004 Datenverarbeitung; Informatik
Refine
Document Type
- Doctoral Thesis (6) (remove)
Keywords
- Maschinelles Lernen (2)
- Annotation (1)
- Augenbewegung (1)
- Auslese (1)
- Auswahl (1)
- Bipartiter Graph (1)
- Blickbewegung (1)
- Enhanced Representation (1)
- Eye Tracking (1)
- Eyetracking (1)
Institute
- Institute for Web Science and Technologies (6) (remove)
The Web is an essential component of moving our society to the digital age. We use it for communication, shopping, and doing our work. Most user interaction in the Web happens with Web page interfaces. Thus, the usability and accessibility of Web page interfaces are relevant areas of research to make the Web more useful. Eye tracking is a tool that can be helpful in both areas, performing usability testing and improving accessibility. It can be used to understand users' attention on Web pages and to support usability experts in their decision-making process. Moreover, eye tracking can be used as an input method to control an interface. This is especially useful for people with motor impairment, who cannot use traditional input devices like mouse and keyboard. However, interfaces on Web pages become more and more complex due to dynamics, i.e., changing contents like animated menus and photo carousels. We need general approaches to comprehend dynamics on Web pages, allowing for efficient usability analysis and enjoyable interaction with eye tracking. In the first part of this thesis, we report our work on improving gaze-based analysis of dynamic Web pages. Eye tracking can be used to collect the gaze signals of users, who browse a Web site and its pages. The gaze signals show a usability expert what parts in the Web page interface have been read, glanced at, or skipped. The aggregation of gaze signals allows a usability expert insight into the users' attention on a high-level, before looking into individual behavior. For this, all gaze signals must be aligned to the interface as experienced by the users. However, the user experience is heavily influenced by changing contents, as these may cover a substantial portion of the screen. We delineate unique states in Web page interfaces including changing contents, such that gaze signals from multiple users can be aggregated correctly. In the second part of this thesis, we report our work on improving the gaze-based interaction with dynamic Web pages. Eye tracking can be used to retrieve gaze signals while a user operates a computer. The gaze signals may be interpreted as input controlling an interface. Nowadays, eye tracking as an input method is mostly used to emulate mouse and keyboard functionality, hindering an enjoyable user experience. There exist a few Web browser prototypes that directly interpret gaze signals for control, but they do not work on dynamic Web pages. We have developed a method to extract interaction elements like hyperlinks and text inputs efficiently on Web pages, including changing contents. We adapt the interaction with those elements for eye tracking as the input method, such that a user can conveniently browse the Web hands-free. Both parts of this thesis conclude with user-centered evaluations of our methods, assessing the improvements in the user experience for usability experts and people with motor impairment, respectively.
Graph-based data formats are flexible in representing data. In particular semantic data models, where the schema is part of the data, gained traction and commercial success in recent years. Semantic data models are also the basis for the Semantic Web - a Web of data governed by open standards in which computer programs can freely access the provided data. This thesis is concerned with the correctness of programs that access semantic data. While the flexibility of semantic data models is one of their biggest strengths, it can easily lead to programmers accidentally not accounting for unintuitive edge cases. Often, such exceptions surface during program execution as run-time errors or unintended side-effects. Depending on the exact condition, a program may run for a long time before the error occurs and the program crashes.
This thesis defines type systems that can detect and avoid such run-time errors based on schema languages available for the Semantic Web. In particular, this thesis uses the Web Ontology Language (OWL) and its theoretic underpinnings, i.e., description logics, as well as the Shapes Constraint Language (SHACL) to define type systems that provide type-safe data access to semantic data graphs. Providing a safe type system is an established methodology for proving the absence of run-time errors in programs without requiring execution. Both schema languages are based on possible world semantics but differ in the treatment of incomplete knowledge. While OWL allows for modelling incomplete knowledge through an open-world semantics, SHACL relies on a fixed domain and closed-world semantics. We provide the formal underpinnings for type systems based on each of the two schema languages. In particular, we base our notion of types on sets of values which allows us to specify a subtype relation based on subset semantics. In case of description logics, subsumption is a routine problem. For
the type system based on SHACL, we are able to translate it into a description
logic subsumption problem.
