Matching Items (29)
Description
Lecture videos are a widely used resource for learning. A simple way to create
videos is to record live lectures, but these videos end up being lengthy, include long
pauses and repetitive words making the viewing experience time consuming. While
pauses are useful in live learning environments where students take notes, I question
the value of pauses in video lectures. Techniques and algorithms that can shorten such
videos can have a huge impact in saving students’ time and reducing storage space.
I study this problem of shortening videos by removing long pauses and adaptively
modifying the playback rate by emphasizing the most important sections of the video
and its effect on the student community. The playback rate is designed in such a
way to play uneventful sections faster and significant sections slower. Important and
unimportant sections of a video are identified using textual analysis. I use an existing
speech-to-text algorithm to extract the transcript and apply latent semantic analysis
and standard information retrieval techniques to identify the relevant segments of
the video. I compute relevance scores of different segments and propose a variable
playback rate for each of these segments. The aim is to reduce the amount of time
students spend on passive learning while watching videos without harming their ability
to follow the lecture. I validate the approach by conducting a user study among
computer science students and measuring their engagement. The results indicate
no significant difference in their engagement when this method is compared to the
original unedited video.
videos is to record live lectures, but these videos end up being lengthy, include long
pauses and repetitive words making the viewing experience time consuming. While
pauses are useful in live learning environments where students take notes, I question
the value of pauses in video lectures. Techniques and algorithms that can shorten such
videos can have a huge impact in saving students’ time and reducing storage space.
I study this problem of shortening videos by removing long pauses and adaptively
modifying the playback rate by emphasizing the most important sections of the video
and its effect on the student community. The playback rate is designed in such a
way to play uneventful sections faster and significant sections slower. Important and
unimportant sections of a video are identified using textual analysis. I use an existing
speech-to-text algorithm to extract the transcript and apply latent semantic analysis
and standard information retrieval techniques to identify the relevant segments of
the video. I compute relevance scores of different segments and propose a variable
playback rate for each of these segments. The aim is to reduce the amount of time
students spend on passive learning while watching videos without harming their ability
to follow the lecture. I validate the approach by conducting a user study among
computer science students and measuring their engagement. The results indicate
no significant difference in their engagement when this method is compared to the
original unedited video.
ContributorsPurushothama Shenoy, Sreenivas (Author) / Amresh, Ashish (Thesis advisor) / Femiani, John (Committee member) / Walker, Erin (Committee member) / Arizona State University (Publisher)
Created2016
Description
Large datasets of sub-meter aerial imagery represented as orthophoto mosaics are widely available today, and these data sets may hold a great deal of untapped information. This imagery has a potential to locate several types of features; for example, forests, parking lots, airports, residential areas, or freeways in the imagery. However, the appearances of these things vary based on many things including the time that the image is captured, the sensor settings, processing done to rectify the image, and the geographical and cultural context of the region captured by the image. This thesis explores the use of deep convolutional neural networks to classify land use from very high spatial resolution (VHR), orthorectified, visible band multispectral imagery. Recent technological and commercial applications have driven the collection a massive amount of VHR images in the visible red, green, blue (RGB) spectral bands, this work explores the potential for deep learning algorithms to exploit this imagery for automatic land use/ land cover (LULC) classification. The benefits of automatic visible band VHR LULC classifications may include applications such as automatic change detection or mapping. Recent work has shown the potential of Deep Learning approaches for land use classification; however, this thesis improves on the state-of-the-art by applying additional dataset augmenting approaches that are well suited for geospatial data. Furthermore, the generalizability of the classifiers is tested by extensively evaluating the classifiers on unseen datasets and we present the accuracy levels of the classifier in order to show that the results actually generalize beyond the small benchmarks used in training. Deep networks have many parameters, and therefore they are often built with very large sets of labeled data. Suitably large datasets for LULC are not easy to come by, but techniques such as refinement learning allow networks trained for one task to be retrained to perform another recognition task. Contributions of this thesis include demonstrating that deep networks trained for image recognition in one task (ImageNet) can be efficiently transferred to remote sensing applications and perform as well or better than manually crafted classifiers without requiring massive training data sets. This is demonstrated on the UC Merced dataset, where 96% mean accuracy is achieved using a CNN (Convolutional Neural Network) and 5-fold cross validation. These results are further tested on unrelated VHR images at the same resolution as the training set.
