Caroline Ziemkiewicz

WireVis: Visualization of Categorical, Time-Varying Data From Financial Transactions

Large financial institutions such as Bank of America handle hundreds of thousands of wire transactions per day. Although most transactions are legitimate, these institutions have legal and financial obligations in discovering those that are suspicious. With the methods of fraudulent activities ever changing, searching on predefined patterns is often insufficient in detecting previously undiscovered methods. In this paper, we present a set of coordinated visualizations based on identifying specific keywords within the wire transactions. The different views used in our system depict relationships among keywords and accounts over time. Furthermore, we introduce a search-by-example technique which extracts accounts that show similar transaction patterns. In collaboration with the Anti-Money Laundering division at Bank of America, we demonstrate that using our tool, investigators are able to detect accounts and transactions that exhibit suspicious behaviors.

The Shaping of Information by Visual Metaphors

The nature of an information visualization can be considered to lie in the visual metaphors it uses to structure information. The process of understanding a visualization therefore involves an interaction between these external visual metaphors and the users internal knowledge representations. To investigate this claim, we conducted an experiment to test the effects of visual metaphor and verbal metaphor on the understanding of tree visualizations. Participants answered simple data comprehension questions while viewing either a treemap or a node-link diagram. Questions were worded to reflect a verbal metaphor that was either compatible or incompatible with the visualization a participant was using. The results suggest that the visual metaphor indeed affects how a user derives information from a visualization. Additionally, we found that the degree to which a user is affected by the metaphor is strongly correlated with the users ability to answer task questions correctly. These findings are a first step towards illuminating how visual metaphors shape user understanding, and have significant implications for the evaluation, application, and theory of visualization.

Do Mechanical Turks dream of square pie charts

Online studies are an attractive alternative to the laborintensive lab study, and promise the possibility of reaching a larger variety and number of people than at a typical university. There are also a number of draw-backs, however, that have made these studies largely impractical so far. Amazons Mechanical Turk is a web service that facilitates the assignment of small, web-based tasks to a large pool of anonymous workers. We used it to conduct several perception and cognition studies, one of which was identical to a previous study performed in our lab. We report on our experiences and present ways to avoid common problems by taking them into account in the study design, and taking advantage of Mechanical Turks features.

Legible Simplification of Textured Urban Models

Most of the algorithms used for research in mesh simplification and discrete levels of detail (LOD) work well for simplifying single objects with a large number of polygons. For a city-sized collection of simple buildings, using these traditional algorithms could mean the disappearance of an entire residential area in which the buildings tend to be smaller than those in commercial regions. To solve this problem, we developed a mesh-simplification algorithm that incorporates concepts from architecture and city planning. Specifically, we rely on the concept of urban legibility, which segments a city into paths, edges, districts, nodes, and landmarks. If we preserve these elements of legibility during the simplification process, we can maintain the citys image and create urban models that users can understand more effectively. To accomplish this goal, we divide our algorithm into five steps. During preprocessing, it performs hierarchical clustering, cluster merging, model simplification, and hierarchical texturing, at runtime, it employs LOD to select the appropriate models for rendering.

Preconceptions and individual differences in understanding visual metaphors

Understanding information visualization is more than a matter of reading a series of data values; it is also a matter of incorporating a visual structure into ones own thinking about a problem. We have proposed visual metaphors as a framework for understanding high‐level visual structure and its effect on visualization use. Although there is some evidence that visual metaphors can affect visualization use, the nature of this effect is still ambiguous. We propose that a users preconceived metaphors for data and other individual differences play an important role in her ability to think in a variety of visual metaphors, and subsequently in her ability to use a visualization. We test this hypothesis by conducting a study in which a participants preconceptions and thinking style were compared with the degree to which she is affected by conflicting metaphors in a visualization and its task questions. The results show that metaphor compatibility has a significant effect on accuracy, but that factors such as spatial ability and personality can lessen this effect. We also find a complex influence of self‐reported metaphor preference on performance. These findings shed light on how people use visual metaphors to understand a visualization.

