Israel Spiegler

Technology and knowledge: bridging a generating gap

Refuting the notion of technology as a replacement of knowledge, this paper focuses on a gap between them that needs to be bridged. The idea is that technology represents the means, and knowledge the end of a process that includes many explicit and implicit methods for generating knowledge by using technology. Among these methods is data mining (DM), the leading thrust in the effort to gain actionable information from operational databases of organizations; this is particularly evident in direct marketing, customer relationship management (CRM), user profiling, and e-commerce applications.Two models of knowledge are reviewed. The first follows a conventional hierarchy of data, information and knowledge with a spiral and recursive way of generating knowledge. The other presents a reverse hierarchy where knowledge precedes the data-to-information process. The models are compared and discussed in the context of knowledge management (KM), using DM as an example.

Investigating diversity of clustering methods: An empirical comparison

The paper aims to shed some light on the question why clustering algorithms, despite being quantitative and hence supposedly objective in nature, yield different and varied results. To do that, we took 10 common clustering algorithms and tested them over four known datasets, used in the literature as baselines with agreed upon clusters. One additional method, Binary-Positive, developed by our team, was added to the analysis. The results affirm the unpredictable nature of the clustering process, point to different assumptions taken by different methods. One conclusion of the study is to carefully choose the appropriate clustering method for any given application.

Integrating system analysis and project management tools

Abstract Currently, computer-aided tools for system analysis are distinct from project management tools. This study proposes and prototypes a model that integrates these two aspects of the Information System Life Cycle (ISLC) by automatically mapping system analysis objects into project management objects. To validate the feasibility of our model and without loss of generality, the conversion of Data Flow Diagrams (DFD) objects into Gantt and Pert diagrams is demonstrated in this study. Experiments with the prototype confirm that integrating common tools for system analysis and standard tools for project management, during system development, helps improve system building tasks and their management. In addition, project managers using the proposed mapping approach can better assess project duration and system performance parameters such as response time and data traffic. We address implications of our work to both academics and practitioners, discussing directions future research might take as well as opportunities and prospects for commercialization of the proposed approach.

Data Mining by Means of Binary Representation: A Model for Similarity and Clustering

In this paper we outline a new method for clustering that is based on a binary representation of data records. The binary database relates each entity to all possible attribute values (domain) that entity may assume. The resulting binary matrix allows for similarity and clustering calculation by using the positive (‘1’ bits) of the entity vector. We formulate two indexes: Pair Similarity Index (PSI) to measure similarity between two entities and Group Similarity Index (GSI) to measure similarity within a group of entities. A threshold factor for each attribute domain is defined that is dependent on the domain but independent of the number of entities in the group. The similarity measure provides simplicity of storage and efficiency of calculation. A comparison of our similarity index to other indexes is made. Experiments with sample data indicate a 48% improvement of group similarity over standard methods pointing to the potential and merit of the binary approach to clustering and data mining.

Information as inventory: A new conceptual view

Abstract A new approach to viewing and handling information in the organization is presented. Information is treated as inventory in its three stages: raw materials (data to be processed), work in process (data being converted into information), and finished goods (information being stored). In line with modern inventory management methods, such as Just In Time, we introduce the concept of Information In Process (IIP). We show that IIP is costly, hard to control, and hinders the decision-making ability of those for whom information is gathered and processed in the first place — the users. Mismanaging the information resource by stocking IIP is suggested to be a cause of many system failures and malfunctions as is the case with production systems accumulating inventories. A methodology for applying inventory control techniques to the area of information systems is outlined aiming at the reduction of IIP.

Hempel's raven paradox: a positive approach to cluster analysis

Abstract A practical conclusion of the Hampel Raven paradox suggests a logical preference for using positive predicates in formulating scientific hypotheses. This led us to outline a new cluster analysis and grouping technique. We define a positive attribute distance (PAD) index that uses a binary representation of the existence or absence of an attribute value in a given object being observed. The resulting binary string representing an entity is then used to calculate distance to other strings using only the ‘1’ bits. This measure, with a matching grouping technique, simplifies clustering and grouping and yields equivalent or better results, as well as more efficient and compact calculations. Scope and purpose Cluster analysis is widely used in many fields of social science. Its basic aim is to assign individuals or objects under study into groups so that they have a high degree of similarity within the group, and that the groups are to be distinct. Various methods have been developed for clustering including regression and other statistical techniques. This paper introduces a new approach for clustering by using a computer representation form – binary 1 and 0 digits. A binary matrix is constructed from the data where rows represent the individuals (entities) and columns are values of attributes measured. The binary content of the matrix indicates which entity has or lacks certain attributes. This representation, simple, compact, and efficient in terms of computer application, allows clustering and grouping calculations that take into account only the positive attributes. Such technique compares favorably with conventional binary representation and has potential for use in cluster analysis.

