教学课程资源库 更多>>

关键词： Computer science   Elementary education   Expansive framing   In-service teacher learning   Teacher professional development  

摘要： Elementary school teachers are being increasingly asked to teach computer science—something that most teacher certification programs do not prepare them for. In an attempt to study how elementary teachers learn to teach computer science, I analyzed the ways that teachers behaved during a professional development accompanying the implementation of a fifth-grade computer science curriculum. My findings suggest that teachers benefit from professional development that encourages collaboration and active participation in teachers through discussion and modeling. Furthermore, my findings suggest that teachers benefit from using curriculum that deliberately connects new concepts to content that they are already familiar and comfortable with—a model known as expansive framing. By encouraging active teacher participation in professional development and by using curriculum that relates to teachers’ existing content knowledge, we may be able to help elementary teachers prepare to teach computer science with more confidence and accuracy.

关键词： Predictive models   Correlation   Urban areas   Neural networks   Trajectory   Task analysis   Computer science   Deep learning   urban crowd flows   neural networks   long short-term memory   convolutional neural network  

摘要： Prediction of traffic crowd movement is one of the most important component in many applications ' domains ranging from urban management to transportation schedule. The key challenge of citywide crowd flows prediction is how to model spatial and dynamic temporal correlation. However, in recent years several studies have been done, but they lack the ability to effectively and simultaneously model spatial and temporal dependencies among traffic crowd flows. To address this issue, in this article a novel spatio-temporal deep hybrid neural network proposed termed STD-Net to forecast citywide crowd traffic flows. More specifically, STD-Net contains four major branches, i.e., closeness, period volume, weekly volume, and external branches, respectively. We design a residual neural network unit for each property to depict the spatio-temporal features of traffic flows. For various branches, STD-Net provides distinct weights and then combines the outputs of four branches together. Extensive experiments on two large-scale datasets from New York bike and Beijing taxi have demonstrated that STD-Net achieves competitive performances the existing state-of-the-art prediction baselines.

关键词： Servers   Browsers   Web pages   Time-frequency analysis   Prefetching   Prediction algorithms   Computer science   Locally normalized intervals   proxy   web cache   pre-fetching  

摘要： Caching is one of the most promising areas in the field of future internet architecture like Information-centric Networking, Software Defined Networking, and IoT. In Web caching, most of the web content is readily available across the network, even if the webserver is not reachable. Several existing traditional caching methods and cache replacement strategies are evaluated based on the metrics like hit ratio and byte hit Ratio. However, these metrics have not been improved over the period because of the traditional caching policies. So, in this paper, we have used an intelligent function like locally normalized intervals of page visit, website duration, users' interest between user groups is proposed. These intervals are combined with multiple distance metrics like Manhattan, squared Euclidean, and 3-,4-,5-norm Minkowski. In order to obtain significant common user navigation patterns, the clustering relation between the users using different intervals and distances is thoroughly analyzed. These patterns are successfully coupled with greedy web cache replacement strategies to improve the efficiency of the proposed web cache system. Particularly for improving the caching metrics more, we used an AI-based intelligent approach like Random Forest classifier to boost the prefetch buffer performance and achieves the maximum hit rate of 0.89, 0.90, and byte hit rate of 0.87, 0.89 for Greedy Dual Size Frequency and Weighted Greedy Dual Size Frequency algorithms, respectively. Our experiments show good hit/byte hit rates than the frequently used algorithms like least recently used and least frequently used.

关键词： Streaming media   Urban areas   Image segmentation   Annotations   Semantics   Computer science   Automobiles   Autonomous driving   night-time vision   scene analysis   adverse conditions  

摘要： Although huge progress has been made on scene analysis in recent years, most existing works assume the input images to be in day-time with good lighting conditions. In this work, we aim to address the night-time scene parsing (NTSP) problem, which has two main challenges: 1) labeled night-time data are scarce, and 2) over- and under-exposures may co-occur in the input night-time images and are not explicitly modeled in existing pipelines. To tackle the scarcity of night-time data, we collect a novel labeled dataset, named NightCity, of 4,297 real night-time images with ground truth pixel-level semantic annotations. To our knowledge, NightCity is the largest dataset for NTSP. In addition, we also propose an exposure-aware framework to address the NTSP problem through augmenting the segmentation process with explicitly learned exposure features. Extensive experiments show that training on NightCity can significantly improve NTSP performances and that our exposure-aware model outperforms the state-of-the-art methods, yielding top performances on our dataset as well as existing datasets.