Through the increasing availability of access to the web, more and more interactions between people take place in online social networks, such as Twitter or Facebook, or sites where opinions can be exchanged. At the same time, knowledge is made openly available for many people, such as by the biggest collaborative encyclopedia Wikipedia and diverse information in Internet forums and on websites. These two kinds of networks - social networks and knowledge networks - are highly dynamic in the sense that the links that contain the important information about the relationships between people or the relations between knowledge items are frequently updated or changed. These changes follow particular structural patterns and characteristics that are far less random than expected.
The goal of this thesis is to predict three characteristic link patterns for the two network types of interest: the addition of new links, the removal of existing links and the presence of latent negative links. First, we show that the prediction of link removal is indeed a new and challenging problem. Even if the sociological literature suggests that reasons for the formation and resolution of ties are often complementary, we show that the two respective prediction problems are not. In particular, we show that the dynamics of new links and unlinks lead to the four link states of growth, decay, stability and instability. For knowledge networks we show that the prediction of link changes greatly benefits from the usage of temporal information; the timestamp of link creation and deletion events improves the prediction of future link changes. For that, we present and evaluate four temporal models that resemble different exploitation strategies. Focusing on directed social networks, we conceptualize and evaluate sociological constructs that explain the formation and dissolution of relationships between users. Measures based on information about past relationships are extremely valuable for predicting the dissolution of social ties. Hence, consistent for knowledge networks and social networks, temporal information in a network greatly improves the prediction quality. Turning again to social networks, we show that negative relationship information such as distrust or enmity can be predicted from positive known relationships in the network. This is particularly interesting in networks where users cannot label their relationships to other users as negative. For this scenario we show how latent negative relationships can be predicted.
The availability of digital cameras and the possibility to take photos at no cost lead to an increasing amount of digital photos online and on private computers. The pure amount of data makes approaches that support users in the administration of the photo necessary. As the automatic understanding of photo content is still an unsolved task, metadata is needed for supporting administrative tasks like search or photo work such as the generation of photo books. Meta-information textually describes the depicted scene or consists of information on how good or interesting a photo is.
In this thesis, an approach for creating meta-information without additional effort for the user is investigated. Eye tracking data is used to measure the human visual attention. This attention is analyzed with the objective of information creation in the form of metadata. The gaze paths of users working with photos are recorded, for example, while they are searching for photos or while they are just viewing photo collections.
Eye tracking hardware is developing fast within the last years. Because of falling prices for sensor hardware such as cameras and more competition on the eye tracker market, the prices are falling, and the usability is increasing. It can be assumed that eye tracking technology can soon be used in everyday devices such as laptops or mobile phones. The exploitation of data, recorded in the background while the user is performing daily tasks with photos, has great potential to generate information without additional effort for the users.
The first part of this work deals with the labeling of image region by means of gaze data for describing the depicted scenes in detail. Labeling takes place by assigning object names to specific photo regions. In total, three experiments were conducted for investigating the quality of these assignments in different contexts. In the first experiment, users decided whether a given object can be seen on a photo by pressing a button. In the second study, participants searched for specific photos in an image search application. In the third experiment, gaze data was collected from users playing a game with the task to classify photos regarding given categories. The results of the experiments showed that gaze-based region labeling outperforms baseline approaches in various contexts. In the second part, most important photos in a collection of photos are identified by means of visual attention for the creation of individual photo selections. Users freely viewed photos of a collection without any specific instruction on what to fixate, while their gaze paths were recorded. By comparing gaze-based and baseline photo selections to manually created selections, the worth of eye tracking data in the identification of important photos is shown. In the analysis of the data, the characteristics of gaze data has to be considered, for example, inaccurate and ambiguous data. The aggregation of gaze data, collected from several users, is one suggested approach for dealing with this kind of data.
