http://idea.library.drexel.edu/handle/1860/804 Search the Channel search http://idea.library.drexel.edu/simple-search http://idea.library.drexel.edu/handle/1860/4050 Title: Content-Based Music Genre Classification Using Sparse Approximation Techniques <br/> <br/>Authors: Aryafar, Kamelia; Adams, Trevor R.; Shokoufandeh, Ali <br/> <br/>Abstract: In this study we evaluated the performance of genre classification systems using various feature vectors and learning methods. Using a fixed classifier, i.e., the Gaussian mixture models we were able to create a suboptimal feature vector to characterize the audio signals in a low dimensional feature space. We then utilized this modified feature representation to solve the problem of music genre classification. We evaluated the performance of the recent sparsity-eager support vector machines classifier using the proposed feature vector and compared the results to the classic support vector machines and Gaussian mixture models as the baseline classifiers. http://idea.library.drexel.edu/handle/1860/4049 Title: Dominant Color Learning by Subject Extraction <br/> <br/>Authors: Aryafar, Kamelia; Attenberg, Josh; Condon, Fiona <br/> <br/>Abstract: Advances in the digital media industry have resulted in an exponential growth in available image data sets. This exponential growth has in turn spurred great interest in various methods for acquiring, processing, analyzing, and understanding images in order to produce numerical or symbolic information such as color and texture characteristics. Detecting the dominant color of an object in the image without any prior knowledge about the background model, the object characteristics or the scene geometry is a challenging problem. The two major challenges in assigning a dominant color to the image subject are the isolation of the subject by background subtraction and the extraction of dominant color from the approximated subject region. In this work, we combine an estimated subject mask with the image color histogram to detect the dominant image color. http://idea.library.drexel.edu/handle/1860/4048 Title: Automatic Classification of Digital Music by Genre <br/> <br/>Authors: Aryafar, Kamelia; Shokoufandeh, Ali <br/> <br/>Abstract: Over the past two decades, advances in the digital music industry have resulted in an exponential growth in music data sets. This exponential growth has in turn spurred great interest in music information retrieval (MIR) problems, organizing large music collections, and content-based search methods for digital music libraries. Equally important are the related problems in music classification such as genre classification, music mood analysis, and artist identification. Music genre classification is a well-studied problem in the music information retrieval community and has a wide range of applications. In this project we address the problem of genre classification by representing the MFCC feature vectors in an extended semantic space. We combine this audio representation with machine learning techniques to perform genre classification with the goal of obtaining higher classification accuracy. http://idea.library.drexel.edu/handle/1860/4047 Title: Music Genre Classification Using Explicit Semantic Analysis <br/> <br/>Authors: Aryafar, Kamelia; Shokoufandeh, Ali <br/> <br/>Abstract: Music genre classification is the categorization of a piece of music into its corresponding categorical labels created by humans and has been traditionally performed through a manual process. Automatic music genre classification, a fundamental problem in the musical information retrieval community, has been gaining more attention with advances in the development of the digital music industry. Most current genre classification methods tend to be based on the extraction of short-time features in combination with high-level audio features to perform genre classification. However, the representation of short-time features, using time windows, in a semantic space has received little attention. This paper proposes a vector space model of mel-frequency cepstral coefficients (MFCCs) that can, in turn, be used by a supervised learning schema for music genre classification. Inspired by explicit semantic analysis of textual documents using term frequency-inverse document frequency (tf-idf), a semantic space model is proposed to represent music samples. The effectiveness of this representation of audio samples is then demonstrated in music genre classification using various machine learning classification algorithms, including support vector machines (SVMs) and k-nearest neighbor clustering. Our preliminary results suggest that the proposed method is comparable to genre classification methods that use low-level audio features. http://idea.library.drexel.edu/handle/1860/2717 Title: The Athens affair <br/> <br/>Authors: Prevelakis, Vassilis; Spinellis, Diomidis http://idea.library.drexel.edu/handle/1860/2701 Title: Mixture of spherical distributions for single-view relighting <br/> <br/>Authors: Hara, Kenji; Nishino, Ko; Ikeuchi, Katsushi <br/> <br/>Abstract: Wepresent a method for simultaneously estimating the illumination of a scene and the reflectance property of an object from single view images—a single image or a small number of images taken from the same viewpoint.Weassume that the illumination consists of multiple point light sources, and the shape of the object is known. First, we represent the illumination on the surface of a unit sphere as a finite mixture of von Mises-Fisher distributions based on a novel spherical specular reflection model that well approximates the Torrance- Sparrow reflection model. Next, we estimate the parameters of this mixture model including the number of its component distributions and the standard deviation of them, which correspond to the number of light sources and the surface roughness, respectively. Finally, using these results as the initial