Visual opinion analysis of customer feedback data
2009-10, Oelke, Daniela, Hao, Ming C., Rohrdantz, Christian, Keim, Daniel A., Dayal, Umeshwar, Haug, Lars-Erik, Janetzko, Halldor
Today, online stores collect a lot of customer feedback in the form of surveys, reviews, and comments. This feedback is categorized and in some cases responded to, but in general it is underutilized even though customer satisfaction is essential to the success of their business. In this paper, we introduce several new techniques to interactively analyze customer comments and ratings to determine the positive and negative opinions expressed by the customers. First, we introduce a new discrimination-based technique to automatically extract the terms that are the subject of the positive or negative opinion (such as price or customer service) and that are frequently commented on. Second, we derive a Reverse-Distance-Weighting method to map the attributes to the related positive and negative opinions in the text. Third, the resulting high-dimensional feature vectors are visualized in a new summary representation that provides a quick overview. We also cluster the reviews according to the similarity of the comments. Special thumbnails are used to provide insight into the composition of the clusters and their relationship. In addition, an interactive circular correlation map is provided to allow analysts to detect the relationships of the comments to other important attributes and the scores. We have applied these techniques to customer comments from real-world online stores and product reviews from web sites to identify the strength and problems of different products and services, and show the potential of our technique.
Document Cards : A Top Trumps Visualization for Documents
2009, Strobelt, Hendrik, Oelke, Daniela, Rohrdantz, Christian, Stoffel, Andreas, Keim, Daniel A., Deussen, Oliver
Finding suitable, less space consuming views for a document s main content is crucial to provide convenient access to large document collections on display devices of different size. We present a novel compact visualization which represents the document s key semantic as a mixture of images and important key terms, similar to cards in a top trumps game. The key terms are extracted using an advanced text mining approach based on a fully automatic document structure extraction. The images and their captions are extracted using a graphical heuristic and the captions are used for a semi-semantic image weighting. Furthermore, we use the image color histogram for classification and show at least one representative from each non-empty image class. The approach is demonstrated for the IEEE InfoVis publications of a complete year. The method can easily be applied to other publication collections and sets of documents which contain images.
Visual Analytics : Combining Automated Discovery with Interactive Visualizations
2008, Keim, Daniel A., Mansmann, Florian, Oelke, Daniela, Ziegler, Hartmut
In numerous application areas fast growing data sets develop with ever higher complexity and dynamics. A central challenge is to filter the substantial information and to communicate it to humans in an appropriate way. Approaches, which work either on a purely analytical or on a purely visual level, do not sufficiently help due to the dynamics and complexity of the underlying processes or due to a situation with intelligent opponents. Only a combination of data analysis and visualization techniques make an effective access to the otherwise unmanageably complex data sets possible.
Visual analysis techniques extend the perceptual and cognitive abilities of humans with automatic data analysis techniques, and help to gain insights for optimizing and steering complicated processes. In the paper, we introduce the basic idea of Visual Analytics, explain how automated discovery and visual analysis methods can be combined, discuss the main challenges of Visual Analytics, and show that combining automatic and visual analysis is the only chance to capture the complex, changing characteristics of the data. To further explain the Visual Analytics process, we provide examples from the area of document analysis.
Exploration of the Local Distribution of Major Ethnic Groups in the USA
2006, Belle, Sebastian Kay, Oelke, Daniela, Oettl, Sonja, Sips, Mike
Knowledge about the local distribution of major ethnic groups in the USA is an important source of information upon which the success of political and economic decisions may depend. Enhancing this information with additional attributes, such as the specific income or spoken languages, reveals interesting aspects on social constellations as well as their interdependencies. The presented visualization facilitates the intuitive exploration of such multidimensional data sets with references to geographical units. Thus, a quick insight into the inherent patterns and characteristic features is provided. Thanks to the high scalability of our visualization technique it can even be used with an iPod-resolution.
Visual Sentiment Analysis of RSS News Feeds Featuring the US Presidential Election in 2008
2009, Wanner, Franz, Rohrdantz, Christian, Mansmann, Florian, Oelke, Daniela, Keim, Daniel A.
The technology behind RSS feeds offers great possibilities to retrieve more news items than ever. In contrast to these technical developments, human capabilities to read all these news items have not increased likewise. To bridge this gap, this paper presents a visual analytics tool for conducting semi-automatic sentiment analysis of large news feeds. While the tool automatically retrieves and analyzes RSS feeds with respect to positive and negative opinion words, the more demanding news analysis of finding trends, spotting peculiarities and putting events into context is left to the human expert. For a solid analysis the news similarity filter enables highlighting of similar or redundant news items. A case study about news related to the US presidential election in 2008 shows how the visual interface of the tool empowers the analyst to draw meaningful conclusions without the effort of reading all news postings.
