教学课程资源库 更多>>

关键词： Computer engineering   Computer science  

摘要： The research work comprises of two parts. Part one is using an Arbitrary CRCW PRAM algorithm for sorting integers into a linked list. There are various algorithms and techniques to sort the integers in LinkedList. Arbitrary CRCW PRAM model, being the weakest model is able to sort n integers in a LinkedList in “constant time” using nlogm processors and if we use nt processors, then it can be sorted in O(loglogm/logt) time by converting Arbitrary CRCW PRAM model to Priority CRCW PRAM model. Part two is Chaining on a Trie. This research paper solves the problem of chaining on a Trie by providing more efficient complexity. This Algorithm takes “constant time” using n(logm+1) processors to chain the nodes on a Trie for n input integers on the Arbitrary CRCW PRAM model.

关键词： Computer science  

摘要： Higher education has evolved to an open and engaging environment but has generally not invested in endpoint and system security or security training needed to keep its valuable data safe (AACU, 2017). Six trillion dollars in global losses are suspected to be caused by cybersecurity breaches in 2019 (Geronimo, 2019). In 2019 thirty five percent of all security breaches involved higher education institutions (JAMF, 2019). The outlook of endpoint security is grim as higher education has not been a common target for cyberattacks but within the past five years cyberattacks involving higher education have risen dramatically (Bonderud, 2019). If this rise continues and changes are not made to increase security more breaches will occur. New challenges and threats are evolving each day and higher education is being outpaced and breaches are increasing (Athena, 2017). Recovery for many institutions will cost millions of dollars (Geronimo, 2019). Higher education institutions are implementing endpoint security as a reactive action to a breach rather than a proactive action to mitigate or prevent an attack from becoming a breach. Many breaches may have been prevented if higher education institutions had implemented security in a proactive manner rather than reactive to a breach that had already occurred. Keywords: Dr. Chris Riddell, Cybersecurity, Higher Education, Endpoint Security, Cost, Data, Breach, Response, IT, Challenge

关键词： Mechanical engineering   Computer science  

摘要： Kinematic synthesis of mechanisms is an important phase in the conceptual design of machines. This requires the synthesis to be prolific in terms of concept generation to realize the potential of attainable design possibilities and to have the agility to adapt a design to evolving requirements. The goal of this research is to advance the science of kinematic synthesis by bringing together machine learning and theoretical kinematics to create a data-driven computational framework for kinematic synthesis of mechanisms. At the heart of this novel framework lies a probabilistic generative model that learns joint probability distribution of various linkage parameters and their interdependence to perform useful inference tasks. In doing so, we leverage the emerging machine learning techniques to learn meaningful representations and combine them with simultaneous type and dimensional methods of kinematic synthesis to enhance users' computational creativity. The approach is particularly amenable to mechanism synthesis when the input from mechanism designers is deliberately imprecise or inherently uncertain due to the nature of the problem. The approach models the input as a probability distribution and employs a deep generative model to capture the inherent uncertainty in the input. In addition, it gives feedback on the input quality and provides corrections for a more conducive input. Lastly, the outputs are post-processed and the designer is presented with a set of distributions of solutions, where each set consists of a concept with different variations. We define this approach, where the input uncertainty is intelligently managed to generate a distribution of solutions as Variational Synthesis of Mechanisms.

关键词： Information technology   Criminology   Computer science  

摘要： Cyberbullying on social media is a growing problem causing mental and physical harm often leading to anxiety, depression, and suicidal thoughts and tendencies. The purpose of this research was to analyze the issues and dangers of cyberbullying utilizing social media against adolescents. The research outlined the methods that loved ones and social media companies can use to recognize cyberbullying. This study examined the effects of cyberbullying on adolescents, what can be done to protect victims, how they can overcome the mental and physical abuse, and how parents, loved ones, and businesses can prevent further bullying. This research found cyberbullying to be a huge problem. There are several ways to detect and prevent cyberbullying but many of them have limited success. The most effective methods were shown to be automation and forensic investigations. Protections from governments and businesses are coming along slowly putting further burden on families to help and protect victims. This research recommended further development of automation and forensic investigation because they both offer the highest probability of success. Governments at the state and local levels need to create laws and protections more quickly. Businesses need to be more proactive to prevent misuse of their platforms.

关键词： Artificial intelligence   Computer science  

摘要： Image segmentation is a fundamental and challenging problem in computer vision with applications spanning multiple areas, such as medical imaging, remote sensing, and autonomous vehicles. Recently, convolutional neural networks (CNNs) have gained traction in the design of automated segmentation pipelines. Although CNN-based models are adept at learning abstract features from raw image data, their performance is dependent on the availability and size of suitable training datasets. Additionally, these models are often unable to capture the details of object boundaries and generalize poorly to unseen classes. In this thesis, we devise novel methodologies that address these issues and establish robust representation learning frameworks for fully-automatic semantic segmentation in medical imaging and mainstream computer vision. In particular, our contributions include (1) state-of-the-art 2D and 3D image segmentation networks for computer vision and medical image analysis, (2) an end-to-end trainable image segmentation framework that unifies CNNs and active contour models with learnable parameters for fast and robust object delineation, (3) a novel approach for disentangling edge and texture processing in segmentation networks, and (4) a novel few-shot learning model in both supervised settings and semi-supervised settings where synergies between latent and image spaces are leveraged to learn to segment images given limited training data.

