@phdthesis{Perscheid2013,
  author    = {Perscheid, Michael},
  title     = {Test-driven fault navigation for debugging reproducible failures},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-68155},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {The correction of software failures tends to be very cost-intensive because their debugging is an often time-consuming development activity. During this activity, developers largely attempt to understand what causes failures: Starting with a test case that reproduces the observable failure they have to follow failure causes on the infection chain back to the root cause (defect). This idealized procedure requires deep knowledge of the system and its behavior because failures and defects can be far apart from each other. Unfortunately, common debugging tools are inadequate for systematically investigating such infection chains in detail. Thus, developers have to rely primarily on their intuition and the localization of failure causes is not time-efficient. To prevent debugging by disorganized trial and error, experienced developers apply the scientific method and its systematic hypothesis-testing. However, even when using the scientific method, the search for failure causes can still be a laborious task. First, lacking expertise about the system makes it hard to understand incorrect behavior and to create reasonable hypotheses. Second, contemporary debugging approaches provide no or only partial support for the scientific method. In this dissertation, we present test-driven fault navigation as a debugging guide for localizing reproducible failures with the scientific method. Based on the analysis of passing and failing test cases, we reveal anomalies and integrate them into a breadth-first search that leads developers to defects. This systematic search consists of four specific navigation techniques that together support the creation, evaluation, and refinement of failure cause hypotheses for the scientific method. First, structure navigation localizes suspicious system parts and restricts the initial search space. Second, team navigation recommends experienced developers for helping with failures. Third, behavior navigation allows developers to follow emphasized infection chains back to root causes. Fourth, state navigation identifies corrupted state and reveals parts of the infection chain automatically. We implement test-driven fault navigation in our Path Tools framework for the Squeak/Smalltalk development environment and limit its computation cost with the help of our incremental dynamic analysis. This lightweight dynamic analysis ensures an immediate debugging experience with our tools by splitting the run-time overhead over multiple test runs depending on developers' needs. Hence, our test-driven fault navigation in combination with our incremental dynamic analysis answers important questions in a short time: where to start debugging, who understands failure causes best, what happened before failures, and which state properties are infected.},
  language  = {en}
}
@phdthesis{Berg2013,
  author    = {Berg, Gregor},
  title     = {Virtual prototypes for the model-based elicitation and validation of collaborative scenarios},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-69729},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {Requirements engineers have to elicit, document, and validate how stakeholders act and interact to achieve their common goals in collaborative scenarios. Only after gathering all information concerning who interacts with whom to do what and why, can a software system be designed and realized which supports the stakeholders to do their work. To capture and structure requirements of different (groups of) stakeholders, scenario-based approaches have been widely used and investigated. Still, the elicitation and validation of requirements covering collaborative scenarios remains complicated, since the required information is highly intertwined, fragmented, and distributed over several stakeholders. Hence, it can only be elicited and validated collaboratively. In times of globally distributed companies, scheduling and conducting workshops with groups of stakeholders is usually not feasible due to budget and time constraints. Talking to individual stakeholders, on the other hand, is feasible but leads to fragmented and incomplete stakeholder scenarios. Going back and forth between different individual stakeholders to resolve this fragmentation and explore uncovered alternatives is an error-prone, time-consuming, and expensive task for the requirements engineers. While formal modeling methods can be employed to automatically check and ensure consistency of stakeholder scenarios, such methods introduce additional overhead since their formal notations have to be explained in each interaction between stakeholders and requirements engineers. Tangible prototypes as they are used in other disciplines such as design, on the other hand, allow designers to feasibly validate and iterate concepts and requirements with stakeholders. This thesis proposes a model-based approach for prototyping formal behavioral specifications of stakeholders who are involved in collaborative scenarios. By simulating and animating such specifications in a remote domain-specific visualization, stakeholders can experience and validate the scenarios captured so far, i.e., how other stakeholders act and react. This interactive scenario simulation is referred to as a model-based virtual prototype. Moreover, through observing how stakeholders interact with a virtual prototype of their collaborative scenarios, formal behavioral specifications can be automatically derived which complete the otherwise fragmented scenarios. This, in turn, enables requirements engineers to elicit and validate collaborative scenarios in individual stakeholder sessions - decoupled, since stakeholders can participate remotely and are not forced to be available for a joint session at the same time. This thesis discusses and evaluates the feasibility, understandability, and modifiability of model-based virtual prototypes. Similarly to how physical prototypes are perceived, the presented approach brings behavioral models closer to being tangible for stakeholders and, moreover, combines the advantages of joint stakeholder sessions and decoupled sessions.},
  language  = {en}
}
@phdthesis{Boehm2013,
  author    = {B{\"o}hm, Christoph},
  title     = {Enriching the Web of Data with topics and links},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-68624},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {This thesis presents novel ideas and research findings for the Web of Data - a global data space spanning many so-called Linked Open Data sources. Linked Open Data adheres to a set of simple principles to allow easy access and reuse for data published on the Web. Linked Open Data is by now an established concept and many (mostly academic) publishers adopted the principles building a powerful web of structured knowledge available to everybody. However, so far, Linked Open Data does not yet play a significant role among common web technologies that currently facilitate a high-standard Web experience. In this work, we thoroughly discuss the state-of-the-art for Linked Open Data and highlight several shortcomings - some of them we tackle in the main part of this work. First, we propose a novel type of data source meta-information, namely the topics of a dataset. This information could be published with dataset descriptions and support a variety of use cases, such as data source exploration and selection. For the topic retrieval, we present an approach coined Annotated Pattern Percolation (APP), which we evaluate with respect to topics extracted from Wikipedia portals. Second, we contribute to entity linking research by presenting an optimization model for joint entity linking, showing its hardness, and proposing three heuristics implemented in the LINked Data Alignment (LINDA) system. Our first solution can exploit multi-core machines, whereas the second and third approach are designed to run in a distributed shared-nothing environment. We discuss and evaluate the properties of our approaches leading to recommendations which algorithm to use in a specific scenario. The distributed algorithms are among the first of their kind, i.e., approaches for joint entity linking in a distributed fashion. Also, we illustrate that we can tackle the entity linking problem on the very large scale with data comprising more than 100 millions of entity representations from very many sources. Finally, we approach a sub-problem of entity linking, namely the alignment of concepts. We again target a method that looks at the data in its entirety and does not neglect existing relations. Also, this concept alignment method shall execute very fast to serve as a preprocessing for further computations. Our approach, called Holistic Concept Matching (HCM), achieves the required speed through grouping the input by comparing so-called knowledge representations. Within the groups, we perform complex similarity computations, relation conclusions, and detect semantic contradictions. The quality of our result is again evaluated on a large and heterogeneous dataset from the real Web. In summary, this work contributes a set of techniques for enhancing the current state of the Web of Data. All approaches have been tested on large and heterogeneous real-world input.},