Visual Evaluation of Text Features for Document Summarization and Analysis
2008-10, Oelke, Daniela, Bak, Peter, Keim, Daniel A., Last, Mark, Danon, Guy
Thanks to the web-related and other advanced technologies, textual information is increasingly being stored in digital form and posted online. Automatic methods to analyze such textual information are becoming inevitable. Many of those methods are based on quantitative text features. Analysts face the challenge to choose the most appropriate features for their tasks. This requires effective approaches for evaluation and feature-engineering. In this paper we suggest an approach to visually evaluate textanalysis features as part of an interactive feedback loop between evaluation and feature engineering. We apply documentfingerprinting for visualizing text features as an integral part of the analytic process. Consequently, analysts are able to access interim results of the applied automatic methods and alter their properties to achieve better results.
We implement and evaluate the methodology on two different tasks, namely opinion analysis and document summarization and show that our iterative method leads to improved performance.
Literature Fingerprinting : A New Method for Visual Literary Analysis
2007-10, Keim, Daniel A., Oelke, Daniela
In computer-based literary analysis different types of features are used to characterize a text. Usually, only a single feature value or vector is calculated for the whole text. In this paper, we combine automatic literature analysis methods with an effective visualization technique to analyze the behavior of the feature values across the text. For an interactive visual analysis, we calculate a sequence of feature values per text and present them to the user as a characteristic fingerprint. The feature values may be calculated on different hierarchy levels, allowing the analysis to be done on different resolution levels. A case study shows several successful applications of our new method to known literature problems and demonstrates the advantage of our new visual literature fingerprinting.
Large-scale Comparative Sentiment Analysis of News Articles
2009, Wanner, Franz, Rohrdantz, Christian, Mansmann, Florian, Stoffel, Andreas, Oelke, Daniela, Krstajic, Milos, Keim, Daniel A., Luo, Dongning, Yang, Jing, Atkinson, Martin
Online media offers great possibilities to retrieve more news items than ever. In contrast to these technical developments, human capabilities to read all these news items have not increased likewise. To bridge this gap, this poster presents a visual analytics tool for conducting semi-automatic sentiment analysis of large news feeds. The tool retrieves and analyzes the news of two categories (Terrorist Attack and Natural Disasters) and news which belong to both categories of the Europe Media Monitor (EMM) with respect to positive and negative opinion words. While this happens automatically, the more demanding news analysis of finding trends, spotting peculiarities and putting events into context is left to the human expert.
Density Displays for Data Stream Monitoring
2008-05, Hao, Ming C., Keim, Daniel A., Dayal, Umeshwar, Oelke, Daniela, Tremblay, Chantal
In many business applications, large data workloads such as sales figures or process performance measures need to be monitored in real-time. The data analysts want to catch problems in flight to reveal the root cause of anomalies. Immediate actions need to be taken before the problems become too expensive or consume too many resources. In the meantime, analysts need to have the big picture of what the information is about. In this paper, we derive and analyze two real-time visualization techniques for managing density displays: (1) circular overlay displays which visualize large volumes of data without data shift movements after the display is full, thus freeing the analyst from adjusting the mental picture of the data after each data shift; and (2) variable resolution density displays which allow users to get the entire view without cluttering. We evaluate these techniques with respect to a number of evaluation measures, such as constancy of the display and usage of display space, and compare them to conventional displays with periodic shifts. Our real time data monitoring system also provides advanced interactions such as a local root cause analysis for further exploration. The applications using a number of real-world data sets show the wide applicability and usefulness of our ideas.
Finding Correlations in Functionally Equivalent Proteins by Integrating Automated and Visual Data Exploration
2006-10, Keim, Daniel A., Oelke, Daniela, Truman, Royal, Neuhaus, Klaus
The analysis of alignments of functionally equivalent proteins can reveal regularities such as correlated positions or residue patterns which are important to ensure a specific fold and various cellular functions. Many approaches are found in the literature which try to identify correlated positions to predict the residues that are close to each other in the three-dimensional folded structure. However, the quality of the predictions remains disappointing. One of the problems is that the statistical correlation measures that were used cannot do justice to the underlying complex biological and physicochemical realities. In this paper we evaluate the biological requirements for a correlation measure and explain why a completely automatic approach is unlikely to succeed. We then propose a novel and flexible criteria for correlation of residue positions in protein sequences, which can be optimized for different requirements. To apply this definition we developed the tool VisAlign that combines an automatic calculation of correlations with an interactive visualization. This allows the user to visually explore alternative alignments and thereby conveniently test various hypothesis and to detect regularities in the aligned sequences