@inproceedings{10789, author = {{Trier, Matthias and Fung, Magdalene and Hansen, Abigail}}, booktitle = {{25th European Conference on Information Systems, {ECIS} 2017, Guimar{\~{a}}es, Portugal, June 5-10, 2017}}, pages = {{104}}, title = {{{Uncertainties as Barriers for Knowledge Sharing with Enterprise Social Media}}}, year = {{2017}}, } @misc{1079, author = {{Hamacher, Dustin Stefan}}, publisher = {{Universität Paderborn}}, title = {{{Das Zusammenschlussvorhaben von Edeka und Kaiser's Tengelmann - eine ökonomische Analyse}}}, year = {{2017}}, } @misc{1080, author = {{Bürmann, Jan}}, publisher = {{Universität Paderborn}}, title = {{{Complexity of Signalling in Routing Games under Uncertainty}}}, year = {{2017}}, } @misc{1081, author = {{Vijayalakshmi, Vipin Ravindran}}, publisher = {{Universität Paderborn}}, title = {{{Bounding the Inefficiency of Equilibria in Congestion Games under Taxation}}}, year = {{2017}}, } @techreport{1083, abstract = {{In actual school choice applications the theoretical underpinnings of the Boston School Choice Mechanism (BM) (complete information and rationality of the agents) are often not given. We analyze the actual behavior of agents in such a matching mechanism, using data from the matching mechanism currently used in a clearinghouse at a faculty of Business Administration and Economics at a German university, where a variant of the BM is used, and supplement this data with data generated in a survey among students who participated in the clearinghouse. We find that under the current mechanism over 70% of students act strategically. Controlling for students' limited information, we find that they do act rationally in their decision to act strategically. While students thus seem to react to the incentives to act strategically under the BM, they do not seem to be able to use this to their own advantage. However, those students acting in line with their beliefs manage a significantly better personal outcome than those who do not. We also run simulations by using a variant of the deferred acceptance algorithm, adapted to our situation, to show that the use of a different algorithm may be to the students' advantage.}}, author = {{Hoyer, Britta and Stroh-Maraun, Nadja}}, publisher = {{CIE Working Paper Series, Paderborn University}}, title = {{{Matching Strategies of Heterogeneous Agents under Incomplete Information in a University Clearinghouse}}}, volume = {{110}}, year = {{2017}}, } @misc{109, author = {{Pauck, Felix}}, publisher = {{Universität Paderborn}}, title = {{{Cooperative static analysis of Android applications}}}, year = {{2017}}, } @inproceedings{1094, abstract = {{Many university students struggle with motivational problems, and gamification has the potential to address these problems. However, gamification is hardly used in education, because current approaches to gamification require instructors to engage in the time-consuming preparation of their course contents for use in quizzes, mini-games and the like. Drawing on research on limited attention and present bias, we propose a "lean" approach to gamification, which relies on gamifying learning activities (rather than learning contents) and increasing their salience. In this paper, we present the app StudyNow that implements such a lean gamification approach. With this app, we aim to enable more students and instructors to benefit from the advantages of gamification.}}, author = {{Feldotto, Matthias and John, Thomas and Kundisch, Dennis and Hemsen, Paul and Klingsieck, Katrin and Skopalik, Alexander}}, booktitle = {{Proceedings of the 12th International Conference on Design Science Research in Information Systems and Technology (DESRIST)}}, pages = {{462--467}}, title = {{{Making Gamification Easy for the Professor: Decoupling Game and Content with the StudyNow Mobile App}}}, doi = {{10.1007/978-3-319-59144-5_32}}, year = {{2017}}, } @inproceedings{1095, abstract = {{Many university students struggle with motivational problems, and gamification has the potential to address these problems. However, using gamification currently is rather tedious and time-consuming for instructors because current approaches to gamification require instructors to engage in the time-consuming preparation of course contents (e.g., for quizzes or mini-games). In reply to this issue, we propose a “lean” approach to gamification, which relies on gamifying learning activities rather than learning contents. The learning activities that are gamified in the lean approach can typically be drawn from existing course syllabi (e.g., attend certain lectures, hand in assignments, read book chapters and articles). Hence, compared to existing approaches, lean gamification