Seunghyun Im

Action rule extraction from a decision table: ARED

In this paper, we present an algorithm that discovers action rules from a decision table. Action rules describe possible transitions of objects from one state to another with respect to a distinguished attribute. The previous research on action rule discovery required the extraction of classification rules before constructing any action rule. The new proposed algorithm does not require pre-existing classification rules, and it uses a bottom up approach to generate action rules having minimal attribute involvement.

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.

Validating Secure Connections between Wireless Devices in Pervasive Computing Using Data Matrix

This paper describes the techniques for establishing secure associations between mobile devices in pervasive computing where the communications are often established with minimal priori knowledge. An improved method for securing the association by comparing a hashed key using Data Matrix will be presented. A Data Matrix displayed or printed on another device in the association is captured with a built-in digital camera. The image is then used to validate the devices public key.

Reclassification Rules

The ultimate goal of knowledge discovery (KD) is to extract sets of patterns leading to useful knowledge for obtaining user desirable outcomes. The key characteristics of knowledge usefulness is that these patterns are actionable. In the last decade, KD algorithms such as mining for association rules, clustering, and classification rules, have made a tremendous progress and have been demonstrated to be of significant value in a variety of real-world data mining applications. However, the results of the existing methods require to be further processed in order to suggest actions that achieve the desired outcome, by giving only previously acquired data. To address this issue, we present a novel technique, called reclassification rules, to gather all facts, to understand their causes and effects, and to list all potential solutions and the responding effects. Algorithm, Strategy Generator-II, is proposed to discover a complete set of reclassification rules which meets pre-specified constraints.

Protection of Sensitive Data based on Reducts in a Distributed Knowledge Discovery System

This paper discusses data confidentiality in a distributed knowledge discovery system (DKDS) . In particular, we provide a method that protects values of confidential attributes from being reveled by chase algorithm (Ras and Dardzinska, 2005) using reducts (Skowron and Rauszer, 1992). The method presented in this paper is intended for use in DKDS that the set of rules used by chase algorithm is not completely known in advance. Reduct is used to determine the optimal set of data to be additionally hidden.

Mining Non-redundant Reclassification Rules

The increased competition faced by todays companies can wield data mining tools to extract actionable knowledge and then use it as a weapon to outmaneuver competitors and boost revenue. Mining reclassification rules is a way to model actionable patterns directly from a given data set. The previous work on reclassification rule mining has shown that they are effective when variables are weakly correlated. However, when the data set is correlated, some redundant rules are in the result set. This problem becomes critical for discovering rules in correlated data which may have long frequent factor-sets. In this paper, we investigate properties of reclassification rules and offer a new method to discovery a set of non-redundant reclassification rules without information loss.

Data Confidentiality Versus Chase

We present a generalization of a strategy, called SCIKD, proposed in [7] that allows to reduce a disclosure risk of confidential data in an information system S[10] using methods based on knowledge discovery. The method proposed in [7] protects confidential data against Rule-based Chase, the null value imputation algorithm driven by certain rules [2], [4]. This method identifies a minimal subset of additional data in Swhich needs to be hidden to guarantee that the confidential data are not revealed by Chase. In this paper we propose a bottom-up strategy which identifies, for each object xin S, a maximal set of values of attributes which do not have to be hidden and still the information associated with secure attribute values of xis protected. It is achieved without examining all possible combinations of values of attributes. Our method is driven by classification rules extracted from Sand takes into consideration their confidence and support.

Multi-granularity classification rule discovery using ERID

This paper introduces the use of ERID [1] algorithm for classification rule discovery at various levels of granularity. We use an incomplete information system and attribute value hierarchy to extract rules. The incomplete information system is capable of storing weighted attribute values and the domains of those attributes are organized using a hierarchical tree structure. The granularity of attribute values can be adjusted using the attribute value hierarchy. The result is then processed through ERID, which is designed to discover rules from partially incomplete information systems. The capability of handling incomplete data enables to build more specific and general classification rules.

A visual way to talk to strangers: authentication in wireless pervasive computing

In this paper, we discuss a novel approach to identifying entities involved in ad-hoc wireless communications through using an effficient visual code system called UbiCode. UbiCode, used in an out-of-band channel in order to bootstrap trust between entities unknown to each other, facilitates a mechanism for demonstrative identification of entities involved, thereby enabling secure spontaneous communications among them in wireless pervasive computing environments. We present the design of the visual code system as well as different types of identification protocols leveraging the visual code system. We also demonstrate our approach through a proof-of-concept implementation.

Mining Generalized Actionable Rules Using Concept Hierarchies

A series of mining actionable rule methods have been proposed from various aspects, but the existing models do not incorporate the concept of hierarchy/taxonomy into the mining process and restrict the terms used to build actionable rules to atomic concepts. In order to resolve this problem, an integrated framework for extracting multiple-level actionable rules with ontology support is proposed so more generalized knowledge from data can be extracted. This type of generalized rules will contain not only the attribute values contained in data, but also some concepts encoded in a given taxonomy. Obtaining generalized actionable rules are a necessity since they provide a more general view of the domain. The proposed framework is based on a breadth-first top-downward model to be developed by extending the existing single-level actionable rule discovery methods. This framework can improve the quality of the extracted actionable rules in terms of their interestingness and understandability.