  language  = {en}
}
@phdthesis{Bauckmann2013,
  author    = {Bauckmann, Jana},
  title     = {Dependency discovery for data integration},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-66645},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {Data integration aims to combine data of different sources and to provide users with a unified view on these data. This task is as challenging as valuable. In this thesis we propose algorithms for dependency discovery to provide necessary information for data integration. We focus on inclusion dependencies (INDs) in general and a special form named conditional inclusion dependencies (CINDs): (i) INDs enable the discovery of structure in a given schema. (ii) INDs and CINDs support the discovery of cross-references or links between schemas. An IND "A in B" simply states that all values of attribute A are included in the set of values of attribute B. We propose an algorithm that discovers all inclusion dependencies in a relational data source. The challenge of this task is the complexity of testing all attribute pairs and further of comparing all of each attribute pair's values. The complexity of existing approaches depends on the number of attribute pairs, while ours depends only on the number of attributes. Thus, our algorithm enables to profile entirely unknown data sources with large schemas by discovering all INDs. Further, we provide an approach to extract foreign keys from the identified INDs. We extend our IND discovery algorithm to also find three special types of INDs: (i) Composite INDs, such as "AB in CD", (ii) approximate INDs that allow a certain amount of values of A to be not included in B, and (iii) prefix and suffix INDs that represent special cross-references between schemas. Conditional inclusion dependencies are inclusion dependencies with a limited scope defined by conditions over several attributes. Only the matching part of the instance must adhere the dependency. We generalize the definition of CINDs distinguishing covering and completeness conditions and define quality measures for conditions. We propose efficient algorithms that identify covering and completeness conditions conforming to given quality thresholds. The challenge for this task is twofold: (i) Which (and how many) attributes should be used for the conditions? (ii) Which attribute values should be chosen for the conditions? Previous approaches rely on pre-selected condition attributes or can only discover conditions applying to quality thresholds of 100\%. Our approaches were motivated by two application domains: data integration in the life sciences and link discovery for linked open data. We show the efficiency and the benefits of our approaches for use cases in these domains.},
  language  = {en}
}
@book{MeyerPufahlFahlandetal.2013,
  author    = {Meyer, Andreas and Pufahl, Luise and Fahland, Dirk and Weske, Mathias},
  title     = {Modeling and enacting complex data dependencies in business processes},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-245-2},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-65103},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {40},
  year      = {2013},
  abstract  = {Enacting business processes in process engines requires the coverage of control flow, resource assignments, and process data. While the first two aspects are well supported in current process engines, data dependencies need to be added and maintained manually by a process engineer. Thus, this task is error-prone and time-consuming. In this report, we address the problem of modeling processes with complex data dependencies, e.g., m:n relationships, and their automatic enactment from process models. First, we extend BPMN data objects with few annotations to allow data dependency handling as well as data instance differentiation. Second, we introduce a pattern-based approach to derive SQL queries from process models utilizing the above mentioned extensions. Therewith, we allow automatic enactment of data-aware BPMN process models. We implemented our approach for the Activiti process engine to show applicability.},
  language  = {en}
}
@phdthesis{Becker2013,
  author    = {Becker, Basil},
  title     = {Architectural modelling and verification of open service-oriented systems of systems},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-70158},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {Systems of Systems (SoS) have received a lot of attention recently. In this thesis we will focus on SoS that are built atop the techniques of Service-Oriented Architectures and thus combine the benefits and challenges of both paradigms. For this thesis we will understand SoS as ensembles of single autonomous systems that are integrated to a larger system, the SoS. The interesting fact about these systems is that the previously isolated systems are still maintained, improved and developed on their own. Structural dynamics is an issue in SoS, as at every point in time systems can join and leave the ensemble. This and the fact that the cooperation among the constituent systems is not necessarily observable means that we will consider these systems as open systems. Of course, the system has a clear boundary at each point in time, but this can only be identified by halting the complete SoS. However, halting a system of that size is practically impossible. Often SoS are combinations of software systems and physical systems. Hence a failure in the software system can have a serious physical impact what makes an SoS of this kind easily a safety-critical system. The contribution of this thesis is a modelling approach that extends OMG's SoaML and basically relies on collaborations and roles as an abstraction layer above the components. This will allow us to describe SoS at an architectural level. We will also give a formal semantics for our modelling approach which employs hybrid graph-transformation systems. The modelling approach is accompanied by a modular verification scheme that will be able to cope with the complexity constraints implied by the SoS' structural dynamics and size. Building such autonomous systems as SoS without evolution at the architectural level --- i. e. adding and removing of components and services --- is inadequate. Therefore our approach directly supports the modelling and verification of evolution.},
  language  = {en}
}
@phdthesis{Lange2013,
  author    = {Lange, Dustin},
