05 Fakultät Informatik, Elektrotechnik und Informationstechnik
Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/6
Browse
10 results
Search Results
Item Open Access Large-scale analysis of textual and multivariate data combining machine learning and visualization(2022) Knittel, Johannes; Ertl, Thomas (Prof. Dr.)Item Open Access Deep learning based prediction and visual analytics for temporal environmental data(2022) Harbola, Shubhi; Coors, Volker (Prof. Dr.)The objective of this thesis is to focus on developing Machine Learning methods and their visualisation for environmental data. The presented approaches primarily focus on devising an accurate Machine Learning framework that supports the user in understanding and comparing the model accuracy in relation to essential aspects of the respective parameter selection, trends, time frame, and correlating together with considered meteorological and pollution parameters. Later, this thesis develops approaches for the interactive visualisation of environmental data that are wrapped over the time series prediction as an application. Moreover, these approaches provide an interactive application that supports: 1. a Visual Analytics platform to interact with the sensors data and enhance the representation of the environmental data visually by identifying patterns that mostly go unnoticed in large temporal datasets, 2. a seasonality deduction platform presenting analyses of the results that clearly demonstrate the relationship between these parameters in a combined temporal activities frame, and 3. air quality analyses that successfully discovers spatio-temporal relationships among complex air quality data interactively in different time frames by harnessing the user’s knowledge of factors influencing the past, present, and future behaviour with Machine Learning models' aid. Some of the above pieces of work contribute to the field of Explainable Artificial Intelligence which is an area concerned with the development of methods that help understand, explain and interpret Machine Learning algorithms. In summary, this thesis describes Machine Learning prediction algorithms together with several visualisation approaches for visually analysing the temporal relationships among complex environmental data in different time frames interactively in a robust web platform. The developed interactive visualisation system for environmental data assimilates visual prediction, sensors’ spatial locations, measurements of the parameters, detailed patterns analyses, and change in conditions over time. This provides a new combined approach to the existing visual analytics research. The algorithms developed in this thesis can be used to infer spatio-temporal environmental data, enabling the interactive exploration processes, thus helping manage the cities smartly.Item Open Access Reshaping ubiquitous interaction through sensory augmentation(2020) Kiss, FranciscoCurrent technological advances enable unprecedented approaches for aiding human perception through digital technologies. Powerful Sensory Augmentation becomes feasible. Our concept aims to improve the natural sensory capabilities of users while transcending the traditional interaction concept of tools. This approach promises to facilitate a more natural interaction with ubiquitous computers and support the enhancement of interaction with reality through computing machines. The goal of this research is to investigate the possibilities of this emerging paradigm and to provide a formal structure for future efforts in this field. In this thesis, we propose a definition for Sensory Augmentation and a design space to structure its study. We present a vision for its application to human activities and identify opportunities and challenges for the incorporation of Sensory Augmentation to human activities. Further, we report on three research probes that help investigate these opportunities and challenges, as well as users’ experiences and preferences for this technology. Each of these probes explores a specific area of augmentation, both assessing specific theoretical aspects of Sensory Augmentation and providing practical insights in the technical challenges posed by the development of applications. Additionally, we investigate and discuss the implications of augmenting the senses from the perspectives of users and society, focusing on benefits but also addressing the main foreseeable negative effects and possible strategies to minimize them. Finally, we present a design process for Sensory Augmentation applications alongside with practical recommendations and discuss future research directions of this new field.Item Open Access Designing smart home appliances displaying non-urgent everyday information(2021) Voit, Alexandra; Henze, Niels (Prof. Dr.)In the smart home era living environments are significantly changing. Smart