substantially lowers the time requirements posed on instructors for gamifying a given course. Drawing on research on limited attention and the present bias, we provide the theoretical foundation for the lean gamification approach. In addition, we present a mobile application that implements lean gamification and outline a mixed-methods study that is currently under way for evaluating whether lean gamification does indeed have the potential to increase students’ motivation. We thereby hope to allow more students and instructors to benefit from the advantages of gamification. }}, author = {{John, Thomas and Feldotto, Matthias and Hemsen, Paul and Klingsieck, Katrin and Kundisch, Dennis and Langendorf, Mike}}, booktitle = {{Proceedings of the 25th European Conference on Information Systems (ECIS)}}, pages = {{2970--2979}}, title = {{{Towards a Lean Approach for Gamifying Education}}}, year = {{2017}}, } @article{1098, abstract = {{An end user generally writes down software requirements in ambiguous expressions using natural language; hence, a software developer attuned to programming language finds it difficult to understand th meaning of the requirements. To solve this problem we define semantic categories for disambiguation and classify/annotate the requirement into the categories by using machine-learning models. We extensively use a language frame closely related to such categories for designing features to overcome the problem of insufficient training data compare to the large number of classes. Our proposed model obtained a micro-average F1-score of 0.75, outperforming the previous model, REaCT.}}, author = {{Kim, Yeong-Su and Lee, Seung-Woo and Dollmann, Markus and Geierhos, Michaela}}, issn = {{2205-8494}}, journal = {{International Journal of Software Engineering for Smart Device}}, keywords = {{Natural Language Processing, Semantic Annotation, Machine Learning}}, number = {{2}}, pages = {{1--6}}, publisher = {{Global Vision School Publication}}, title = {{{Semantic Annotation of Software Requirements with Language Frame}}}, volume = {{4}}, year = {{2017}}, } @article{110, abstract = {{We consider an extension of the dynamic speed scaling scheduling model introduced by Yao et al.: A set of jobs, each with a release time, deadline, and workload, has to be scheduled on a single, speed-scalable processor. Both the maximum allowed speed of the processor and the energy costs may vary continuously over time. The objective is to find a feasible schedule that minimizes the total energy costs. Theoretical algorithm design for speed scaling problems often tends to discretize problems, as our tools in the discrete realm are often better developed or understood. Using the above speed scaling variant with variable, continuous maximal processor speeds and energy prices as an example, we demonstrate that a more direct approach via tools from variational calculus can not only lead to a very concise and elegant formulation and analysis, but also avoids the “explosion of variables/constraints” that often comes with discretizing. Using well-known tools from calculus of variations, we derive combinatorial optimality characteristics for our continuous problem and provide a quite concise and simple correctness proof.}}, author = {{Antoniadis, Antonios and Kling, Peter and Ott, Sebastian and Riechers, Sören}}, journal = {{Theoretical Computer Science}}, pages = {{1--13}}, publisher = {{Elsevier}}, title = {{{Continuous Speed Scaling with Variability: A Simple and Direct Approach}}}, doi = {{10.1016/j.tcs.2017.03.021}}, year = {{2017}}, } @misc{117, author = {{Bemmann, Pascal}}, publisher = {{Universität Paderborn}}, title = {{{Attribute-based Signatures using Structure Preserving Signatures}}}, year = {{2017}}, } @inproceedings{11717, abstract = {{In this work, we address the limited availability of large annotated databases for real-life audio event detection by utilizing the concept of transfer learning. This technique aims to transfer knowledge from a source domain to a target domain, even if source and target have different feature distributions and label sets. We hypothesize that all acoustic events share the same inventory of basic acoustic building blocks and differ only in the temporal order of these acoustic units. We then construct a deep neural network with convolutional layers for extracting the acoustic units and a recurrent layer for capturing the temporal order. Under the above hypothesis, transfer learning from a source to a target domain with a different acoustic event inventory is realized by transferring the convolutional layers from the source to the target domain. The recurrent layer is, however, learnt directly from the target domain. Experiments on the transfer from a synthetic source database to the reallife target database of DCASE 2016 demonstrate that transfer learning leads to improved detection performance on average. However, the successful transfer to detect events which are very different from what was seen in the source domain, could not be verified.