  title     = {Effective and efficient similarity search in databases},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-65712},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {Given a large set of records in a database and a query record, similarity search aims to find all records sufficiently similar to the query record. To solve this problem, two main aspects need to be considered: First, to perform effective search, the set of relevant records is defined using a similarity measure. Second, an efficient access method is to be found that performs only few database accesses and comparisons using the similarity measure. This thesis solves both aspects with an emphasis on the latter. In the first part of this thesis, a frequency-aware similarity measure is introduced. Compared record pairs are partitioned according to frequencies of attribute values. For each partition, a different similarity measure is created: machine learning techniques combine a set of base similarity measures into an overall similarity measure. After that, a similarity index for string attributes is proposed, the State Set Index (SSI), which is based on a trie (prefix tree) that is interpreted as a nondeterministic finite automaton. For processing range queries, the notion of query plans is introduced in this thesis to describe which similarity indexes to access and which thresholds to apply. The query result should be as complete as possible under some cost threshold. Two query planning variants are introduced: (1) Static planning selects a plan at compile time that is used for all queries. (2) Query-specific planning selects a different plan for each query. For answering top-k queries, the Bulk Sorted Access Algorithm (BSA) is introduced, which retrieves large chunks of records from the similarity indexes using fixed thresholds, and which focuses its efforts on records that are ranked high in more than one attribute and thus promising candidates. The described components form a complete similarity search system. Based on prototypical implementations, this thesis shows comparative evaluation results for all proposed approaches on different real-world data sets, one of which is a large person data set from a German credit rating agency.},
  language  = {en}
}
@book{CalmezHesseSiegmundetal.2013,
  author    = {Calmez, Conrad and Hesse, Hubert and Siegmund, Benjamin and Stamm, Sebastian and Thomschke, Astrid and Hirschfeld, Robert and Ingalls, Dan and Lincke, Jens},
  title     = {Explorative authoring of Active Web content in a mobile environment},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-232-2},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-64054},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {132},
  year      = {2013},
  abstract  = {Developing rich Web applications can be a complex job - especially when it comes to mobile device support. Web-based environments such as Lively Webwerkstatt can help developers implement such applications by making the development process more direct and interactive. Further the process of developing software is collaborative which creates the need that the development environment offers collaboration facilities. This report describes extensions of the webbased development environment Lively Webwerkstatt such that it can be used in a mobile environment. The extensions are collaboration mechanisms, user interface adaptations but as well event processing and performance measuring on mobile devices.},
  language  = {en}
}
@phdthesis{Kyprianidis2013,
  author    = {Kyprianidis, Jan Eric},
  title     = {Structure adaptive stylization of images and video},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-64104},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {In the early days of computer graphics, research was mainly driven by the goal to create realistic synthetic imagery. By contrast, non-photorealistic computer graphics, established as its own branch of computer graphics in the early 1990s, is mainly motivated by concepts and principles found in traditional art forms, such as painting, illustration, and graphic design, and it investigates concepts and techniques that abstract from reality using expressive, stylized, or illustrative rendering techniques. This thesis focuses on the artistic stylization of two-dimensional content and presents several novel automatic techniques for the creation of simplified stylistic illustrations from color images, video, and 3D renderings. Primary innovation of these novel techniques is that they utilize the smooth structure tensor as a simple and efficient way to obtain information about the local structure of an image. More specifically, this thesis contributes to knowledge in this field in the following ways. First, a comprehensive review of the structure tensor is provided. In particular, different methods for integrating the minor eigenvector field of the smoothed structure tensor are developed, and the superiority of the smoothed structure tensor over the popular edge tangent flow is demonstrated. Second, separable implementations of the popular bilateral and difference of Gaussians filters that adapt to the local structure are presented. These filters avoid artifacts while being computationally highly efficient. Taken together, both provide an effective way to create a cartoon-style effect. Third, a generalization of the Kuwahara filter is presented that avoids artifacts by adapting the shape, scale, and orientation of the filter to the local structure. This causes directional image features to be better preserved and emphasized, resulting in overall sharper edges and a more feature-abiding painterly effect. In addition to the single-scale variant, a multi-scale variant is presented, which is capable of performing a highly aggressive abstraction. Fourth, a technique that builds upon the idea of combining flow-guided smoothing with shock filtering is presented, allowing for an aggressive exaggeration and an emphasis of directional image features. All presented techniques are suitable for temporally coherent per-frame filtering of video or dynamic 3D renderings, without requiring expensive extra processing, such as optical flow. Moreover, they can be efficiently implemented to process content in real-time on a GPU.},
  language  = {en}
}
@phdthesis{Trapp2013,
  author    = {Trapp, Matthias},
  title     = {Interactive rendering techniques for focus+context visualization of 3D geovirtual environments},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-66824},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {This thesis introduces a collection of new real-time rendering techniques and applications for focus+context visualization of interactive 3D geovirtual environments such as virtual 3D city and landscape models. These environments are generally characterized by a large number of objects and are of high complexity with respect to geometry and textures. For these reasons, their interactive 3D rendering represents a major challenge. Their 3D depiction implies a number of weaknesses such as occlusions, cluttered image contents, and partial screen-space usage. To overcome these limitations and, thus, to facilitate the effective communication of geo-information, principles of focus+context visualization can be used for the design of real-time 3D rendering techniques for 3D geovirtual environments (see Figure). In general, detailed views of a 3D geovirtual environment are combined seamlessly with abstracted views of the context within a single image. To perform the real-time image synthesis required for interactive visualization, dedicated parallel processors (GPUs) for rasterization of computer graphics primitives are used. For this purpose, the design and implementation of appropriate data structures and rendering pipelines are necessary. The contribution of this work comprises the following five real-time rendering methods: • The rendering technique for 3D generalization lenses enables the combination of different 3D city geometries (e.g., generalized versions of a 3D city model) in a single image in real time. The method is based on a generalized and fragment-precise clipping approach, which uses a compressible, raster-based data structure. It enables the combination of detailed views in the focus area with the representation of abstracted variants in the context area. • The rendering technique for the interactive visualization of dynamic raster data in 3D geovirtual environments facilitates the rendering of 2D surface lenses. It enables a flexible combination of different raster layers (e.g., aerial images or videos) using projective texturing for decoupling image and geometry data. Thus, various overlapping and nested 2D surface lenses of different contents can be visualized interactively. • The interactive rendering technique for image-based deformation of 3D geovirtual environments enables the real-time image synthesis of non-planar projections, such as cylindrical and spherical projections, as well as multi-focal 3D fisheye-lenses and the combination of planar and non-planar projections. • The rendering technique for view-dependent multi-perspective views of 3D geovirtual environments, based on the application of global deformations to the 3D scene geometry, can be used for synthesizing interactive panorama maps to combine detailed views close to the camera (focus) with abstract views in the background (context). This approach reduces occlusions, increases the usage the available screen space, and reduces the overload of image contents. • The object-based and image-based rendering techniques for highlighting objects and focus areas inside and outside the view frustum facilitate preattentive perception. The concepts and implementations of interactive image synthesis for focus+context visualization and their selected applications enable a more effective communication of spatial information, and provide building blocks for design and development of new applications and systems in the field of 3D geovirtual environments.},
