Reinhold Decker

Data Analysis, Machine Learning, and Applications

We consider distance-based similarity measures for real-valued vectors of interest in kernel-based machine learning algorithms. In particular, a truncated Euclidean similarity measure and a self-normalized similarity measure related to the Canberra distance. It is proved that they are positive semi-definite (p.s.d.), thus facilitating their use in kernel-based methods, like the Support Vector Machine, a very popular machine learning tool. These kernels may be better suited than standard kernels (like the RBF) in certain situations, that are described in the paper. Some rather general results concerning positivity properties are presented in detail as well as some interesting ways of proving the p.s.d. property.

Advances in Data Analysis

Clustering.- Mixture Models for Classification.- How to Choose the Number of Clusters: The Cramer Multiplicity Solution.- Model Selection Criteria for Model-Based Clustering of Categorical Time Series Data: A Monte Carlo Study.- Cluster Quality Indexes for Symbolic Classification - An Examination.- Semi-Supervised Clustering: Application to Image Segmentation.- A Method for Analyzing the Asymptotic Behavior of the Walk Process in Restricted Random Walk Cluster Algorithm.- Cluster and Select Approach to Classifier Fusion.- Random Intersection Graphs and Classification.- Optimized Alignment and Visualization of Clustering Results.- Finding Cliques in Directed Weighted Graphs Using Complex Hermitian Adjacency Matrices.- Text Clustering with String Kernels in R.- Automatic Classification of Functional Data with Extremal Information.- Typicality Degrees and Fuzzy Prototypes for Clustering.- On Validation of Hierarchical Clustering.- Classification.- Rearranging Classified Items in Hierarchies Using Categorization Uncertainty.- Localized Linear Discriminant Analysis.- Calibrating Classifier Scores into Probabilities.- Nonlinear Support Vector Machines Through Iterative Majorization and I-Splines.- Deriving Consensus Rankings from Benchmarking Experiments.- Classification of Contradiction Patterns.- Selecting SVM Kernels and Input Variable Subsets in Credit Scoring Models.- Data and Time Series Analysis.- Simultaneous Selection of Variables and Smoothing Parameters in Geoadditive Regression Models.- Modelling and Analysing Interval Data.- Testing for Genuine Multimodality in Finite Mixture Models: Application to Linear Regression Models.- Happy Birthday to You, Mr. Wilcoxon!.- Equivalent Number of Degrees of Freedom for Neural Networks.- Model Choice for Panel Spatial Models: Crime Modeling in Japan.- A Boosting Approach to Generalized Monotonic Regression.- From Eigenspots to Fisherspots - Latent Spaces in the Nonlinear Detection of Spot Patterns in a Highly Varying Background.- Identifying and Exploiting Ultrametricity.- Factor Analysis for Extraction of Structural Components and Prediction in Time Series.- Classification of the U.S. Business Cycle by Dynamic Linear Discriminant Analysis.- Examination of Several Results of Different Cluster Analyses with a Separate View to Balancing the Economic and Ecological Performance Potential of Towns and Cities.- Visualization and Scaling Methods.- VOS: A New Method for Visualizing Similarities Between Objects.- Multidimensional Scaling of Asymmetric Proximities with a Dominance Point.- Single Cluster Visualization to Optimize Air Traffic Management.- Rescaling Proximity Matrix Using Entropy Analyzed by INDSCAL.- Information Retrieval, Data and Web Mining.- Canonical Forms for Frequent Graph Mining.- Applying Clickstream Data Mining to Real-Time Web Crawler Detection and Containment Using ClickTips Platform.- Plagiarism Detection Without Reference Collections.- Putting Successor Variety Stemming to Work.- Collaborative Filtering Based on User Trends.- Investigating Unstructured Texts with Latent Semantic Analysis.- Marketing, Management Science and Economics.- Heterogeneity in Preferences for Odd Prices.- Classification of Reference Models.- Adaptive Conjoint Analysis for Pricing Music Downloads.- Improving the Probabilistic Modeling of Market Basket Data.- Classification in Marketing Research by Means of LEM2-generated Rules.- Pricing Energy in a Multi-Utility Market.- Disproportionate Samples in Hierarchical Bayes CBC Analysis.- Building on the Arules Infrastructure for Analyzing Transaction Data with R.- Balanced Scorecard Simulator - A Tool for Stochastic Business Figures.- Integration of Customer Value into Revenue Management.- Womens Occupational Mobility and Segregation in the Labour Market: Asymmetric Multidimensional Scaling.- Multilevel Dimensions of Consumer Relationships in the Healthcare Service Market M-L IRT vs. M-L SEM Approach.- Data Mining in Higher Education.- Attribute Aware Anonymous Recommender Systems.- Banking and Finance.- On the Notions and Properties of Risk and Risk Aversion in the Time Optimal Approach to Decision Making.- A Model of Rational Choice Among Distributions of Goal Reaching Times.- On Goal Reaching Time Distributions Estimated from DAX Stock Index Investments.- Credit Risk of Collaterals: Examining the Systematic Linkage between Insolvencies and Physical Assets in Germany.- Foreign Exchange Trading with Support Vector Machines.- The Influence of Specific Information on the Credit Risk Level.- Bio- and Health Sciences.- Enhancing Bluejay with Scalability, Genome Comparison and Microarray Visualization.- Discovering Biomarkers for Myocardial Infarction from SELDI-TOF Spectra.- Joint Analysis of In-situ Hybridization and Gene Expression Data.- Unsupervised Decision Trees Structured by Gene Ontology (GO-UDTs) for the Interpretation of Microarray Data.- Linguistics and Text Analysis.- Clustering of Polysemic Words.- Classifying German Questions According to Ontology-Based Answer Types.- The Relationship of Word Length and Sentence Length: The Inter-Textual Perspective.- Comparing the Stability of Different Clustering Results of Dialect Data.- Part-of-Speech Discovery by Clustering Contextual Features.- Statistical Musicology and Sound Classification.- A Probabilistic Framework for Audio-Based Tonal Key and Chord Recognition.- Using MCMC as a Stochastic Optimization Procedure for Monophonic and Polyphonic Sound.- Vowel Classification by a Neurophysiologically Parameterized Auditory Model.- Archaeology.- Uncovering the Internal Structure of the Roman Brick and Tile Making in Frankfurt-Nied by Cluster Validation.- Where Did I See You Before... A Holistic Method to Compare and Find Archaeological Artifacts.