}}, author = {{Arora, Prerna and Haeb-Umbach, Reinhold}}, booktitle = {{IEEE 19th International Workshop on Multimedia Signal Processing (MMSP)}}, title = {{{A Study on Transfer Learning for Acoustic Event Detection in a Real Life Scenario}}}, year = {{2017}}, } @techreport{11735, abstract = {{This report describes the computation of gradients by algorithmic differentiation for statistically optimum beamforming operations. Especially the derivation of complex-valued functions is a key component of this approach. Therefore the real-valued algorithmic differentiation is extended via the complex-valued chain rule. In addition to the basic mathematic operations the derivative of the eigenvalue problem with complex-valued eigenvectors is one of the key results of this report. The potential of this approach is shown with experimental results on the CHiME-3 challenge database. There, the beamforming task is used as a front-end for an ASR system. With the developed derivatives a joint optimization of a speech enhancement and speech recognition system w.r.t. the recognition optimization criterion is possible.}}, author = {{Boeddeker, Christoph and Hanebrink, Patrick and Drude, Lukas and Heymann, Jahn and Haeb-Umbach, Reinhold}}, title = {{{On the Computation of Complex-valued Gradients with Application to Statistically Optimum Beamforming}}}, year = {{2017}}, } @inproceedings{11736, abstract = {{In this paper we show how a neural network for spectral mask estimation for an acoustic beamformer can be optimized by algorithmic differentiation. Using the beamformer output SNR as the objective function to maximize, the gradient is propagated through the beamformer all the way to the neural network which provides the clean speech and noise masks from which the beamformer coefficients are estimated by eigenvalue decomposition. A key theoretical result is the derivative of an eigenvalue problem involving complex-valued eigenvectors. Experimental results on the CHiME-3 challenge database demonstrate the effectiveness of the approach. The tools developed in this paper are a key component for an end-to-end optimization of speech enhancement and speech recognition.}}, author = {{Boeddeker, Christoph and Hanebrink, Patrick and Drude, Lukas and Heymann, Jahn and Haeb-Umbach, Reinhold}}, booktitle = {{Proc. IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP)}}, title = {{{Optimizing Neural-Network Supported Acoustic Beamforming by Algorithmic Differentiation}}}, year = {{2017}}, } @inproceedings{11737, abstract = {{The benefits of both a logarithmic spectral amplitude (LSA) estimation and a modeling in a generalized spectral domain (where short-time amplitudes are raised to a generalized power exponent, not restricted to magnitude or power spectrum) are combined in this contribution to achieve a better tradeoff between speech quality and noise suppression in single-channel speech enhancement. A novel gain function is derived to enhance the logarithmic generalized spectral amplitudes of noisy speech. Experiments on the CHiME-3 dataset show that it outperforms the famous minimum mean squared error (MMSE) LSA gain function of Ephraim and Malah in terms of noise suppression by 1.4 dB, while the good speech quality of the MMSE-LSA estimator is maintained.}}, author = {{Chinaev, Alleksej and Haeb-Umbach, Reinhold}}, booktitle = {{Proc. IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP)}}, title = {{{A Generalized Log-Spectral Amplitude Estimator for Single-Channel Speech Enhancement}}}, year = {{2017}}, } @inproceedings{11754, abstract = {{Recent advances in discriminatively trained mask estimation networks to extract a single source utilizing beamforming techniques demonstrate, that the integration of statistical models and deep neural networks (DNNs) are a promising approach for robust automatic speech recognition (ASR) applications. In this contribution we demonstrate how discriminatively trained embeddings on spectral features can be tightly integrated into statistical model-based source separation to separate and transcribe overlapping speech. Good generalization to unseen spatial configurations is achieved by estimating a statistical model at test time, while still leveraging discriminative training of deep clustering embeddings on a separate training set. We formulate an expectation maximization (EM) algorithm which jointly estimates a model for deep clustering embeddings and complex-valued spatial observations in the short time Fourier transform (STFT) domain at test time. Extensive simulations confirm, that the integrated model outperforms (a) a deep clustering model with a subsequent beamforming step and (b) an EM-based model with a beamforming step alone in terms of signal to distortion ratio (SDR) and perceptually motivated metric (PESQ) gains. ASR results on a reverberated dataset further show, that the aforementioned gains translate to reduced word error rates (WERs) even in reverberant environments.