  language  = {en}
}
@article{MomtaziNaumann2013,
  author    = {Momtazi, Saeedeh and Naumann, Felix},
  title     = {Topic modeling for expert finding using latent Dirichlet allocation},
  series = {Wiley interdisciplinary reviews : Data mining and knowledge discovery},
  volume    = {3},
  journal   = {Wiley interdisciplinary reviews : Data mining and knowledge discovery},
  number    = {5},
  publisher = {Wiley},
  address   = {San Fransisco},
  issn      = {1942-4787},
  doi       = {10.1002/widm.1102},
  pages     = {346 -- 353},
  year      = {2013},
  abstract  = {The task of expert finding is to rank the experts in the search space given a field of expertise as an input query. In this paper, we propose a topic modeling approach for this task. The proposed model uses latent Dirichlet allocation (LDA) to induce probabilistic topics. In the first step of our algorithm, the main topics of a document collection are extracted using LDA. The extracted topics present the connection between expert candidates and user queries. In the second step, the topics are used as a bridge to find the probability of selecting each candidate for a given query. The candidates are then ranked based on these probabilities. The experimental results on the Text REtrieval Conference (TREC) Enterprise track for 2005 and 2006 show that the proposed topic-based approach outperforms the state-of-the-art profile- and document-based models, which use information retrieval methods to rank experts. Moreover, we present the superiority of the proposed topic-based approach to the improved document-based expert finding systems, which consider additional information such as local context, candidate prior, and query expansion.},
  language  = {en}
}
@unpublished{FishLambers2013,
  author    = {Fish, Andrew and Lambers, Leen},
  title     = {Special issue on graph transformation and visual modeling techniques - guest editors' introduction},
  series = {Journal of visual languages and computing},
  volume    = {24},
  journal   = {Journal of visual languages and computing},
  number    = {6},
  publisher = {Elsevier},
  address   = {London},
  issn      = {1045-926X},
  doi       = {10.1016/j.jvlc.2013.08.004},
  pages     = {419 -- 420},
  year      = {2013},
  language  = {en}
}
@phdthesis{Gustafson2013,
  author    = {Gustafson, Sean},
  title     = {Imaginary Interfaces},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-68960},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {The size of a mobile device is primarily determined by the size of the touchscreen. As such, researchers have found that the way to achieve ultimate mobility is to abandon the screen altogether. These wearable devices are operated using hand gestures, voice commands or a small number of physical buttons. By abandoning the screen these devices also abandon the currently dominant spatial interaction style (such as tapping on buttons), because, seemingly, there is nothing to tap on. Unfortunately this design prevents users from transferring their learned interaction knowledge gained from traditional touchscreen-based devices. In this dissertation, I present Imaginary Interfaces, which return spatial interaction to screenless mobile devices. With these interfaces, users point and draw in the empty space in front of them or on the palm of their hands. While they cannot see the results of their interaction, they obtain some visual and tactile feedback by watching and feeling their hands interact. After introducing the concept of Imaginary Interfaces, I present two hardware prototypes that showcase two different forms of interaction with an imaginary interface, each with its own advantages: mid-air imaginary interfaces can be large and expressive, while palm-based imaginary interfaces offer an abundance of tactile features that encourage learning. Given that imaginary interfaces offer no visual output, one of the key challenges is to enable users to discover the interface's layout. This dissertation offers three main solutions: offline learning with coordinates, browsing with audio feedback and learning by transfer. The latter I demonstrate with the Imaginary Phone, a palm-based imaginary interface that mimics the layout of a physical mobile phone that users are already familiar with. Although these designs enable interaction with Imaginary Interfaces, they tell us little about why this interaction is possible. In the final part of this dissertation, I present an exploration into which human perceptual abilities are used when interacting with a palm-based imaginary interface and how much each accounts for performance with the interface. These findings deepen our understanding of Imaginary Interfaces and suggest that palm-based Imaginary Interfaces can enable stand-alone eyes-free use for many applications, including interfaces for visually impaired users.},
  language  = {en}
}
@phdthesis{Holz2013,
  author    = {Holz, Christian},
  title     = {3D from 2D touch},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-67796},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {While interaction with computers used to be dominated by mice and keyboards, new types of sensors now allow users to interact through touch, speech, or using their whole body in 3D space. These new interaction modalities are often referred to as "natural user interfaces" or "NUIs." While 2D NUIs have experienced major success on billions of mobile touch devices sold, 3D NUI systems have so far been unable to deliver a mobile form factor, mainly due to their use of cameras. The fact that cameras require a certain distance from the capture volume has prevented 3D NUI systems from reaching the flat form factor mobile users expect. In this dissertation, we address this issue by sensing 3D input using flat 2D sensors. The systems we present observe the input from 3D objects as 2D imprints upon physical contact. By sampling these imprints at very high resolutions, we obtain the objects' textures. In some cases, a texture uniquely identifies a biometric feature, such as the user's fingerprint. In other cases, an imprint stems from the user's clothing, such as when walking on multitouch floors. By analyzing from which part of the 3D object the 2D imprint results, we reconstruct the object's pose in 3D space. While our main contribution is a general approach to sensing 3D input on 2D sensors upon physical contact, we also demonstrate three applications of our approach. (1) We present high-accuracy touch devices that allow users to reliably touch targets that are a third of the size of those on current touch devices. We show that different users and 3D finger poses systematically affect touch sensing, which current devices perceive as random input noise. We introduce a model for touch that compensates for this systematic effect by deriving the 3D finger pose and the user's identity from each touch imprint. We then investigate this systematic effect in detail and explore how users conceptually touch targets. Our findings indicate that users aim by aligning visual features of their fingers with the target. We present a visual model for touch input that eliminates virtually all systematic effects on touch accuracy. (2) From each touch, we identify users biometrically by analyzing their fingerprints. Our prototype Fiberio integrates fingerprint scanning and a display into the same flat surface, solving a long-standing problem in human-computer interaction: secure authentication on touchscreens. Sensing 3D input and authenticating users upon touch allows Fiberio to implement a variety of applications that traditionally require the bulky setups of current 3D NUI systems. (3) To demonstrate the versatility of 3D reconstruction on larger touch surfaces, we present a high-resolution pressure-sensitive floor that resolves the texture of objects upon touch. Using the same principles as before, our system GravitySpace analyzes all imprints and identifies users based on their shoe soles, detects furniture, and enables accurate touch input using feet. By classifying all imprints, GravitySpace detects the users' body parts that are in contact with the floor and then reconstructs their 3D body poses using inverse kinematics. GravitySpace thus enables a range of applications for future 3D NUI systems based on a flat sensor, such as smart rooms in future homes. We conclude this dissertation by projecting into the future of mobile devices. Focusing on the mobility aspect of our work, we explore how NUI devices may one day augment users directly in the form of implanted devices.},