home technologies offer new opportunities for the users to control or monitor their smart home appliances. In the near future, smart home appliances may need to inform their users about everyday home details, such as their current states. Previous work already investigated the users' acceptance of smart home notifications presenting everyday information. However, little research has been done on how users can access the everyday home information. This thesis examines how smart home appliances presenting everyday home information should be designed. It reports about nine user studies investigating either the context of use or the user requirements for smart home appliances presenting everyday home information or evaluating the design solutions for the investigated research probes. As a result of this, we focus on which modalities, locations and information strategies should be used to convey everyday information in a smart home context. In addition, we study how smart home appliances cab be designed to suit to the users' complex daily routines. We further investigate which evaluation methods are suitable for evaluating of smart home appliances presenting everyday home information. This thesis contributes insights into advantages and disadvantages for various evaluation methods and design guidelines for the development of smart home appliances that present everyday home information.Item Open Access Displaying text using head-mounted displays(2021) Rzayev, Rufat; Henze, Niels (Prof. Dr.)Reading text is a fundamental activity to gain information both in the real and the digital world. Although digital text is ubiquitous and accessible through various devices, several challenges need to be addressed for efficient display of text on head-mounted displays (HMDs). Specifically, the display of text in the user's field of view or in the vicinity can lead to occlusion of objects in the real or the virtual environment, distraction from the current activity and decrease of visibility of the environment or the text itself. To overcome these challenges and to understand when they occur, it is important to investigate the presentation of text on HMDs. In this thesis, through a series of lab studies and an online survey, we investigate text presentation using HMDs. Particularly, we study the effects of spatial characteristics and presentation types of text on HMDs and the social implication of displaying text using HMDs. First, we show how short texts can be efficiently displayed using HMDs considering the context. Second, we present design recommendations for displaying long texts on HMDs regarding their spatial characteristics and presentation types. Finally, we show how displaying text on HMDs in a social setting affects the social interaction and how the priority of the text can affect the users' preference during social interaction. This thesis contributes with three main insights to provide designers with recommendations for designing reading interfaces for HMD-based applications.Item Open Access Interactive tactile representations to support document accessibility for people with visual impairments(2020) Ávila Soto, Mauro; Schmidt, Albrecht (Prof. Dr.)Since the early beginnings of writing, humans exploited text layout as a primary means to facilitate reading access to a document. White space, spatial arrangements of text elements on the page, font sizes, and font weight are but a few examples. These features are judiciously employed to structure the document visually, in order to emphasize its logical structure. Sighted users enjoy the advantages that a well-designed layout confers, often without even realizing it; only when documents suer from illogical layoutsùe.g., missing white spaces or clear sectionsùdo users realize the benets of a well-organized layout. In contrast, it is the norm for visually impaired people to be provided little to no information about the spatial layout of documents. Braille text, sonification, and Text-To-Speech (TTS) can provide access to digital documents, albeit in a linearized form. This means that structural information, namely a bird's-eye-view, is mostly absent. For linear reading, this is a minor inconvenience that users can work around. However, spatial structures can be expected to strongly contribute to activities besides linear reading, such as document skimming, revising for a test, memorizing, understanding concepts, and text comparison. This lack of layout cues and structural information can provoke distinct types of reading hindrances. A reader with visual impairment may start reading multiple sidebar paragraphs before starting to read the main text without noticing, which is not optimal for reading a textbook. If readers want to revise a paragraph or access a certain element of the document, they must to go through each element on the page before reaching the