The results of the performed experiments show the value of gaze data as source of information. It allows to benefit from human abilities where algorithms still have problems to perform satisfyingly.
Tagging systems are intriguing dynamic systems, in which users collaboratively index resources with the so-called tags. In order to leverage the full potential of tagging systems, it is important to understand the relationship between the micro-level behavior of the individual users and the macro-level properties of the whole tagging system. In this thesis, we present the Epistemic Dynamic Model, which tries to bridge this gap between the micro-level behavior and the macro-level properties by developing a theory of tagging systems. The model is based on the assumption that the combined influence of the shared background knowledge of the users and the imitation of tag recommendations are sufficient for explaining the emergence of the tag frequency distribution and the vocabulary growth in tagging systems. Both macro-level properties of tagging systems are closely related to the emergence of the shared community vocabulary. rnrnWith the help of the Epistemic Dynamic Model, we show that the general shape of the tag frequency distribution and of the vocabulary growth have their origin in the shared background knowledge of the users. Tag recommendations can then be used for selectively influencing this general shape. In this thesis, we especially concentrate on studying the influence of recommending a set of popular tags. Recommending popular tags adds a feedback mechanism between the vocabularies of individual users that increases the inter-indexer consistency of the tag assignments. How does this influence the indexing quality in a tagging system? For this purpose, we investigate a methodology for measuring the inter-resource consistency of tag assignments. The inter-resource consistency is an indicator of the indexing quality, which positively correlates with the precision and recall of query results. It measures the degree to which the tag vectors of indexed resources reflect how the users perceive the similarity between resources. We argue with our model, and show it with a user experiment, that recommending popular tags decreases the inter-resource consistency in a tagging system. Furthermore, we show that recommending the user his/her previously used tags helps to increase the inter-resource consistency. Our measure of the inter-resource consistency complements existing measures for the evaluation and comparison of tag recommendation algorithms, moving the focus to evaluating their influence on the indexing quality.
In this thesis, I study the spectral characteristics of large dynamic networks and formulate the spectral evolution model. The spectral evolution model applies to networks that evolve over time, and describes their spectral decompositions such as the eigenvalue and singular value decomposition. The spectral evolution model states that over time, the eigenvalues of a network change while its eigenvectors stay approximately constant.
I validate the spectral evolution model empirically on over a hundred network datasets, and theoretically by showing that it generalizes arncertain number of known link prediction functions, including graph kernels, path counting methods, rank reduction and triangle closing. The collection of datasets I use contains 118 distinct network datasets. One dataset, the signed social network of the Slashdot Zoo, was specifically extracted during work on this thesis. I also show that the spectral evolution model can be understood as a generalization of the preferential attachment model, if we consider growth in latent dimensions of a network individually. As applications of the spectral evolution model, I introduce two new link prediction algorithms that can be used for recommender systems, search engines, collaborative filtering, rating prediction, link sign prediction and more.
The first link prediction algorithm reduces to a one-dimensional curve fitting problem from which a spectral transformation is learned. The second method uses extrapolation of eigenvalues to predict future eigenvalues. As special cases, I show that the spectral evolution model applies to directed, undirected, weighted, unweighted, signed and bipartite networks. For signed graphs, I introduce new applications of the Laplacian matrix for graph drawing, spectral clustering, and describe new Laplacian graph kernels. I also define the algebraic conflict, a measure of the conflict present in a signed graph based on the signed graph Laplacian. I describe the problem of link sign prediction spectrally, and introduce the signed resistance distance. For bipartite and directed graphs, I introduce the hyperbolic sine and odd Neumann kernels, which generalize the exponential and Neumann kernels for undirected unipartite graphs. I show that the problem of directed and bipartite link prediction are related by the fact that both can be solved by considering spectral evolution in the singular value decomposition.