  language  = {en}
}
@book{EidSabbaghHeweltWeske2013,
  author    = {Eid-Sabbagh, Rami-Habib and Hewelt, Marcin and Weske, Mathias},
  title     = {Business process architectures with multiplicities : transformation and correctness},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-257-5},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-66780},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {18},
  year      = {2013},
  abstract  = {Business processes are instrumental to manage work in organisations. To study the interdependencies between business processes, Business Process Architectures have been introduced. These express trigger and message ow relations between business processes. When we investigate real world Business Process Architectures, we find complex interdependencies, involving multiple process instances. These aspects have not been studied in detail so far, especially concerning correctness properties. In this paper, we propose a modular transformation of BPAs to open nets for the analysis of behavior involving multiple business processes with multiplicities. For this purpose we introduce intermediary nets to portray semantics of multiplicity specifications. We evaluate our approach on a use case from the public sector.},
  language  = {en}
}
@book{OPUS4-6986,
  title     = {HPI Future SOC Lab},
  editor    = {Meinel, Christoph and Polze, Andreas and Oswald, Gerhard and Strotmann, Rolf and Seibold, Ulrich and Schulzki, Bernard},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-276-6},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-68991},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {ii, 118},
  year      = {2013},
  abstract  = {The "HPI Future SOC Lab" is a cooperation of the Hasso-Plattner-Institut (HPI) and industrial partners. Its mission is to enable and promote exchange and interaction between the research community and the industrial partners. The HPI Future SOC Lab provides researchers with free of charge access to a complete infrastructure of state of the art hard- and software. This infrastructure includes components, which might be too expensive for an ordinary research environment, such as servers with up to 64 cores. The offerings address researchers particularly from but not limited to the areas of computer science and business information systems. Main areas of research include cloud computing, parallelization, and In-Memory technologies. This technical report presents results of research projects executed in 2012. Selected projects have presented their results on June 18th and November 26th 2012 at the Future SOC Lab Day events.},
  language  = {en}
}
@phdthesis{Gumienny2013,
  author    = {Gumienny, Raja Carola},
  title     = {Understanding the adoption of digital whiteboard systems for collaborative design work},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-72417},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {User-centered design processes are the first choice when new interactive systems or services are developed to address real customer needs and provide a good user experience. Common tools for collecting user research data, conducting brainstormings, or sketching ideas are whiteboards and sticky notes. They are ubiquitously available, and no technical or domain knowledge is necessary to use them. However, traditional pen and paper tools fall short when saving the content and sharing it with others unable to be in the same location. They are also missing further digital advantages such as searching or sorting content. Although research on digital whiteboard and sticky note applications has been conducted for over 20 years, these tools are not widely adopted in company contexts. While many research prototypes exist, they have not been used for an extended period of time in a real-world context. The goal of this thesis is to investigate what the enablers and obstacles for the adoption of digital whiteboard systems are. As an instrument for different studies, we developed the Tele-Board software system for collaborative creative work. Based on interviews, observations, and findings from former research, we tried to transfer the analog way of working to the digital world. Being a software system, Tele-Board can be used with a variety of hardware and does not depend on special devices. This feature became one of the main factors for adoption on a larger scale. In this thesis, I will present three studies on the use of Tele-Board with different user groups and foci. I will use a combination of research methods (laboratory case studies and data from field research) with the overall goal of finding out when a digital whiteboard system is used and in which cases not. Not surprisingly, the system is used and accepted if a user sees a main benefit that neither analog tools nor other applications can offer. However, I found that these perceived benefits are very different for each user and usage context. If a tool provides possibilities to use in different ways and with different equipment, the chances of its adoption by a larger group increase. Tele-Board has now been in use for over 1.5 years in a global IT company in at least five countries with a constantly growing user base. Its use, advantages, and disadvantages will be described based on 42 interviews and usage statistics from server logs. Through these insights and findings from laboratory case studies, I will present a detailed analysis of digital whiteboard use in different contexts with design implications for future systems.},
  language  = {en}
}
@phdthesis{Dawoud2013,
  author    = {Dawoud, Wesam},
  title     = {Scalability and performance management of internet applications in the cloud},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-68187},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {Cloud computing is a model for enabling on-demand access to a shared pool of computing resources. With virtually limitless on-demand resources, a cloud environment enables the hosted Internet application to quickly cope when there is an increase in the workload. However, the overhead of provisioning resources exposes the Internet application to periods of under-provisioning and performance degradation. Moreover, the performance interference, due to the consolidation in the cloud environment, complicates the performance management of the Internet applications. In this dissertation, we propose two approaches to mitigate the impact of the resources provisioning overhead. The first approach employs control theory to scale resources vertically and cope fast with workload. This approach assumes that the provider has knowledge and control over the platform running in the virtual machines (VMs), which limits it to Platform as a Service (PaaS) and Software as a Service (SaaS) providers. The second approach is a customer-side one that deals with the horizontal scalability in an Infrastructure as a Service (IaaS) model. It addresses the trade-off problem between cost and performance with a multi-goal optimization solution. This approach finds the scale thresholds that achieve the highest performance with the lowest increase in the cost. Moreover, the second approach employs a proposed time series forecasting algorithm to scale the application proactively and avoid under-utilization periods. Furthermore, to mitigate the interference impact on the Internet application performance, we developed a system which finds and eliminates the VMs suffering from performance interference. The developed system is a light-weight solution which does not imply provider involvement. To evaluate our approaches and the designed algorithms at large-scale level, we developed a simulator called (ScaleSim). In the simulator, we implemented scalability components acting as the scalability components of Amazon EC2. The current scalability implementation in Amazon EC2 is used as a reference point for evaluating the improvement in the scalable application performance. ScaleSim is fed with realistic models of the RUBiS benchmark extracted from the real environment. The workload is generated from the access logs of the 1998 world cup website. The results show that optimizing the scalability thresholds and adopting proactive scalability can mitigate 88\% of the resources provisioning overhead impact with only a 9\% increase in the cost.},