ContributorsUba, Nagesh Kumar (Author) / Femiani, John (Thesis advisor) / Razdan, Anshuman (Committee member) / Amresh, Ashish (Committee member) / Arizona State University (Publisher)
Created2016
Description
In the last decade, the number of people who own a mobile phone or portable electronic communication device has grown exponentially. Recent advances in smartphone technology have enabled mobile devices to provide applications (“mHealth apps”) to support delivering interventions, tracking health treatments, or involving a healthcare team into the treatment process and symptom monitoring. Although the popularity of mHealth apps is increasing, few lessons have been shared regarding user experience design and evaluation for such innovations as they relate to clinical outcomes. Studies assessing usability for mobile apps primarily rely on survey instruments. Though surveys are effective in determining user perception of usability and positive attitudes towards an app, they do not directly assess app feature usage, and whether feature usage and related aspects of app design are indicative of whether intended tasks are completed by users. This is significant in the area of mHealth apps, as proper utilization of the app determines compliance to a clinical study protocol. Therefore it is important to understand how design directly impacts compliance, specifically what design factors are prevalent in non-compliant users. This research studies the impact of usability features on clinical protocol compliance by applying a mixed methods approach to usability assessment, combining traditional surveys, log analysis, and clickstream analysis to determine the connection of design to outcomes. This research is novel in its construction of the mixed methods approach and in its attempt to tie usability results to impacts on clinical protocol compliance. The validation is a case study approach, applying the methods to an mHealth app developed for early prevention of anxiety in middle school students. The results of three empirical studies are shared that support the construction of the mixed methods approach.
ContributorsPatwardhan, Mandar (Author) / Gary, Kevin A (Thesis advisor) / Pina, Armando (Committee member) / Amresh, Ashish (Committee member) / Arizona State University (Publisher)
Created2016
Description
For the past decade, mobile health applications are seeing greater acceptance due to their potential to remotely monitor and increase patient engagement, particularly for chronic disease. Sickle Cell Disease is an inherited chronic disorder of red blood cells requiring careful pain management. A significant number of mHealth applications have been developed in the market to help clinicians collect and monitor information of SCD patients. Surveys are the most common way to self-report patient conditions. These are non-engaging and suffer from poor compliance. The quality of data gathered from survey instruments while using technology can be questioned as patients may be motivated to complete a task but not motivated to do it well. A compromise in quality and quantity of the collected patient data hinders the clinicians' effort to be able to monitor patient's health on a regular basis and derive effective treatment measures. This research study has two goals. The first is to monitor user compliance and data quality in mHealth apps with long and repetitive surveys delivered. The second is to identify possible motivational interventions to help improve compliance and data quality. As a form of intervention, will introduce intrinsic and extrinsic motivational factors within the application and test it on a small target population. I will validate the impact of these motivational factors by performing a comparative analysis on the test results to determine improvements in user performance. This study is relevant, as it will help analyze user behavior in long and repetitive self-reporting tasks and derive measures to improve user performance. The results will assist software engineers working with doctors in designing and developing improved self-reporting mHealth applications for collecting better quality data and enhance user compliance.
ContributorsRallabhandi, Pooja (Author) / Gary, Kevin A (Thesis advisor) / Gaffar, Ashraf (Committee member) / Bansal, Srividya (Committee member) / Amresh, Ashish (Committee member) / Arizona State University (Publisher)
Created2017
Description
Tessellation and Screen-Space Ambient Occlusion are algorithms which have been widely-used in real-time rendering in the past decade. They aim to enhance the details of the mesh, cast better shadow effects and improve the quality of the rendered images in real time. WebGL is a web-based graphics library derived from OpenGL ES used for rendering in web applications. It is relatively new and has been rapidly evolving, this has resulted in it supporting a subset of rendering features normally supported by desktop applications. In this thesis, the research is focusing on evaluating Curved PN-Triangles tessellation with Screen Space Ambient Occlusion (SSAO), Horizon-Based Ambient Occlusion (HBAO) and Horizon-Based Ambient Occlusion Plus (HBAO+) in WebGL-based real-time application and comparing its performance to desktop based application and to discuss the capabilities, limitations and bottlenecks of WebGL 1.0.