CM-tree: A dynamic clustered index for similarity search in metric databases

Repositories of unstructured data types, such as free text, images, audio and video, have been recently emerging in various fields. A general searching approach for such data types is that of similarity search, where the search is for similar objects and similarity is modeled by a metric distance function. In this article we propose a new dynamic paged and balanced access method for similarity search in metric data sets, named CM-tree (Clustered Metric tree). It fully supports dynamic capabilities of insertions and deletions both of single objects and in bulk. Distinctive from other methods, it is especially designed to achieve a structure of tight and low overlapping clusters via its primary construction algorithms (instead of post-processing), yielding significantly improved performance. Several new methods are introduced to achieve this: a strategy for selecting representative objects of nodes, clustering based node split algorithm and criteria for triggering a node split, and an improved sub-tree pruning method used during search. To facilitate these methods the pairwise distances between the objects of a node are maintained within each node. Results from an extensive experimental study show that the CM-tree outperforms the M-tree and the Slim-tree, improving search performance by up to 312% for I/O costs and 303% for CPU costs.

Bulk construction of dynamic clustered metric trees

Repositories of complex data types, such as images, audio, video and free text, are becoming increasingly frequent in various fields. A general searching approach for such data types is that of similarity search, where the search is for similar objects and similarity is modeled by a metric distance function. An important class of access methods for similarity search in metric data is that of dynamic clustered metric trees, where the index is structured as a paged and balanced tree and the space is partitioned hierarchically into compact regions. While access methods of this class allow dynamic insertions typically of single objects, the problem of efficiently inserting a given data set into the index in bulk is largely open. In this article we address this problem and propose novel algorithms corresponding to its two cases, where the index is initially empty (i.e. bulk loading), and where the index is initially non empty (i.e. bulk insertion). The proposed bulk loading algorithm builds the index bottom-up layer by layer, using a new sampling based clustering method, which improves clustering results by improving the quality of the selected sample sets. The proposed bulk insertion algorithm employs the bulk loading algorithm to load the given data into a new index structure, and then merges the new and the existing structures into a unified high quality index, using a novel decomposition method to reduce overlaps between the structures. Both algorithms yield significantly improved construction and search performance, and are applicable to all dynamic clustered metric trees. Results from an extensive experimental study show that the proposed algorithms outperform alternative methods, reducing construction costs by up to 47% for CPU costs and 99% for I/O costs, and search costs by up to 48% for CPU costs and 30% for I/O costs.

Using DSS methods in selecting operations management software

Abstract The multitude of software packages available today for production and operations management creates a selection problem. This paper proposes a methodology and a tool to aid decision makers in evaluating and selecting the most appropriate package for their firm. The methodology deals with manufacturing resource planning (MRP) software and outlines two phases: the enumeration of an extensive list of criteria for MRP and the identification of a decision-making method to be used on the basis of these criteria. The analytic hierarchy process (AHP), also selected by the methodology, is applied to an actual case where a software package is selected for an industrial enterprise. Organizational and human implications resulting from the introduction of an MRP package are discussed.

Synergetic expert systems

Abstract A model for integrating Expert Systems is presented. The model — Synergetic Expert System (SES) — contains several expert systems which can be arranged synergetically to suit the particular needs of a problem. An object-oriented approach is used to design the model and handle its various components. The building blocks of the model, arranged in series or parallel, are defined together with a formal delineation of efficient and economic expert systems. These lead to a definition of marginal cost and value of an expert to a system. The model may be applied when different experts or expert systems are needed to tackle a complex problem. Treating experts or expert systems in parallel may also be viewed as a form of Group Decision Support System (GDSS).