  language  = {en}
}
@book{OPUS4-6297,
  title     = {HPI Future SOC Lab},
  editor    = {Meinel, Christoph and Polze, Andreas and Oswald, Gerhard and Strotmann, Rolf and Seibold, Ulrich and D'Errico, Doc},
  isbn      = {978-3-86956-230-8},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-64004},
  publisher      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {Together with industrial partners Hasso-Plattner-Institut (HPI) is currently establishing a "HPI Future SOC Lab," which will provide a complete infrastructure for research on on-demand systems. The lab utilizes the latest, multi/many-core hardware and its practical implementation and testing as well as further development. The necessary components for such a highly ambitious project are provided by renowned companies: Fujitsu and Hewlett Packard provide their latest 4 and 8-way servers with 1-2 TB RAM, SAP will make available its latest Business byDesign (ByD) system in its most complete version. EMC² provides high performance storage systems and VMware offers virtualization solutions. The lab will operate on the basis of real data from large enterprises. The HPI Future SOC Lab, which will be open for use by interested researchers also from other universities, will provide an opportunity to study real-life complex systems and follow new ideas all the way to their practical implementation and testing. This technical report presents results of research projects executed in 2011. Selected projects have presented their results on June 15th and October 26th 2011 at the Future SOC Lab Day events.},
  language  = {en}
}
@article{RoschkeChengMeinel2013,
  author    = {Roschke, Sebastian and Cheng, Feng and Meinel, Christoph},
  title     = {High-quality attack graph-based IDS correlation},
  series = {Logic journal of the IGPL},
  volume    = {21},
  journal   = {Logic journal of the IGPL},
  number    = {4},
  publisher = {Oxford Univ. Press},
  address   = {Oxford},
  issn      = {1367-0751},
  doi       = {10.1093/jigpal/jzs034},
  pages     = {571 -- 591},
  year      = {2013},
  abstract  = {Intrusion Detection Systems are widely deployed in computer networks. As modern attacks are getting more sophisticated and the number of sensors and network nodes grow, the problem of false positives and alert analysis becomes more difficult to solve. Alert correlation was proposed to analyse alerts and to decrease false positives. Knowledge about the target system or environment is usually necessary for efficient alert correlation. For representing the environment information as well as potential exploits, the existing vulnerabilities and their Attack Graph (AG) is used. It is useful for networks to generate an AG and to organize certain vulnerabilities in a reasonable way. In this article, a correlation algorithm based on AGs is designed that is capable of detecting multiple attack scenarios for forensic analysis. It can be parameterized to adjust the robustness and accuracy. A formal model of the algorithm is presented and an implementation is tested to analyse the different parameters on a real set of alerts from a local network. To improve the speed of the algorithm, a multi-core version is proposed and a HMM-supported version can be used to further improve the quality. The parallel implementation is tested on a multi-core correlation platform, using CPUs and GPUs.},
  language  = {en}
}
@book{FelgentreffBorningHirschfeld2013,
  author    = {Felgentreff, Tim and Borning, Alan and Hirschfeld, Robert},
  title     = {Babelsberg : specifying and solving constraints on object behavior},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-265-0},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-67296},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {53},
  year      = {2013},
  abstract  = {Constraints allow developers to specify desired properties of systems in a number of domains, and have those properties be maintained automatically. This results in compact, declarative code, avoiding scattered code to check and imperatively re-satisfy invariants. Despite these advantages, constraint programming is not yet widespread, with standard imperative programming still the norm. There is a long history of research on integrating constraint programming with the imperative paradigm. However, this integration typically does not unify the constructs for encapsulation and abstraction from both paradigms. This impedes re-use of modules, as client code written in one paradigm can only use modules written to support that paradigm. Modules require redundant definitions if they are to be used in both paradigms. We present a language - Babelsberg - that unifies the constructs for en- capsulation and abstraction by using only object-oriented method definitions for both declarative and imperative code. Our prototype - Babelsberg/R - is an extension to Ruby, and continues to support Ruby's object-oriented se- mantics. It allows programmers to add constraints to existing Ruby programs in incremental steps by placing them on the results of normal object-oriented message sends. It is implemented by modifying a state-of-the-art Ruby virtual machine. The performance of standard object-oriented code without con- straints is only modestly impacted, with typically less than 10\% overhead compared with the unmodified virtual machine. Furthermore, our architec- ture for adding multiple constraint solvers allows Babelsberg to deal with constraints in a variety of domains. We argue that our approach provides a useful step toward making con- straint solving a generic tool for object-oriented programmers. We also provide example applications, written in our Ruby-based implementation, which use constraints in a variety of application domains, including interactive graphics, circuit simulations, data streaming with both hard and soft constraints on performance, and configuration file Management.},
  language  = {en}
}
@book{GieseBecker2013,
  author    = {Giese, Holger and Becker, Basil},
  title     = {Modeling and verifying dynamic evolving service-oriented architectures},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-246-9},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-65112},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {97},
  year      = {2013},
  abstract  = {The service-oriented architecture supports the dynamic assembly and runtime reconfiguration of complex open IT landscapes by means of runtime binding of service contracts, launching of new components and termination of outdated ones. Furthermore, the evolution of these IT landscapes is not restricted to exchanging components with other ones using the same service contracts, as new services contracts can be added as well. However, current approaches for modeling and verification of service-oriented architectures do not support these important capabilities to their full extend.In this report we present an extension of the current OMG proposal for service modeling with UML - SoaML - which overcomes these limitations. It permits modeling services and their service contracts at different levels of abstraction, provides a formal semantics for all modeling concepts, and enables verifying critical properties. Our compositional and incremental verification approach allows for complex properties including communication parameters and time and covers besides the dynamic binding of service contracts and the replacement of components also the evolution of the systems by means of new service contracts. The modeling as well as verification capabilities of the presented approach are demonstrated by means of a supply chain example and the verification results of a first prototype are shown.},
