Roy Gelbard

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.

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.

Decision-making method using a visual approach for cluster analysis problems; indicative classification algorithms and grouping scope

: Currently, classifying samples into a fixed number of clusters (i.e. supervised cluster analysis) as well as unsupervised cluster analysis are limited in their ability to support ‘cross-algorithms’ analysis. It is well known that each cluster analysis algorithm yields different results (i.e. a different classification); even running the same algorithm with two different similarity measures commonly yields different results. Researchers usually choose the preferred algorithm and similarity measure according to analysis objectives and data set features, but they have neither a formal method nor tool that supports comparisons and evaluations of the different classifications that result from the diverse algorithms. Current research development and prototype decisions support a methodology based upon formal quantitative measures and a visual approach, enabling presentation, comparison and evaluation of multiple classification suggestions resulting from diverse algorithms. This methodology and tool were used in two basic scenarios: (I) a classification problem in which a ‘true result’ is known, using the Fisher iris data set; (II) a classification problem in which there is no ‘true result’ to compare with. In this case, we used a small data set from a user profile study (a study that tries to relate users to a set of stereotypes based on sociological aspects and interests). In each scenario, ten diverse algorithms were executed. The suggested methodology and decision support system produced a cross-algorithms presentation; all ten resultant classifications are presented together in a ‘Tetris-like’ format. Each column represents a specific classification algorithm, each line represents a specific sample, and formal quantitative measures analyse the ‘Tetris blocks’, arranging them according to their best structures, i.e. best classification.

Visualization of multi-algorithm clustering for better economic decisions - The case of car pricing

Clustering decisions frequently arise in business applications such as recommendations concerning products, markets, human resources, etc. Currently, decision makers must analyze diverse algorithms and parameters on an individual basis in order to establish preferences on the decision-making issues they face; because there is no supportive model or tool which enables comparing different result-clusters generated by these algorithms and parameters combinations. The Multi-Algorithm-Voting (MAV) methodology enables not only visualization of results produced by diverse clustering algorithms, but also provides quantitative analysis of the results. The current research applies MAV methodology to the case of recommending new-car pricing. The findings illustrate the impact and the benefits of such decision support system.

Linking perceived external prestige and collective identification to collaborative behaviors in R&D teams

Research efforts have long been directed at understanding variations in collaborative behaviors among work teams with burgeoning interest in teams operating in knowledge-intensive settings. One of the largely unexplained issues is how does team image and collective identification facilitate collaborative behaviors. Here, survey data were collected from nineteen highly technical work teams engaging in software development in an R&D division of a multinational NASDAQ firm involved in multimedia communications and information processing technology. The relationships between perceived external prestige, collective team identification and team collaborative behaviors were examined. The results of the team-level analyses suggest that perceived external prestige augments collective team identification (measured at Time 1), which in turn engenders a high degree of collaboration and interaction within the team (measured at Time 2). When past team performance was controlled for, the results consistently supported the hypothesized model.

Cluster analysis using multi-algorithm voting in cross-cultural studies

The goal of this study was to overcome three main shortcomings in using a single algorithm to determine a particular clustering of a phenomenon. We addressed this issue by considering cross-cultural research as a case in point and applied Multi-Algorithm Voting (MAV) methodology to cluster analysis. Specifically, this study was designed to provide more systematic supportive decision tools for researchers and managers alike when attempting to cluster analyzing phenomena. To assess the merits of the methodology of MAV for cluster analysis, we analytically examined cross-cultural data from Merritt [Merritt, A. (2000). Culture in the cockpit Do Hofstedes dimensions replicate? Journal of Cross-Cultural Psychology, 31, 283-301] study as well as data scored and ranked by Hofstede [Hofstede, G. (1980). Cultures consequences: International differences in work-related values. Beverly Hills, CA: Sage; Hofstede, G. (1982). Values survey module (Tech. Paper). Maastricht, The Netherlands: Institute for Research on Intercultural Cooperation]. Our study contributes to the literature in several ways. From a methodological point of view, we show how researchers can avoid arbitrary decisions in determining the number of clusters. We provide the researcher with more compelling and robust methodologies not only for analyzing the results of cluster analysis, but also for more better-grounded decision-making through which theoretical insights and implications can be drawn.

Integrated IT management tool kit

The tools track IT assets, performance, budgets, and resources, managing system development and deployment in the interests of the organizations strategic goals.

Optimizing version release dates of research and development long-term processes

This paper develops and compares several optimization approaches for the version planning and release problem. This problem is new, challenging for scholars and practitioners, and was not fully addressed in the OR literature. Version releases are part of a wide-spread phenomenon. Mobile phones, operating systems (e.g. MS-Windows) and digital printers are well known examples. However, version release can be found in many other product development fields, such as software products and games, and hardware versions (e.g. TV, screens, communication equipment etc.). In some fields (such as the automotive field) the version release is so well-established that it became an annual routine.