关键词： Computer science   Artificial intelligence   Neural networks (Computer science)   Machine learning  

摘要： Artificial intelligence powered by deep neural networks has seen tremendous improvements in the last decade, achieving superhuman performance on a diverse range of tasks. Many worry that it can one day develop the ability to recursively self-improve itself, leading to an intelligence explosion known as the Singularity. Autogenerative networks, or neural networks generating neural networks, is one major plausible pathway towards realizing this possibility. The object of this thesis is to study various challenges and applications of small-scale autogenerative networks in domains such as artificial life, reinforcement learning, neural network initialization and optimization, gradient-based meta-learning, and logical networks. Chapters 2 and 3 describe novel mechanisms for generating neural network weights and embeddings. Chapters 4 and 5 identify problems and propose solutions to fix optimization difficulties in differentiable mechanisms of neural network generation known as Hypernetworks. Chapters 6 and 7 study implicit models of network generation like backpropagating through gradient descent itself and integrating discrete solvers into continuous functions. Together, the chapters in this thesis contribute novel proposals for non-differentiable neural network generation mechanisms, significant improvements to existing differentiable network generation mechanisms, and an assimilation of different learning paradigms in autogenerative networks.

关键词： Computer science  

摘要： From the early days of the mid to late nineteenth century when scientific research first began to focus on fingerprints, to the present day fingerprint recognition systems we find deployed on our day to day devices, the science of fingerprint recognition has come a long way. In spite of this progress, there remains challenging problems to be solved. This thesis highlights a few of these problems, and proposes solutions to address them. One area of further research that must be conducted on fingerprint recognition systems is that of robust, operational evaluations. In chapter two of this thesis, we show how the current practices of using calibration patterns to evaluate fingerprint readers are limited. We then propose a realistic fake finger called the Universal Target. The Universal Target is a realistic, 3D, fake finger (or phantom) which can be imaged by all major types of fingerprint sensing technologies. We show the entire manufacturing (molding and casting) process for fabricating the Universal Targets. Then, we show a series of evaluations which demonstrate how the Universal Targets can be used to operationally evaluate current commercial fingerprint readers. Our Universal Target is a significant step forward in enabling more realistic, standardized evaluations of fingerprint readers. In our third chapter, we shift gears from improving the evaluation standards of fingerprint readers to instead focus on the security of fingerprint readers. In particular, we turn our attention towards detecting fake fingerprint (spoof) attacks. To do so, we open source a fingerprint reader (built from low-cost ubiquitous components), called RaspiReader. RaspiReader is a high-resolution fingerprint reader customized with both direct-view imaging and FTIR imaging in order to better detect fingerprint spoofs. We show through a number of experiments that RaspiReader enables state-of-the-art fingerprint spoof detection accuracy. We also demonstrate that RaspiReader enables better gen

关键词： Approximation algorithms   Artificial intelligence   Computer science   Dragon boat festivals   Machine learning   Neural networks  

摘要： We investigate approximating the Dragon Boat Partition problem, a practical real-world variant of the Partition problem. A team of dragon boat participants must be partitioned with an approximately balanced arrangement with a preferable weight difference of 0. We present two approaches that capture the participant characteristics. The first approach takes a heuristic route. The second approach applies Deep Convolutional Neural Networks to the problem, with two versions. In our 10,000 episodes per experiment, our heuristic implementation had an average episode runtime of 1.84ms, an average of 7.39 steps per episode,perfect left-right approximation rate of 98.53%, perfect front-back approximation rate of89.16%, and a perfect combined approximation rate of 90.15%. Whereas our best deep learning model has an average episode runtime of 1.23ms, an average of 4.65 steps perepisode, perfect left-right approximate rate of 98.00%, perfect front-back approximation rate of 95.13%, and a perfect combined approximation rate of 94.28%.