ContributorsLi, Chenyang (Author) / Amresh, Ashish (Thesis advisor) / Wang, Yalin (Thesis advisor) / Kobayashi, Yoshihiro (Committee member) / Arizona State University (Publisher)
Created2017
Description
Ensemble learning methods like bagging, boosting, adaptive boosting, stacking have traditionally shown promising results in improving the predictive accuracy in classification. These techniques have recently been widely used in various domains and applications owing to the improvements in computational efficiency and distributed computing advances. However, with the advent of wide variety of applications of machine learning techniques to class imbalance problems, further focus is needed to evaluate, improve and optimize other performance measures such as sensitivity (true positive rate) and specificity (true negative rate) in classification. This thesis demonstrates a novel approach to evaluate and optimize the performance measures (specifically sensitivity and specificity) using ensemble learning methods for classification that can be especially useful in class imbalanced datasets. In this thesis, ensemble learning methods (specifically bagging and boosting) are used to optimize the performance measures (sensitivity and specificity) on a UC Irvine (UCI) 130 hospital diabetes dataset to predict if a patient will be readmitted to the hospital based on various feature vectors. From the experiments conducted, it can be empirically concluded that, by using ensemble learning methods, although accuracy does improve to some margin, both sensitivity and specificity are optimized significantly and consistently over different cross validation approaches. The implementation and evaluation has been done on a subset of the large UCI 130 hospital diabetes dataset. The performance measures of ensemble learners are compared to the base machine learning classification algorithms such as Naive Bayes, Logistic Regression, k Nearest Neighbor, Decision Trees and Support Vector Machines.
ContributorsBahl, Neeraj Dharampal (Author) / Bansal, Ajay (Thesis advisor) / Amresh, Ashish (Committee member) / Bansal, Srividya (Committee member) / Arizona State University (Publisher)
Created2017
Description
Machine learning methodologies are widely used in almost all aspects of software engineering. An effective machine learning model requires large amounts of data to achieve high accuracy. The data used for classification is mostly labeled, which is difficult to obtain. The dataset requires both high costs and effort to accurately label the data into different classes. With abundance of data, it becomes necessary that all the data should be labeled for its proper utilization and this work focuses on reducing the labeling effort for large dataset. The thesis presents a comparison of different classifiers performance to test if small set of labeled data can be utilized to build accurate models for high prediction rate. The use of small dataset for classification is then extended to active machine learning methodology where, first a one class classifier will predict the outliers in the data and then the outlier samples are added to a training set for support vector machine classifier for labeling the unlabeled data. The labeling of dataset can be scaled up to avoid manual labeling and building more robust machine learning methodologies.
ContributorsBatra, Salil (Author) / Femiani, John (Thesis advisor) / Amresh, Ashish (Thesis advisor) / Bansal, Ajay (Committee member) / Arizona State University (Publisher)
Created2017
Description
Subdivision surfaces have gained more and more traction since it became the standard surface representation in the movie industry for many years. And Catmull-Clark subdivision scheme is the most popular one for handling polygonal meshes. After its introduction, Catmull-Clark surfaces have been extended to several eminent ways, including the handling of boundaries, infinitely sharp creases, semi-sharp creases, and hierarchically defined detail. For ray tracing of subdivision surfaces, a common way is to construct spatial bounding volume hierarchies on top of input control mesh. However, a high-level refined subdivision surface not only requires a substantial amount of memory storage, but also causes slow and inefficient ray tracing. In this thesis, it presents a new way to improve the efficiency of ray tracing of subdivision surfaces, while the quality is not as good as general methods.
ContributorsKe, Shujian (Author) / Amresh, Ashish (Thesis advisor) / Femiani, John (Committee member) / Gonzalez-Sanchez, Javier (Committee member) / Arizona State University (Publisher)
Created2017
Description
Image processing has changed the way we store, view and share images. One important component of sharing images over the networks is image compression. Lossy image compression techniques compromise the quality of images to reduce their size. To ensure that the distortion of images due to image compression is not highly detectable by humans, the perceived quality of an image needs to be maintained over a certain threshold. Determining this threshold is best done using human subjects, but that is impractical in real-world scenarios. As a solution to this issue, image quality assessment (IQA) algorithms are used to automatically compute a fidelity score of an image.