targeted paragraph, due screen readers iterate through each paragraph linearly. This can be very frustrating and exhausting, especially if it would be required multiple times during a long study session. To identify different document's elements through a screen reader is also challenging. For example the user would not be able to distinguish between the main text and a sidebar paragraph. In the present dissertation, we portray in detail the research process where we investigated the access of digital textual contents by readers with visual impairments, we define strength and flaws of this content pursuit process. Then, we aim to provide answers and solutions related to the design and usability of touchable and auditory modalities to supply access to digital textual content to readers with visual impairments. We describe the observations and conclusions related to point out different features and characteristics of the reading process based on TTS screen readers for readers with visual impairments. We studied existing skim-reading strategies in sighted and visually impaired individuals, we identified key features of visual documents that allow sighted users to adopt a highly flexible and goal-oriented approach to skim-reading, elements which are mostly lost when documents are accessed in auditory form, as is commonly the case for readers with visual impairments. Different approaches have been proposed to combine electronic devices with tactile media, such as combining tactile elements with touch screen devices to produce interactive tactile maps. tactile and tangible elements have been added to enhance the accessibility of mainstream touch screen interfaces. We envisioned, designed and developed through three different design stages Tactile sheets, a system which has worked on conveying information about visual layout of digital textual documents, this system is based on differentiated tactile elements layouted on teh surface of overlays; these overlays are placed on a touchscreen tablet, creating a tactile interface which makes possible the users to navigate through a document, recongnize its layout, to acquire implicit information related to the position and clasification of different document elements like titles, paragraphs, tables or highlighted text, and access its content through auditory feedback. As a system's design iteration we have introduced an automatic identification method which allows the user to update the digital content in relation to a concrete tactile overlay placed on the tochscreen device, this improved the usability of tactile overlays and granted a innovative lebel of interactivity and fluidity to the tactual pages interface system. This solution was assessed conducting a set of studies with visually impaired participants who were requested to perform different tasks with our system and with auditory screen readers. The approach received positive quantitative and qualitative assessments and was preferred by most participants over the screen readers. Our experimental setting demonstrated the feasibility of a multimodal interface dedicated for visually impaired users, modalities are based on tactual and auditory channels. We also demonstrated that this interface can convey a concrete flow of information to the users, and those users can interact with that information within teh context of the mentioned interface. The possibility to generate a refreshable interface based on a mean initially conceived as a fixed setting like tactile cardboard overlays was proved. This feature has been obtained utilizing affordable resources and through an easily replicable method. Thus, we consider that the automatic identified tactile overlays approach is a promising method for touch screens accessibility, that improves the usability and gives the users more control over the touch screen and provides more content on the layout structure and element types of a text document.Item Open Access Understanding stress responses related to digital technologies(2020) Poguntke, Romina; Schmidt, Albrecht (Prof. Dr.)Item Open Access Interaktive visuelle Analysetechniken für die Exploration narrativer Texte(2022) John, Markus; Ertl, Thomas (Prof. Dr.)Durch die fortschreitende Digitalisierung textueller Daten werden enormeMengen an digitalen Informationen täglich generiert und gespeichert. Dieser technologische Fortschritt hat in den vergangenen Jahrzehnten nahezu alle gesellschaftlichen Bereiche erfasst und verändert. Darunter fallen ebenfalls die Geisteswissenschaften, da sich durch die Digitalisierung von Literatur und Webportale, welche einen einfachen Zugang zu diesen digitalen Kopien bieten, viele neue Forschungsmöglichkeiten eröffnet haben. Aus Sicht der Informatik haben sich die digitalen Geisteswissenschaften in den letzten Jahren zu einem viel diskutierten und spannenden Forschungsgebiet entwickelt. Es