}}, author = {{Drude, Lukas and Haeb-Umbach, Reinhold}}, booktitle = {{INTERSPEECH 2017, Stockholm, Schweden}}, title = {{{Tight integration of spatial and spectral features for BSS with Deep Clustering embeddings}}}, year = {{2017}}, } @inproceedings{11770, abstract = {{In this contribution we show how to exploit text data to support word discovery from audio input in an underresourced target language. Given audio, of which a certain amount is transcribed at the word level, and additional unrelated text data, the approach is able to learn a probabilistic mapping from acoustic units to characters and utilize it to segment the audio data into words without the need of a pronunciation dictionary. This is achieved by three components: an unsupervised acoustic unit discovery system, a supervisedly trained acoustic unit-to-grapheme converter, and a word discovery system, which is initialized with a language model trained on the text data. Experiments for multiple setups show that the initialization of the language model with text data improves the word segementation performance by a large margin.}}, author = {{Glarner, Thomas and Boenninghoff, Benedikt and Walter, Oliver and Haeb-Umbach, Reinhold}}, booktitle = {{INTERSPEECH 2017, Stockholm, Schweden}}, title = {{{Leveraging Text Data for Word Segmentation for Underresourced Languages}}}, year = {{2017}}, } @misc{118, author = {{Chi Banh, Ngoc}}, publisher = {{Universität Paderborn}}, title = {{{An Asynchronous Adaptation of a Churn-resistant Overlay Network}}}, year = {{2017}}, } @inproceedings{1180, abstract = {{These days, there is a strong rise in the needs for machine learning applications, requiring an automation of machine learning engineering which is referred to as AutoML. In AutoML the selection, composition and parametrization of machine learning algorithms is automated and tailored to a specific problem, resulting in a machine learning pipeline. Current approaches reduce the AutoML problem to optimization of hyperparameters. Based on recursive task networks, in this paper we present one approach from the field of automated planning and one evolutionary optimization approach. Instead of simply parametrizing a given pipeline, this allows for structure optimization of machine learning pipelines, as well. We evaluate the two approaches in an extensive evaluation, finding both approaches to have their strengths in different areas. Moreover, the two approaches outperform the state-of-the-art tool Auto-WEKA in many settings.}}, author = {{Wever, Marcel Dominik and Mohr, Felix and Hüllermeier, Eyke}}, booktitle = {{27th Workshop Computational Intelligence}}, location = {{Dortmund}}, title = {{{Automatic Machine Learning: Hierachical Planning Versus Evolutionary Optimization}}}, year = {{2017}}, } @inproceedings{11809, abstract = {{This paper presents an end-to-end training approach for a beamformer-supported multi-channel ASR system. A neural network which estimates masks for a statistically optimum beamformer is jointly trained with a network for acoustic modeling. To update its parameters, we propagate the gradients from the acoustic model all the way through feature extraction and the complex valued beamforming operation. Besides avoiding a mismatch between the front-end and the back-end, this approach also eliminates the need for stereo data, i.e., the parallel availability of clean and noisy versions of the signals. Instead, it can be trained with real noisy multichannel data only. Also, relying on the signal statistics for beamforming, the approach makes no assumptions on the configuration of the microphone array. We further observe a performance gain through joint training in terms of word error rate in an evaluation of the system on the CHiME 4 dataset.}}, author = {{Heymann, Jahn and Drude, Lukas and Boeddeker, Christoph and Hanebrink, Patrick and Haeb-Umbach, Reinhold}}, booktitle = {{Proc. IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP)}}, title = {{{BEAMNET: End-to-End Training of a Beamformer-Supported Multi-Channel ASR System}}}, year = {{2017}}, }