increasingly important. Frequently, knowledge management applications and information portals synthesize unstructured text information from the Web, intranets and partner sites. Given this context, we crawl a statistically significant number of Web pages, detect those that contain streaming media links, crawl the media links to extract associated meta-data, then use the crawl data to build a resource list for Web media. We have used these crawl-data findings to build a media indexing application that uses content-based indexing methods     

Time-evolving rule-based knowledge bases

Human knowledge in any expertise area changes with respect to time. Two types of such knowledge can be identified, time independent and time dependent. It is shown that the maintenance effort of the latter is harder than that of the former. The present paper applies research results in the area of temporal databases, in order to maintain a rule-based knowledge base whose content changes with respect to the real world time. It is shown that the approach simplifies the maintenance of time dependent knowledge. It also enables the study of the evolution of knowledge with respect to time, which is knowledge on its own. Three distinct solutions are actually proposed and evaluated. Their common characteristic is that knowledge is stored in a database; therefore, all the advantages of databases are inherited by knowledge bases. Implementations are also reported.     

一种防火墙规则冲突检测算法

在入侵检测系统和状态检测防火墙等应用中,规则冲突检测及冲突解析算法是影响安全性及服务质量的关键。首先对防火墙过滤规则之间的关系进行了建模和分类。然后在过滤规则关系分类的基础上提出了一种冲突检测算法。该算法能够自动检测、发现规则冲突和潜在的问题,并且能够对防火墙过滤规则进行无冲突的插入、删除和修改。实现该算法的工具软件能够显著简化防火墙策略的管理和消除防火墙的规则冲突。     

Combining multiple knowledge bases

Combining knowledge present in multiple knowledge base systems into a single knowledge base is discussed. A knowledge based system can be considered an extension of a deductive database in that it permits function symbols as part of the theory. Alternative knowledge bases that deal with the same subject matter are considered. The authors define the concept of combining knowledge present in a set of knowledge bases and present algorithms to maximally combine them so that the combination is consistent with respect to the integrity constraints associated with the knowledge bases. For this, the authors define the concept of maximality and prove that the algorithms presented combine the knowledge bases to generate a maximal theory. The authors also discuss the relationships between combining multiple knowledge bases and the view update problem     

Large engineering knowledge bases

This paper concerns the development of numerous knowledge bases for engineering applications and integrating them into one knowledge environment applicable in different problem domains. It discusses steps made towards building large heterogeneous engineering knowledge bases. It analyses the requirements to large knowledge bases, presents their architecture and discusses the content and size of general-purpose engineering knowledge bases.     

Multi-level rough set reduction for decision rule mining

Most previous studies on rough sets focused on attribute reduction and decision rule mining on a single concept level. Data with attribute value taxonomies (AVTs) are, however, commonly seen in real-world applications. In this paper, we extend Pawlak’s rough set model, and propose a novel multi-level rough set model (MLRS) based on AVTs and a full-subtree generalization scheme. Paralleling with Pawlak’s rough set model, some conclusions related to the MLRS are given. Meanwhile, a novel concept of cut reduction based on MLRS is presented. A cut reduction can induce the most abstract multi-level decision table with the same classification ability on the raw decision table, and no other multi-level decision table exists that is more abstract. Furthermore, the relationships between attribute reduction in Pawlak’s rough set model and cut reduction in MLRS are discussed. We also prove that the problem of cut reduction generation is NP-hard, and develop a heuristic algorithm named CRTDR for computing the cut reduction. Finally, an approach named RMTDR for mining multi-level decision rule is provided. It can mine decision rules from different concept levels. Example analysis and comparative experiments show that the proposed methods are efficient and effective in handling the problems where data is associated with AVTs.     

Updating knowledge bases II

We consider the problem of updating a knowledge base, where a knowledge base is realised as a (logic) program. In a previous paper, we presented procedures for deleting an atom from a normal program and inserting an atom into a normal program, concentrating particularly on the case when negative literals appear in the bodies of program clauses. We also proved various properties of the procedures including their correctness. Here we present mutually recursive versions of the update procedures and prove their correctness and other properties. We then generalise the procedures so that we can update an (arbitrary) program with an (arbitrary) formula. The correctness of the update procedures for programs is also proved.     

Nonmonotonic reasoning from conditional knowledge bases with system W

Annals of Mathematics and Artificial Intelligence - For nonmonotonic reasoning in the context of a knowledge base $\mathcal {R}$ containing conditionals of the form If A then usually B, system P...