estimates, we iteratively refine the estimates based on the original Torrance-Sparrow reflection model. The final estimates can be used to relight single-view images such as altering the intensities and directions of the individual light sources. The proposed method provides a unified framework based on directional statistics for simultaneously estimating the intensities and directions of an unknown number of light sources, as well as the specular reflection parameter of the object in the scene. http://idea.library.drexel.edu/handle/1860/2590 Title: Lane-change detection using a computational driver model <br/> <br/>Authors: Salvucci, Dario D.; Mandalia, Hiren M.; Kuge, Nobuyuki; Yamamura, Tomohiro <br/> <br/>Abstract: Objective: This paper introduces a robust, real-time system for detecting driver lane changes. Background: As intelligent transportation systems evolve to assist drivers in their intended behaviors, the systems have demonstrated a need for methods of inferring driver intentions and detecting intended maneuvers. Method: Using a “model tracing” methodology, our system simulates a set of possible driver intentions and their resulting behaviors using a simplification of a previously validated computational model of driver behavior. The system compares the model’s simulated behavior with a driver’s actual observed behavior and thus continually infers the driver’s unobservable intentions from her or his observable actions. Results: For data collected in a driving simulator, the system detects 82% of lane changes within 0.5 s of maneuver onset (assuming a 5% false alarm rate), 93% within 1 s, and 95% before the vehicle moves one fourth of the lane width laterally. For data collected from an instrumented vehicle, the system detects 61% within 0.5 s, 77% within 1 s, and 84% before the vehicle moves one-fourth of the lane width laterally. Conclusion: The model-tracing system is the first system to demonstrate high sample-by-sample accuracy at low false alarm rates as well as high accuracy over the course of a lane change with respect to time and lateral movement. Application: By providing robust realtime detection of driver lane changes, the system shows good promise for incorporation into the next generation of intelligent transportation systems. http://idea.library.drexel.edu/handle/1860/2575 Title: A probabilistic approach to source code authorship identification <br/> <br/>Authors: Kothari, Jay; Shevertalov, Maxim; Stehle, Edward; Mancoridis, Spiros <br/> <br/>Abstract: There exists a need for tools to help identify the authorship of source code. This includes situations in which the ownership of code is questionable, such as in plagiarism or intellectual property infringement disputes. Authorship identification can also be used to assist in the apprehension of the creators of malware. In this paper we present an approach to identifying the authors of source code. We begin by computing a set of metrics to build profiles for a population of known authors using code samples that are verified to be authentic. We then compute metrics on unidentified source code to determine the closest matching profile. We demonstrate our approach on a case study that involves two kinds of software: one based on open source developers working on various projects, and another based on students working on assignments with the same requirements. In our case study we are able to determine authorship with greater than 70% accuracy in choosing the single nearest match and greater than 90% accuracy in choosing the top three ordered nearest matches. http://idea.library.drexel.edu/handle/1860/2571 Title: Reducing program comprehension effort in evolving software by recognizing feature implementation convergence <br/> <br/>Authors: Kothari, Jay; Denton, Trip; Shokoufandeh, Ali; Mancoridis, Spiros <br/> <br/>Abstract: The implementations of software features evolve as an application matures. We define a measure of feature implementation overlap that determines how similar features are in their execution by examining their call graphs. We consider how this measure changes over time, and evaluate the hypothesis that over time and subsequent versions of a software application, the implementations of semantically similar features converge. As the features of an application converge in their implementation, we are able to more effectively determine groups of semantically similar features and to reduce the cost of program comprehension by selecting few key features that give an overview of the system. We present a case study analyzing the features of the Jext, Firefox, and Gaim software systems to support our hypothesis. http://idea.library.drexel.edu/handle/1860/2569 Title: On computing the canonical features of software systems <br/> <br/>Authors: Kothari, Jay; Denton, Trip; Mancoridis, Spiros; Shokoufandeh, Ali <br/> <br/>Abstract: Software applications typically have many features that vary in their similarity. We define a measurement of similarity between pairs of features based on their underlying implementations and use this measurement to compute a set of canonical features. The Canonical Features Set (CFS) consists of a small number of features that are as dissimilar as possible to each other, yet are most representative of the features that are not in the CFS. The members of the CFS are distinguishing features and understanding their implementation provides the engineer with an overview of the system undergoing scrutiny. The members of the CFS can also be used as cluster centroids to partition the entire set of features. Partitioning the set of features can simplify the understanding of large and complex software systems. Additionally, when a specific feature must undergo maintenance, it is helpful to know which features are most closely related to it. We demonstrate the utility of our method through the analysis of the Jext, Firefox, and Gaim software systems.