  language  = {en}
}
@book{RoggeSoltiMansvanderAalstetal.2013,
  author    = {Rogge-Solti, Andreas and Mans, Ronny S. and van der Aalst, Wil M. P. and Weske, Mathias},
  title     = {Repairing event logs using stochastic process models},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-258-2},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-66797},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {19},
  year      = {2013},
  abstract  = {Companies strive to improve their business processes in order to remain competitive. Process mining aims to infer meaningful insights from process-related data and attracted the attention of practitioners, tool-vendors, and researchers in recent years. Traditionally, event logs are assumed to describe the as-is situation. But this is not necessarily the case in environments where logging may be compromised due to manual logging. For example, hospital staff may need to manually enter information regarding the patient's treatment. As a result, events or timestamps may be missing or incorrect. In this paper, we make use of process knowledge captured in process models, and provide a method to repair missing events in the logs. This way, we facilitate analysis of incomplete logs. We realize the repair by combining stochastic Petri nets, alignments, and Bayesian networks. We evaluate the results using both synthetic data and real event data from a Dutch hospital.},
  language  = {en}
}
@phdthesis{Kunze2013,
  author    = {Kunze, Matthias},
  title     = {Searching business process models by example},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-68844},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {Business processes are fundamental to the operations of a company. Each product manufactured and every service provided is the result of a series of actions that constitute a business process. Business process management is an organizational principle that makes the processes of a company explicit and offers capabilities to implement procedures, control their execution, analyze their performance, and improve them. Therefore, business processes are documented as process models that capture these actions and their execution ordering, and make them accessible to stakeholders. As these models are an essential knowledge asset, they need to be managed effectively. In particular, the discovery and reuse of existing knowledge becomes challenging in the light of companies maintaining hundreds and thousands of process models. In practice, searching process models has been solved only superficially by means of free-text search of process names and their descriptions. Scientific contributions are limited in their scope, as they either present measures for process similarity or elaborate on query languages to search for particular aspects. However, they fall short in addressing efficient search, the presentation of search results, and the support to reuse discovered models. This thesis presents a novel search method, where a query is expressed by an exemplary business process model that describes the behavior of a possible answer. This method builds upon a formal framework that captures and compares the behavior of process models by the execution ordering of actions. The framework contributes a conceptual notion of behavioral distance that quantifies commonalities and differences of a pair of process models, and enables process model search. Based on behavioral distances, a set of measures is proposed that evaluate the quality of a particular search result to guide the user in assessing the returned matches. A projection of behavioral aspects to a process model enables highlighting relevant fragments that led to a match and facilitates its reuse. The thesis further elaborates on two search techniques that provide concrete behavioral distance functions as an instantiation of the formal framework. Querying enables search with a notion of behavioral inclusion with regard to the query. In contrast, similarity search obtains process models that are similar to a query, even if the query is not precisely matched. For both techniques, indexes are presented that enable efficient search. Methods to evaluate the quality and performance of process model search are introduced and applied to the techniques of this thesis. They show good results with regard to human assessment and scalability in a practical setting.},
  language  = {en}
}
@book{PapeTrefferHirschfeldetal.2013,
  author    = {Pape, Tobias and Treffer, Arian and Hirschfeld, Robert and Haupt, Michael},
  title     = {Extending a Java Virtual Machine to Dynamic Object-oriented Languages},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-266-7},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-67438},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {163},
  year      = {2013},
  abstract  = {There are two common approaches to implement a virtual machine (VM) for a dynamic object-oriented language. On the one hand, it can be implemented in a C-like language for best performance and maximum control over the resulting executable. On the other hand, it can be implemented in a language such as Java that allows for higher-level abstractions. These abstractions, such as proper object-oriented modularization, automatic memory management, or interfaces, are missing in C-like languages but they can simplify the implementation of prevalent but complex concepts in VMs, such as garbage collectors (GCs) or just-in-time compilers (JITs). Yet, the implementation of a dynamic object-oriented language in Java eventually results in two VMs on top of each other (double stack), which impedes performance. For statically typed languages, the Maxine VM solves this problem; it is written in Java but can be executed without a Java virtual machine (JVM). However, it is currently not possible to execute dynamic object-oriented languages in Maxine. This work presents an approach to bringing object models and execution models of dynamic object-oriented languages to the Maxine VM and the application of this approach to Squeak/Smalltalk. The representation of objects in and the execution of dynamic object-oriented languages pose certain challenges to the Maxine VM that lacks certain variation points necessary to enable an effortless and straightforward implementation of dynamic object-oriented languages' execution models. The implementation of Squeak/Smalltalk in Maxine as a feasibility study is to unveil such missing variation points.},
  language  = {en}
}
@book{SchwalbKruegerPlattner2013,
  author    = {Schwalb, David and Kr{\"u}ger, Jens and Plattner, Hasso},
  title     = {Cache conscious column organization in in-memory column stores},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-228-5},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-63890},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {v, 84},
  year      = {2013},
  abstract  = {Cost models are an essential part of database systems, as they are the basis of query performance optimization. Based on predictions made by cost models, the fastest query execution plan can be chosen and executed or algorithms can be tuned and optimised. In-memory databases shifts the focus from disk to main memory accesses and CPU costs, compared to disk based systems where input and output costs dominate the overall costs and other processing costs are often neglected. However, modelling memory accesses is fundamentally different and common models do not apply anymore. This work presents a detailed parameter evaluation for the plan operators scan with equality selection, scan with range selection, positional lookup and insert in in-memory column stores. Based on this evaluation, a cost model based on cache misses for estimating the runtime of the considered plan operators using different data structures is developed. Considered are uncompressed columns, bit compressed and dictionary encoded columns with sorted and unsorted dictionaries. Furthermore, tree indices on the columns and dictionaries are discussed. Finally, partitioned columns consisting of one partition with a sorted and one with an unsorted dictionary are investigated. New values are inserted in the unsorted dictionary partition and moved periodically by a merge process to the sorted partition. An efficient attribute merge algorithm is described, supporting the update performance required to run enterprise applications on read-optimised databases. Further, a memory traffic based cost model for the merge process is provided.},