handelt sich um ein interdisziplinäres Forschungsfeld, das geisteswissenschaftliche Fragestellungen mit computergestützten Verfahren der Informatik untersucht. Eine Herausforderung in vielen digitalen geisteswissenschaftlichen Disziplinen ist die Untersuchung von narrativen Texten. Die Geisteswissenschaftler*innen sind daran interessiert, die Persönlichkeitsmerkmale der Figuren, deren Rollen, Beziehungen und die Entwicklung dieser Aspekte im Verlauf der Erzählung zu untersuchen. Traditionell analysieren die Geisteswissenschaftler*innen ihre zugrunde liegenden Textdokumente durch intensives Lesen und sorgfältiges Interpretieren, sogenanntes Close Reading. Dies ist jedoch eine herausfordernde und zeitaufwendige Vorgehensweise. Diese komplexen Analyseschritte können durch computergestützte Verfahren unterstützt werden. Computerlinguistische Methoden ermöglichen es, gezielt Wissen automatisch aus den Textdokumenten abzuleiten. Die abgeleiteten Informationen können als visuelle Abstraktionen dargestellt werden, sogenannte Distant Reading-Methoden. Visuelle Abstraktionen können Informationen aufzeigen, die es ermöglichen, komplexe Zusammenhänge zu untersuchen, die durch Close Reading nur schwer zu erfassen sind. Um die Zusammenhänge sowie entstandene Ideen und Hypothesen detaillierter analysieren zu können, ist es jedoch notwendig, die zugehörigen Textabschnitte zu studieren. Daher ist es für weiterführende Analysen essenziell, die visuellen Abstraktionen mit dem zugrunde liegenden Textgegenstand zu verknüpfen. Auf diese Weise können entdeckte Sachverhalte oder entstandene Hypothesen weiterführend überprüft werden. Die vorliegende Dissertation fokussiert sich auf die Verknüpfung der Close- und Distant Reading-Techniken. In den nachfolgenden Kapiteln werden neuartige Visualisierungswerkzeuge eingeführt, die sowohl eine abstrakte als auch detaillierte Untersuchung narrativer Texte ermöglichen. Diese Analysetechniken unterstützen die Geisteswissenschaftler*innen bei der Exploration einzelner Texte, beim Vergleich mehrerer Textdokumente sowie bei der Analyse von sehr großen Textkorpora. Des Weiteren werden Unsicherheitsquellen diskutiert, die während der Entwicklung der Visualisierungswerkzeuge festgestellt oder in Diskussionen mit Forscher*innen aus den Geisteswissenschaften und der Computerlinguistik reflektiert wurden. In diesem Kontext wird eine erstellte Übersicht von visuellen Kommunikations- und Interaktionsmöglichkeiten präsentiert sowie Techniken vorgestellt, mit denen Unsicherheitsaspekte untersucht und angepasst werden können. Die erläuterten Möglichkeiten und entworfenen Werkzeuge helfen neuen und weiterführenden Projekten abgeleitete Informationen kritischer zu hinterfragen. Ferner wird eine methodische Vorgehensweise vorgestellt, welche in vielen der nachfolgenden Analysetechniken unterstützt wird, um die visuelle Untersuchung narrativer Texte zu ermöglichen. Die Dissertation konzentriert sich hauptsächlich auf die Untersuchung von Figuren in narrativen Texten, dennoch können viele der vorgestellten Visualisierungswerkzeuge für weitere komplexe Textdokumentanalysen angepasst und angewendet werden. Basierend auf den eingeführten Werkzeugen werden definierte Forschungsfragen diskutiert und offene Herausforderungen für zukünftige Forschungsmöglichkeiten abgeleitet.Item Open Access Data challenges in variational optical flow(2020) Stoll, Michael; Bruhn, Andrés (Prof. Dr.-Ing.)Item Open Access Designing ubiquitous-computing systems for memory alterations(2021) Elagroudy, Passant; Schmidt, Albrecht (Prof. Dr.)Memories shape the identity of civilizations through shared knowledge. However, the increasing digital resources competing to generate content and alter what we remember threaten our natural mechanisms for forgetting inessentials to focus on what is important. In this work, we aim to understand the design process of memory-altering ubiquitous computing systems (MACS) and to raise awareness about accidental alterations from using technology. With the help of 188 participants, we explore: 1) how to design MACS, 2) how to conduct privacy-aware memory research, and 3) the users' expectations for utilizing their memories by others and by AI systems. We consolidate the design insights from eight projects to present “MemAlt”, a framework for designing MACS. We also develop two privacy-aware memory research methods to anonymously collect data and ubiquitously evaluate it. Additionally, we pin sharing expectations on social media to foresee patterns with memory technologies. With the prevalence of tracking devices such as mobiles, fitness devices, and smart home gadgets, we envision a future where domain-specific MACS are ubiquitous in everyday interactions. Thus, this work makes designing MACS more accessible to researchers, and allows them to reflect on adverse effects of their designs on memory.