Measuring Consumer Preferences for Complex Products: A Compositional Approach Based on Paired Comparisons

Conjoint analysis has become a widely accepted tool for preference measurement in marketing research, though its applicability and performance strongly depend on the complexity of the product or service. Therefore, self-explicated approaches are still frequently used because of their simple design, which facilitates preference elicitation when large numbers of attributes need to be considered. However, the direct measurement of preferences, or rather utilities, has been criticized as being imprecise in many cases. Against this background, the authors present a compositional consumer preference measurement approach based on paired comparisons, otherwise known as PCPM. The trade-off character of paired comparisons ensures that the stated judgments are more intuitive than traditional self-explicated preference statements. In contrast to the latter, PCPM accounts for response errors and thus allows for the elicitation of more precise preferences. The authors benchmark PCPM against adaptive conjoint analysis and computer-assisted self-explication of multiattributed preferences to demonstrate its relative validity and predictive accuracy in two empirical studies using complex, high-involvement products. They find that PCPM yields better results than the benchmark approaches with respect to interview length, individual hit rates, and aggregate choice share predictions.

An internet‐based approach to environmental scanning in marketing planning

Purpose – This paper introduces a new approach for autonomous internet‐based environmental scanning, which combines concept of weak signals with “information foraging theory”.Design/methodology/approach – Early detection and rapid action with respect to developments in the operating environment is a prerequisite for successful marketing planning. Accordingly, this paper proposes a three‐stage process for overcoming practical obstacles to the detection and use of weak signals from the operating environment, in particular how to identify relevant and useful documents in harsh information environments such as the internet. Its functionality is demonstrated by means of a human‐machine experiment.Findings – A framework based on information foraging theory is well suited to the task of determining the relevance of documents and facilitates the automation of information search processes. A prototype environmental scanning system of this type outperformed human experts in a typical scanning task.Research limitati...

A survey of the challenges and pifalls of cluster analysis application in market segmentation

Market segmentation is a widely accepted concept in marketing research and planning. Although cluster analysis has been extensively applied to segment markets in the last 50 years, the ways in which the results were obtained have often been reported to be less than satisfactory by both practitioners (Yankelovich & Meer 2006) and academics (Dolnièar 2003). In order to provide guidance to those undertaking market segmentation, this study discusses the critical issues involved when using cluster analysis to segment markets, makes suggestions for best practices and potential improvements, and presents an empirical survey that seeks to provide an up-to-date assessment of cluster analysis application in market segmentation within a six-stage framework. Analyses of more than 200 journal articles published since 2000, in which cluster analysis was empirically used in a marketing research setting, indicate that many critical issues are still ignored rather than addressed adequately.

Eye-tracking Information Processing in Choice-based Conjoint Analysis

Choice models are a common tool in market research for quantifying the influence of product attributes on consumer decisions. Process tracing techniques, on the other hand, try to answer the question of how people process information and make decisions in choice tasks. This paper suggests a combination of both approaches for in-depth investigations of consumer decision processes in preference measurement by means of choice-based conjoint (CBC) analysis. We discuss different process tracing techniques and propose an attribute-specific strategy measure for the analysis of CBC results. In our empirical study we eye-track respondents evaluating CBC choice tasks for single-cup coffee brewers. On the basis of several hypotheses we illustrate the benefits of simultaneously recording eye-tracking information for market research.