关键词： Ontologies   Text categorization   Semantic Web   Semantics   Computer science   Tools   Libraries   Clustering   recommendation system   semantic web   ontology   text categorization   text mining   unsupervised learning  

摘要： Semantic Web is considered as the backbone of web 3.0 and ontologies are an integral part of the Semantic Web. Though an increase of ontologies in different domains is reported due to various benefits which include data heterogeneity, automated information analysis, and reusability, however, finding an appropriate ontology according to user requirement remains cumbersome task due to time and efforts required, context-awareness, and computational complexity. To overcome these issues, an ontology recommendation framework is proposed. The Proposed framework employs text categorization and unsupervised learning techniques. The benefits of the proposed framework are twofold: 1) ontology organization according to the opinion of domain experts and 2) ontology recommendation with respect to user requirement. Moreover, an evaluation model is also proposed to assess the effectiveness of the proposed framework in terms of ontologies organization and recommendation. The main consequences of the proposed framework are 1) ontologies of a corpus can be organized effectively, 2) no effort and time are required to select an appropriate ontology, 3) computational complexity is only limited to the use of unsupervised learning techniques, and 4) due to no requirement of context awareness, the proposed framework can be effective for any corpus or online libraries of ontologies.

关键词： Databases   Bars   Heuristic algorithms   Computer science   Syntactics   Optimization   Licenses   Dynamic database   incomplete database   pairwise comparison   skyline queries  

摘要： Skyline queries have been widely used as an effective query tool in many contemporary database applications. The main concept of skyline queries relies on retrieving the non-dominated tuples in the database which are known skylines. In most database applications, the contents of the databases are dynamic due to the continuous changes made towards the database. Typically, the changes in the contents of the database occur through data manipulation operations (INSERT and/or UPDATE). Performing these operations on the database results in invalidating the most recent skylines before changes are made on the database. Furthermore, the presence of incomplete data in databases becomes frequent phenomena in recent database applications. Data incompleteness causes several challenges on skyline queries such as losing the transitivity property of the skyline technique and the test dominance process between tuples being cyclic. Reapplying skyline technique on the entire updated incomplete database to determine the new skylines is unwise due to the exhaustive pairwise comparisons. Thus, this paper proposes an approach, named Incomplete Dynamic Skyline Algorithm (IDSA) which attempts to determine the skylines on dynamic and incomplete databases. Two optimization techniques have been incorporated in IDSA, namely: pruning and selecting superior local skylines. The pruning process attempts to exploit the derived skylines before the INSERT/UPDATE operation made on the database to identify the new skylines. Moreover, selecting superior local skylines process assists in further eliminating the remaining non-skylines from further processing. These two optimization techniques lead to a large reduction in the number of domination tests due to avoiding re-computing of skylines over the entire updated database to derive the new skylines. Extensive experiments have been accomplished on both real and synthetic datasets, and the results demonstrate that IDSA outperforms the existing solutions in

关键词： African Americans   Computer science   Machine learning   Osteoporosis   SNPs  

摘要： Osteoporosis is a debilitating disease in which an individual’s bones weaken, making bones fragile and more susceptible to fracture. While commonly found amongst postmenopausal Caucasian and Asian women based on previous studies, those of African descent (African American/Black) have largely been ignored when it comes to osteoporotic studies, especially when it comes to Genome Wide Association Studies (GWAS). From GWA studies, we gain access to single nucleotide poly-morphisms (SNPs) that may contribute to certain illnesses, such as osteoporosis. With low Bone Mineral Density (BMD) being one of the primary markers of potential osteoporosis, it is prudent that proper research is done in order to help the African American population circumvent or mitigate the worst symptoms and complications of osteoporosis. In this thesis, we implemented and applied machine learning algorithms to analyze genetic data of African American women in order to make predictions that map SNPs to BMD. Using Coefficient of Determination (R2) and Mean Squared Error (MSE) for evaluation, the machine learning techniques we utilized for this regression task are: regularized linear regression (Ridge, Lasso, and ElasticNet), gradient boosted trees (XGBoost and LightGBM), and artificial neural networks. With these models, we performed an analysis on 3 datasets, compromised of 12,600, 69,476, and 158,444 variants respectively. The first dataset, known as SNP-1, received its highest overall test R2 score of 0.227 through Lasso Regression. The second dataset, SNP-5, received its highest overall test R2 score of 0.437 through LightGBM. And lastly, the third dataset, SNP-10, received its highest overall test R2 score of 0.574 from Ridge Regression.