Adalbert F. X. Wilhelm

Visualizing association rules with interactive mosaic plots

Association rules are amongst the most important patterns that can be discovered using data mining. Their automatic discovery is supported by most, if not all, data mining software tools. Moreover, many techniques have been devised to lter the most interesting (in many senses) rules from the complete set of discovered rules, such that the users are not swamped by results. However, association rules are actually hard to understand; even more so if one only looks at the most interesting rules. For example, rather strong correlations between attributes are not always obvious from the discovered rules. Similarly, a deeper explanation of related association rules may be missing from the rule set. In this paper we show how Mosaic plots and, especially, their variant called Double Decker plots, can be used to visualize association rules. These plots visualize the contingency table that yields the association rule as well as the other potential rules in that table, whether they meet the thresholds or not. This gives a deeper understanding on the nature of the correlation between the left-hand side of the rule and the right-hand side. Moreover, we show how an interactive use of these plots helps the user to understand the relationship between related association rules.

Visual clustering and classification: The Oronsay particle size data set revisited

SummaryInteractive statistical graphics can be effectively used to find natural groupings in observations. In this paper we want to demonstrate how clustering and classification can be done with three approaches based on highly interactive graphical environments: high-dimensional scatterplots as available in XGobi, parallel coordinate plots as available in EXPLORN, and linked low-dimensional views as available in MANET. We will point out the strenghts and the weaknesses of these techniques by comparing their behaviour when applied to the Oronsay particle size data set.

Measuring effects of voluntary attention: A comparison among predictive arrow, colour, and number cues

The Posner cueing paradigm is one of the most widely used paradigms in attention research. Importantly, when employing it, it is critical to understand which type of orienting a cue triggers. It has been suggested that large effects elicited by predictive arrow cues reflect an interaction of involuntary and voluntary orienting. This conclusion is based on comparisons of cueing effects of predictive arrows, nonpredictive arrows (involuntary orienting), and predictive numbers (voluntary orienting). Experiment 1 investigated whether this conclusion is restricted to comparisons with number cues and showed similar results to those of previous studies, but now for comparisons to predictive colour cues, indicating that the earlier conclusion can be generalized. Experiment 2 assessed whether the size of a cueing effect is related to the ease of deriving direction information from a cue, based on the rationale that effects for arrows may be larger, because it may be easier to process direction information given by symbols such as arrows than that given by other cues. Indeed, direction information is derived faster and more accurately from arrows than from colour and number cues in a direction judgement task, and cueing effects are larger for arrows than for the other cues. Importantly though, performance in the two tasks is not correlated. Hence, the large cueing effects of arrows are not a result of the ease of information processing, but of the types of orienting that the arrows elicit.

An empirical study of factors impacting on knowledge processes in online forums: factors of interest and model outline

Based on a literature review of the fields of research of knowledge management, collaborative systems and online communities, this paper presents an outline of factors expected to contribute to collaborative knowledge creation and knowledge transfer processes in online forums, and proposes corresponding hypotheses. The paper depicts the models underlying framework and its general methodological approach, describes the method for the indirect derivation of individual actors knowledge process contributions, and expands in detail on input factors expected to contribute to knowledge process relevant contributions within online forums. Finally, the integration of these factors into a common Measurement and Structural Model is described. The research project is currently being conducted at International University Bremen (IUB).

Projection-Based Partitioning for Large, High-Dimensional Datasets

Recent work in the field of cluster analysis has focused on designing algorithms that address the issue of ever growing datasets and provide meaningful solutions for data with high cardinality and/or dimensionality, under the natural restriction of limited resources. Within this line of research, we propose a method drawing on the principles of projection pursuit and grid partitioning, which focuses on reducing computational requirements for large datasets without loss of performance. To achieve that, we rely on procedures such as sampling of objects, feature selection, and quick density estimation using histograms. The present algorithm searches for low-density points in potentially favorable one-dimensional projections, and partitions the data by a hyperplane passing through the best split point found. Tests on synthetic and reference data indicate that our method can quickly and efficiently recover clusters that are distinguishable from the remaining objects on at least one direction; linearly nonseparable clusters are usually subdivided. The solution is robust in the presence of noise in moderate levels, and when the clusters are partially overlapping. An implementation of the algorithm is available online, as supplemental material.

An empirical study of factors impacting on knowledge processes in online forums: structural equation modelling analysis and results

This paper presents an empirical, Structural Equation Modelling (SEM)-based case study on the identification of factors contributing to collaborative knowledge creation and knowledge transfer processes in forum-based Online Knowledge Communities (OKCs), recently conducted at the International University Bremen (IUB). Following a brief outline of the theoretical background of the study with an outline of individual hypotheses, the paper expands on the methodology of the study. Following the validation of constructs and the validation of the measurement model, structural equation modelling is applied. The paper summarises with a detailed discussion of the individual path value results. The paper continues a prior IJWBC paper by Schmitz-Justen and Wilhelm (2006) and expands on a recent conference publication by Schmitz-Justen and Wilhelm (2005c), which addressed the first empirical results of the project.

Validation of association rules by interactive mosaic plots

Association Rules have been proposed by Agrawal et al. (1993) in the context of market basket analysis. They were invented to provide an automated process, which could find connections among items, that were not known before, especially to answer questions like: “which items are likely to be bought together?”. Typically, the data to be examined consists of customer purchases, i.e. a set of items bought by a customer over a period of time. The standard way of storing such data is the following: To be able to identify each customer the transactions are stored with unique numbers, the transaction identification (TID). Beside that, we have a set of different items, the so-called itemset. \(\mathcal{I} = \{ {i_1},{i_2}, \ldots,{i_m}\}\). The data or database D is a set of purchases (transactions), where each transaction T includes a set of items, such that \(T \subset \mathcal{I}\). A transaction T is said to contain a set of items X, if X is a subset of T.

Analysis of Dialect Features

Interactive graphical analysis has proven to be a seminal method in exploring quantitative spatial data. Using data from a dialect survey we demonstrate that it can also reveal structure in qualitative spatial data. The flexibility offered by such exploratory tools can deal with the special problems caused by multiple responses and by local geographic effects.

Interactive Graphics and Local Statistics

Exploratory spatial data analysis has a series of aims including determining spatial structure in the data, describing and visualizing geographical distributions, exploring spatial dependencies, measuring heterogeneity and identifying outliers. To quantify these phenomena a rich variety of statistics has been proposed. Standard methods use all the data for the entire area under study, yet this area has usually been arbitrarily bounded and may include quite distinctive geographic features. Local statistics are relatively independent of the global boundaries and they attempt to quantify how close a given datum is to the values in its neighbourhood. Since each local statistic focuses on slightly different aspects of the data the use of more than one is suggested. Interactive graphics methods help to link the information from different local statistics and dynamic tools can be used to visualize the effects of changing the neighbourhood definition.

Rank-based classifiers for extremely high-dimensional gene expression data

Predicting phenotypes on the basis of gene expression profiles is a classification task that is becoming increasingly important in the field of precision medicine. Although these expression signals are real-valued, it is questionable if they can be analyzed on an interval scale. As with many biological signals their influence on e.g. protein levels is usually non-linear and thus can be misinterpreted. In this article we study gene expression profiles with up to 54,000 dimensions. We analyze these measurements on an ordinal scale by replacing the real-valued profiles by their ranks. This type of rank transformation can be used for the construction of invariant classifiers that are not affected by noise induced by data transformations which can occur in the measurement setup. Our 10