However, poor performance of IQA algorithms has been observed due to complex statistical computations involved. General Purpose Graphics Processing Unit (GPGPU) programming is one of the solutions proposed to optimize the performance of these algorithms.
This thesis presents a Compute Unified Device Architecture (CUDA) based optimized implementation of full reference IQA algorithm, Visual Signal to Noise Ratio (VSNR) that uses M-level 2D Discrete Wavelet Transform (DWT) with 9/7 biorthogonal filters among other statistical computations. The presented implementation is tested upon four different image quality databases containing images with multiple distortions and sizes ranging from 512 x 512 to 1600 x 1280. The CUDA implementation of VSNR shows a speedup of over 32x for 1600 x 1280 images. It is observed that the speedup scales with the increase in size of images. The results showed that the implementation is fast enough to use VSNR on high definition videos with a frame rate of 60 fps. This work presents the optimizations made due to the use of GPU’s constant memory and reuse of allocated memory on the GPU. Also, it shows the performance improvement using profiler driven GPGPU development in CUDA. The presented implementation can be deployed in production combined with existing applications.
However, poor performance of IQA algorithms has been observed due to complex statistical computations involved. General Purpose Graphics Processing Unit (GPGPU) programming is one of the solutions proposed to optimize the performance of these algorithms.
This thesis presents a Compute Unified Device Architecture (CUDA) based optimized implementation of full reference IQA algorithm, Visual Signal to Noise Ratio (VSNR) that uses M-level 2D Discrete Wavelet Transform (DWT) with 9/7 biorthogonal filters among other statistical computations. The presented implementation is tested upon four different image quality databases containing images with multiple distortions and sizes ranging from 512 x 512 to 1600 x 1280. The CUDA implementation of VSNR shows a speedup of over 32x for 1600 x 1280 images. It is observed that the speedup scales with the increase in size of images. The results showed that the implementation is fast enough to use VSNR on high definition videos with a frame rate of 60 fps. This work presents the optimizations made due to the use of GPU’s constant memory and reuse of allocated memory on the GPU. Also, it shows the performance improvement using profiler driven GPGPU development in CUDA. The presented implementation can be deployed in production combined with existing applications.
ContributorsGupta, Ayush (Author) / Sohoni, Sohum (Thesis advisor) / Amresh, Ashish (Committee member) / Bansal, Ajay (Committee member) / Arizona State University (Publisher)
Created2017
Description
Education of any skill based subject, such as mathematics or language, involves a significant amount of repetition and pratice. According to the National Survey of Student Engagements, students spend on average 17 hours per week reviewing and practicing material previously learned in a classroom, with higher performing students showing a tendency to spend more time practicing. As such, learning software has emerged in the past several decades focusing on providing a wide range of examples, practice problems, and situations for users to exercise their skills. Notably, math students have benefited from software that procedurally generates a virtually infinite number of practice problems and their corresponding solutions. This allows for instantaneous feedback and automatic generation of tests and quizzes. Of course, this is only possible because software is capable of generating and verifying a virtually endless supply of sample problems across a wide range of topics within mathematics. While English learning software has progressed in a similar manner, it faces a series of hurdles distinctly different from those of mathematics. In particular, there is a wide range of exception cases present in English grammar. Some words have unique spellings for their plural forms, some words have identical spelling for plural forms, and some words are conjugated differently for only one particular tense or person-of-speech. These issues combined make the problem of generating grammatically correct sentences complicated. To compound to this problem, the grammar rules in English are vast, and often depend on the context in which they are used. Verb-tense agreement (e.g. "I eat" vs "he eats"), and conjugation of irregular verbs (e.g. swim -> swam) are common examples. This thesis presents an algorithm designed to randomly generate a virtually infinite number of practice problems for students of English as a second language. This approach differs from other generation approaches by generating based on a context set by educators, so that problems can be generated in the context of what students are currently learning. The algorithm is validated through a study in which over 35 000 sentences generated by the algorithm are verified by multiple grammar checking algorithms, and a subset of the sentences are validated against 3 education standards by a subject matter expert in the field. The study found that this approach has a significantly reduced grammar error ratio compared to other generation algorithms, and shows potential where context specification is concerned.
ContributorsMoore, Zachary Christian (Author) / Amresh, Ashish (Thesis director) / Nelson, Brian (Committee member) / Software Engineering (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05