Many information visualization techniques map abstract data into visual representations to present the underlying information in a more understandable way. However, when data is not available, automatic methods are necessary to extract the underlying data from the chart images. This paper focuses on chart image classification, which is a fundamental step of chart analysis and data extraction. The performance of most conventional classifiers depends on the feature extraction technique applied, usually requiring a domain-specific design for creating the features. In this context, Convolutional Neural Network (CNN) has gained attention not only because it has achieved good results in many computer vision tasks but also because it can learn representations of images without the design of a feature extractor specific for a domain problem. Thereby, this paper proposes an evaluation of CNN architectures (VGG19, Resnet-50, and Inception-V3), comparing them with conventional classifiers (HOG features combined with KNN, Naive Bayes, Random Forest and Support Vector Machine). An artificially generated dataset of 10 classes of chart images was used for training and testing. Furthermore, another dataset composed of chart images collected from the internet was used for validation. The results show that Resnet-50 and Inception-V3 performed best achieving an accuracy of 77.76% and 76.77% respectively on the validation set. The CNN models outperformed the conventional methods, where the best combination was HOG+SVM, achieving 45.03 %. Moreover, some similar chart types were expected to be misclassified, and they were discussed based on the confusion matrix.

Data charts are widely used in our daily lives, being present in regular media, such as newspapers, magazines, web pages, books, and many others. In general, a well-constructed data chart leads to an intuitive understanding of its underlying data. In the same way, when data charts have wrong design choices, a redesign of these representations might be needed. However, in most cases, these charts are shown as a static image, which means that the original data are not usually available. Therefore, automatic methods could be applied to extract the underlying data from the chart images to allow these changes. The task of recognizing charts and extracting data from them is complex, largely due to the variety of chart types and their visual characteristics. Other features in real-world images that can make this task difficult are photo distortions, noise, alignment, etc. Two computer vision techniques that can assist this task and have been little explored in this context are perspective detection and correction. These methods transform a distorted and noisy chart in a clear chart, with its type ready for data extraction or other uses. This paper proposes a classification, detection, and perspective correction process that is suitable for real-world usage, when considering the data used for training a state-of-the-art model for the extraction of a chart in real-world photography. The results showed that, with slight changes, chart recognition methods are now ready for real-world charts, when taking time and accuracy into consideration.

This paper presents UXmood, a tool that provides quantitative and qualitative information to assist researchers and practitioners in the evaluation of user experience and usability. The tool uses and combines data from video, audio, interaction logs and eye trackers, presenting them in a configurable dashboard on the web. The UXmood works analogously to a media player, in which evaluators can review the entire user interaction process, fast-forwarding irrelevant sections and rewinding specific interactions to repeat them if necessary. Besides, sentiment analysis techniques are applied to video, audio and transcribed text content to obtain insights on the user experience of participants. The main motivations to develop UXmood are to support joint analysis of usability and user experience, to use sentiment analysis for supporting qualitative analysis, to synchronize different types of data in the same dashboard and to allow the analysis of user interactions from any device with a web browser. We conducted a user study to assess the data communication efficiency of the visualizations, which provided insights on how to improve the dashboard.

Different information visualization techniques can be found in the literature due to the quantity and variety of data stored in computational systems. In this context, the classification of chart images becomes important because it allows various types of graphs to be detected automatically in different contexts, allowing a more specific processing for each type of visualization, for example, data extraction. Several techniques of image classification can be used, where the most common are based on the extraction of features of the images, and a later classification using these features. However, one technique that has been gaining prominence in the context of image classification is the Convolutional Neural Network (CNN). This technique is based on deep learning and, in a way, encapsulates the feature extraction process. In this way, the proposal of this article is to use an architecture of a client-server based model to do the chart image classification and later data extraction from this image. The main advantage is doing the CNN processing on the server side, so the application does not rely on client device limitations. For this, an image dataset was generated from the web, and it has ten classes of graphs. From the experiments done, it was seen that the use of this technique was feasible, and modifications in the architecture can be made as a proposal to improve the accuracy of the model.

Data generators are applications that produce synthetic datasets, which are useful for testing data analytics applications, such as machine learning algorithms and information visualization techniques. Each data generator application has a different approach to generate data. Consequently, each one has functionality gaps that make it unsuitable for some tasks (e.g., lack of ways to create outliers and non-random noise). This paper presents a data generator application that aims to fill relevant gaps scattered across other applications, providing a flexible tool to assist researchers in exhaustively testing their techniques in more diverse ways. The proposed system allows users to define and compose known statistical distributions to produce the desired outcome, visualizing the behavior of the data in real-time to analyze if it has the characteristics needed for efficient testing. This paper presents in detail the tool functionalities and how to create datasets, as well as a usage scenario to illustrate the process of data creation.