Embedding Information Visualization within Visual Representation

Despite its often technical nature, visualization is in many ways a form of visual representation. Just how visualization relates to illustration, information graphics, digital art, visual languages, etc., is nonetheless poorly understood. We propose a theory that embeds information visualizationwithin other visual traditions in terms of criteria that are not purely technical: dependence on data, mapping, interactivity, and notationality. In addition to providing the means for a classification, these criteria also foster a different understanding of information visualization. We further adapt our criteria to differentiate within visualization, using mapping, readability and information loss, and notationality as the criteria. Both sets of criteria are demonstrated in a number of case studies.We believe that our novel taxonomies of visualization methods serve as a step towards a more comprehensive theoretical context to understanding the essential purposes, properties, and functions of information visualization.

Scable and interactive visual analysis of financal wire transactions for fraud detection

Large financial institutions such as Bank of America handle hundreds of thousands of wire transactions per day. Although most transactions are legitimate, these institutions have legal and financial obligations to discover those that are suspicious. With the methods of fraudulent activities ever changing, searching on predefined patterns is often insufficient in detecting previously undiscovered methods. In this paper, we present a set of coordinated visualizations based on identifying specific keywords within the wire transactions. The different views used in our system depict relationships among keywords and accounts over time. Furthermore, we introduce a search-by-example technique, which extracts accounts that show similar transaction patterns. Our system can be connected to a database to handle millions of transactions and still preserve high interactivity. In collaboration with the Anti-Money Laundering division at Bank of America, we demonstrate that using our tool, investigators are able to detect accounts and transactions that exhibit suspicious behaviors.

Understanding Visualization by Understanding Individual Users

Visualization is often seen as a tool to support complex thinking. Although different people can have very different ways of approaching the kind of complex task that visualizations support, as researchers and designers we still rarely consider individual differences in creating and evaluating visualizations. This article reviews recent research on individual differences in visualization and human-computer interaction, showing that both cognitive abilities and personality profiles might significantly affect performance with these tools. The study of individual differences has led to the conclusion that advances in this important area in visualization will require more focused research. Specifically, we must isolate the cognitive factors that are relevant to visualization and the design factors that make one visualization more suited to a user than another. In doing so, we could increase our understanding of the visualization user and reshape how we approach design and evaluation.

Implied dynamics in information visualization

Information visualization is a powerful method for understanding and working with data. However, we still have an incomplete understanding of how people use visualization to think about information. We propose that people use visualization to support comprehension and reasoning by viewing abstract visual representations as physical scenes with a set of implied dynamics between objects. Inferences based on these implied dynamics are metaphorically extended to form inferences about the represented information. This view predicts that even seemingly meaningless properties of a visualization, including such minor design elements as borders, background areas, and the connectedness of parts, may affect how people perceive semantic aspects of data by suggesting different potential dynamics between data points. We present a study that supports this claim and discuss the design implications of this theory of information visualization.

Laws of Attraction: From Perceptual Forces to Conceptual Similarity

Many of the pressing questions in information visualization deal with how exactly a user reads a collection of visual marks as information about relationships between entities. Previous research has suggested that people see parts of a visualization as objects, and may metaphorically interpret apparent physical relationships between these objects as suggestive of data relationships. We explored this hypothesis in detail in a series of user experiments. Inspired by the concept of implied dynamics in psychology, we first studied whether perceived gravity acting on a mark in a scatterplot can lead to errors in a participants recall of the marks position. The results of this study suggested that such position errors exist, but may be more strongly influenced by attraction between marks. We hypothesized that such apparent attraction may be influenced by elements used to suggest relationship between objects, such as connecting lines, grouping elements, and visual similarity. We further studied what visual elements are most likely to cause this attraction effect, and whether the elements that best predicted attraction errors were also those which suggested conceptual relationships most strongly. Our findings show a correlation between attraction errors and intuitions about relatedness, pointing towards a possible mechanism by which the perception of visual marks becomes an interpretation of data relationships.