Classification and Information Processing at the Turn of the Millennium

Data Analysis and Classification: G. Arminger, J. Wittenberg, P. Stein: New Developments in Latent Variable Models.- H.-H. Bock: Regression-Type Models for Kohonens Self-Organizing Networks.-J. Meulman: Relational Data Analysis, Discrimination and Classification.- S. Nishisato: Data Types and Information: Beyond the Current Practice of Data Analysis.-A. Okada, T. Imaizumi: Two-Mode Three-Way Asymmetric Multidimensional Scaling with Constraints on Asymmetry.-H. Bensmail, J. Meulman: Regularized Discriminant Analysis with Optimally Scaled Data.-W. Castillo, J. Trejos: Recurrence Properties in Two-Mode Hierarchical Clustering.-J. Conde, M. Colomer, C. Capdevila, A. Gil: On Pyramidal Classification in the Delimitation of Climatic Regions in Catalonia.-F. Dau, R. Wille: On the Modal Understanding of Triadic Contexts.-S. Gabler, J. Blasius: Clustering and Scaling: Grouping Variables in Burt Matrices.-K. Jajuga, M. Walesiak: Standardisation of Data Set under Different Mesurement Scales.-P. Lory, D. Gietl: Neural Networks for Two-Group Classification Problems with Monotonicity Hints.-R. Pfluger, O. Gefeller: A Bridge from the Nearest Neighbour to the Fixed Bandwidth in Nonparametric Functional Estimation.-W. Theis, C. Weihs: Clustering Techniques for the Detection of Business Cycles.- J. Trejos, W. Castillo: Simulated Annealing Optimization for Two-mode Partitioning.-Computer Science, Computational Statistics, and Data Mining: W. Hauke: Competiveness of Evolutionary Algorithms.-J. Kacprzyk: Intelligent Data Analysis via Linguistic Data Summaries: A fuzzy Logic Approach.-G. Lindner, R. Studer: Algorithm Selection Support for Classification.-K. Mainzer: Society of Knowledge - Interdisciplinary Perspectives of Information Processing in Virtual Networks.-P. Naeve: What is Computational Statistics?.-M.Feldmann: A Development Framework for Nature Analogic Heuristics.-C. Heidsiek, W. Uhr: Systematizing and Evaluating Data Mining Methods.-T. Hermes, A. Miene, P. Kreyenhop: On Textures: A Sketch of a Texture-Based Image Segmentation Approach.-T. Hermes, A. Miene, O. Moehrke: Automatic Texture Classification by Visual Properties.-S. Kuhlins, A. Korthaus: Java Servlets versus CGI - Implications for Remote Data Analysis.-D. Pechtel, K.-D. Kuhnert: Generating Automatically Local Feature Groups of Complex and Deformed Objects.-F. Sauberlich, W. Gaul: Decision Tree Construction by Association Rules.-B. Stein, O. Niggemann, U. Husemeyer: Learning Complex Similarity Measures.-M. Thomas, R.-D. Reiss: Graphical Programming in Statistics: The XGPL Prototype.-Management Science, Marketing, and Finance: U. Wagner, C. Boyer: Measuring Brand Loyalty on the Individual Level: a Comparative Study.-U. Bankhofer: Facility Location Planning with Qualitative Location Factors.-K. Bartels, Y. Boztug, M. Muller: Testing the Multinomial Logit Model.-T. Burkhardt: Time Optimal Portfolio Selection: Mean-variance-efficient Sets for Arithmetic and Geometric Brownian Price Processes.-N. Gorz, L. Hildebrandt, D. Annacker: Analyzing Multigroup Data with Structural Equation Models.-S. Hensel-Borner, H. Sattler: Validity of Customized and Adaptive Hybrid Conjoint Analysis.-Y. Kimura, A. Okada: Analysis of the Assessments of Ones Values Among Different Cohorts in Age and Sex by Multidimensional Scaling.-J. Limperger: Impacts of Hedging with Futures on Optimal Production Levels.-B. Nietert: How to Integrate Stock Price Jumps into Portfolio Selection.-W. Polasek, L. Ren: A Multivariate GARCH-M Model for Exchange Rates in the US, Germany and Japan.-A. Tangian: A Model for Constructing Quadratic

The Number of Clusters in Market Segmentation

Learning the ‘true’ number of clusters in a given data set is a fundamental and largely unsolved problem in data analysis, which seriously affects the identification of customer segments in marketing research.

AHP versus ACA – An Empirical Comparison

The Analytic Hierarchy Process (AHP) has been of substantial impact in business research and particularly in managerial decision making for a long time. Although empirical investigations (e.g. Scholl et al. (2005)) and simulation studies (e.g. Scholz et al. (2006)) have shown its general potential in consumer preference measurement, AHP is still rather unpopular in marketing research.

Platform Launch Strategies

Today, digital platforms mediating between independent groups of users account for a total market value of about US-