  language  = {en}
}
@book{VogelGiese2013,
  author    = {Vogel, Thomas and Giese, Holger},
  title     = {Model-driven engineering of adaptation engines for self-adaptive software : executable runtime megamodels},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-227-8},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-63825},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {vi, 59},
  year      = {2013},
  abstract  = {The development of self-adaptive software requires the engineering of an adaptation engine that controls and adapts the underlying adaptable software by means of feedback loops. The adaptation engine often describes the adaptation by using runtime models representing relevant aspects of the adaptable software and particular activities such as analysis and planning that operate on these runtime models. To systematically address the interplay between runtime models and adaptation activities in adaptation engines, runtime megamodels have been proposed for self-adaptive software. A runtime megamodel is a specific runtime model whose elements are runtime models and adaptation activities. Thus, a megamodel captures the interplay between multiple models and between models and activities as well as the activation of the activities. In this article, we go one step further and present a modeling language for ExecUtable RuntimE MegAmodels (EUREMA) that considerably eases the development of adaptation engines by following a model-driven engineering approach. We provide a domain-specific modeling language and a runtime interpreter for adaptation engines, in particular for feedback loops. Megamodels are kept explicit and alive at runtime and by interpreting them, they are directly executed to run feedback loops. Additionally, they can be dynamically adjusted to adapt feedback loops. Thus, EUREMA supports development by making feedback loops, their runtime models, and adaptation activities explicit at a higher level of abstraction. Moreover, it enables complex solutions where multiple feedback loops interact or even operate on top of each other. Finally, it leverages the co-existence of self-adaptation and off-line adaptation for evolution.},
  language  = {en}
}
@phdthesis{Steinmetz2013,
  author    = {Steinmetz, Nadine},
  title     = {Context-aware semantic analysis of video metadata},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-70551},
  school      = {Universit{\"a}t Potsdam},
  year      = {2013},
  abstract  = {Im Vergleich zu einer stichwortbasierten Suche erm{\"o}glicht die semantische Suche ein pr{\"a}ziseres und anspruchsvolleres Durchsuchen von (Web)-Dokumenten, weil durch die explizite Semantik Mehrdeutigkeiten von nat{\"u}rlicher Sprache vermieden und semantische Beziehungen in das Suchergebnis einbezogen werden k{\"o}nnen. Eine semantische, Entit{\"a}ten-basierte Suche geht von einer Anfrage mit festgelegter Bedeutung aus und liefert nur Dokumente, die mit dieser Entit{\"a}t annotiert sind als Suchergebnis. Die wichtigste Voraussetzung f{\"u}r eine Entit{\"a}ten-zentrierte Suche stellt die Annotation der Dokumente im Archiv mit Entit{\"a}ten und Kategorien dar. Textuelle Informationen werden analysiert und mit den entsprechenden Entit{\"a}ten und Kategorien versehen, um den Inhalt semantisch erschließen zu k{\"o}nnen. Eine manuelle Annotation erfordert Dom{\"a}nenwissen und ist sehr zeitaufwendig. Die semantische Annotation von Videodokumenten erfordert besondere Aufmerksamkeit, da inhaltsbasierte Metadaten von Videos aus verschiedenen Quellen stammen, verschiedene Eigenschaften und Zuverl{\"a}ssigkeiten besitzen und daher nicht wie Fließtext behandelt werden k{\"o}nnen. Die vorliegende Arbeit stellt einen semantischen Analyseprozess f{\"u}r Video-Metadaten vor. Die Eigenschaften der verschiedenen Metadatentypen werden analysiert und ein Konfidenzwert ermittelt. Dieser Wert spiegelt die Korrektheit und die wahrscheinliche Mehrdeutigkeit eines Metadatums wieder. Beginnend mit dem Metadatum mit dem h{\"o}chsten Konfidenzwert wird der Analyseprozess innerhalb eines Kontexts in absteigender Reihenfolge des Konfidenzwerts durchgef{\"u}hrt. Die bereits analysierten Metadaten dienen als Referenzpunkt f{\"u}r die weiteren Analysen. So kann eine m{\"o}glichst korrekte Analyse der heterogen strukturierten Daten eines Kontexts sichergestellt werden. Am Ende der Analyse eines Metadatums wird die f{\"u}r den Kontext relevanteste Entit{\"a}t aus einer Liste von Kandidaten identifiziert - das Metadatum wird disambiguiert. Hierf{\"u}r wurden verschiedene Disambiguierungsalgorithmen entwickelt, die Beschreibungstexte und semantische Beziehungen der Entit{\"a}tenkandidaten zum gegebenen Kontext in Betracht ziehen. Der Kontext f{\"u}r die Disambiguierung wird f{\"u}r jedes Metadatum anhand der Eigenschaften und Konfidenzwerte zusammengestellt. Der vorgestellte Analyseprozess ist an zwei Hypothesen angelehnt: Um die Analyseergebnisse verbessern zu k{\"o}nnen, sollten die Metadaten eines Kontexts in absteigender Reihenfolge ihres Konfidenzwertes verarbeitet werden und die Kontextgrenzen von Videometadaten sollten durch Segmentgrenzen definiert werden, um m{\"o}glichst Kontexte mit koh{\"a}rentem Inhalt zu erhalten. Durch ausf{\"u}hrliche Evaluationen konnten die gestellten Hypothesen best{\"a}tigt werden. Der Analyseprozess wurden gegen mehrere State-of-the-Art Methoden verglichen und erzielt verbesserte Ergebnisse in Bezug auf Recall und Precision, besonders f{\"u}r Metadaten, die aus weniger zuverl{\"a}ssigen Quellen stammen. Der Analyseprozess ist Teil eines Videoanalyse-Frameworks und wurde bereits erfolgreich in verschiedenen Projekten eingesetzt.},
  language  = {en}
}
@book{MeinelWillems2013,
  author    = {Meinel, Christoph and Willems, Christian},
  title     = {openHPI : the MOOC offer at Hasso Plattner Institute},
  publisher = {Universit{\"a}tsverlag Potsdam},
  address   = {Potsdam},
  isbn      = {978-3-86956-264-3},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus-67176},
  publisher      = {Universit{\"a}t Potsdam},
  pages     = {21},
  year      = {2013},
  abstract  = {The new interactive online educational platform openHPI, (https://openHPI.de) from Hasso Plattner Institute (HPI), offers freely accessible courses at no charge for all who are interested in subjects in the field of information technology and computer science. Since 2011, "Massive Open Online Courses," called MOOCs for short, have been offered, first at Stanford University and then later at other U.S. elite universities. Following suit, openHPI provides instructional videos on the Internet and further reading material, combined with learning-supportive self-tests, homework and a social discussion forum. Education is further stimulated by the support of a virtual learning community. In contrast to "traditional" lecture platforms, such as the tele-TASK portal (http://www.tele-task.de) where multimedia recorded lectures are available on demand, openHPI offers didactic online courses. The courses have a fixed start date and offer a balanced schedule of six consecutive weeks presented in multimedia and, whenever possible, interactive learning material. Each week, one chapter of the course subject is treated. In addition, a series of learning videos, texts, self-tests and homework exercises are provided to course participants at the beginning of the week. The course offering is combined with a social discussion platform where participants have the opportunity to enter into an exchange with course instructors and fellow participants. Here, for example, they can get answers to questions and discuss the topics in depth. The participants naturally decide themselves about the type and range of their learning activities. They can make personal contributions to the course, for example, in blog posts or tweets, which they can refer to in the forum. In turn, other participants have the chance to comment on, discuss or expand on what has been said. In this way, the learners become the teachers and the subject matter offered to a virtual community is linked to a social learning network.},
  language  = {en}
}