关键词： Applied mathematics   Chemistry   Computer science  

摘要： This dissertation applies a skeptical but hopeful analytical paradigm and the tools of linear algebra, numerical methods, and machine learning to a diversity of problems in computational chemistry. When the foundation underlying a project is undermined, the primary purpose of the project becomes digging into the nature and structure of the problem. A common theme emerges in which assumptions in an area are challenged and a deeper understanding of the problem structure leads to new insights. In chapter 2, this approach is exploited to approximate derivative coupling vectors, which together with the difference gradient span the branching planes of conical intersections between electronic states. While gradients are commonly available in many electronic structure methods, the derivative coupling vectors are not always implemented and ready for use in characterizing conical intersections. An approach is introduced which computes the derivative coupling vector with high accuracy (direction and magnitude) using energy and gradient information. The new method is based on the combination of a linear-coupling two-state Hamiltonian and a finite-difference Davidson approach for computing the branching plane. Benchmark cases are provided showing these vectors can be efficiently computed near conical intersections. In chapter 3, this approach yields a countercultural explanation for what machine learning algorithms have learned in modeling a chemical reactivity dataset. Data-driven models of chemical reactions, a departure from conventional chemical approaches, have recently been shown to be statistically successful using machine learning. These models, however, are largely black box in character and have not provided the kind of chemical insights that historically advanced the field of chemistry. The chapter examines the knowledgebase of machine learning models—what does the machine learn?—by deconstructing black box machine learning models of a diverse chemical reaction datase

关键词： Economics   Writing   Computer science   Linguistics   Computational modeling   Mathematical model   Economics   macrostructure   research article  

摘要： Background: The cognitive load involved in research article (RA) reading can be overwhelming for L2 novice readers. RA section headings can be used as signals to help novices focus on essential information related to their learning goals to reduce extraneous cognitive processing. There is a need to examine RA macrostructures to inform RA reading instruction. Literature review: RAs do not always follow the Introduction-Methods-Results-Discussion (IMRD) model. Previous research has examined the macrostructure of articles in disciplines such as computer science, applied linguistics, and pure mathematics, but few have investigated the macrostructure of economics RAs. Research questions: 1. Are there any sections frequently used in economics articles apart from the conventional sections? 2. If yes, what are the views of expert economics RA readers on the communicative functions and propositional content of the newly identified sections of economics RAs? Research methods: Eighty RAs were collected from five economics journals using stratified random sampling. Following Yang and Allison's macrostructure analysis method, we conducted an analysis of the overall structure of the RAs based on section headings and the function and content of each section. Results: Compared with the IMRD model, we found six new section types: Background, Theoretical Model, Econometric Model, Robustness, Mechanisms, and Application. Interviews were conducted to explore expert RA readers' genre knowledge on the newly identified sections. Conclusion: The findings can be useful for RA reading and writing instruction and future research on part-genres of economics articles.

关键词： Mathematics   Artificial intelligence   Computer science  

摘要： My dissertation presents several new algorithms incorporating non-parametric and deep learning approaches for computer vision and related tasks, including object localization, object tracking and model compression. With respect to object localization, I introduce a method to perform active localization by modeling spatial and other relationships between objects in a coherent “visual situation” using a set of probability distributions. I further refine this approach with the Multipole Density Estimation with Importance Clustering (MIC-Situate) algorithm. Next, I formulate active, “situation” object search as a Bayesian optimization problem using Gaussian Processes. Using my Gaussian Process Context-Situation Learning (GP-CL) algorithm, I demonstrate improved efficiency for object localization over baseline procedures. In subsequent work, I expand this research to frame object tracking in video as a temporally-evolving, dynamic Bayesian optimization problem. Here I present the Siamese-Dynamic Bayesian Tracking Algorithm (SDBTA), the first integrated dynamic Bayesian optimization framework in combination with deep learning for video tracking. Through experiments, I show improved results for video tracking in comparison with baseline approaches. Finally, I propose a novel data compression algorithm, Regularized L21 Semi-NonNegative Matrix Factorization (L21 SNF) which serves as a general purpose, parts-based compression algorithm, applicable to deep model compression.

关键词： assessment   computer science   critical assessment   education   results   avaliação   avaliação crítica   ciência da computação   educação   resultados   информатика   критическое оценивание   образование   оценивание   результаты   Language of Keywords: English   Portuguese   Russian  

摘要： Durante muito tempo, a principal abordagem da avaliação foi a abordagem normativa, quando as realizações individuais dos estudantes foram comparadas com uma determinada norma (os resultados da maioria dos estudantes). Recentemente, a pesquisa pedagógica doméstica tem desenvolvido uma abordagem baseada em critérios para avaliar o desempenho acadêmico, de acordo com que as realizações dos estudantes são comparadas com o escopo de conhecimento que deve ser adquirido em um estágio específico do treinamento. Este estudo teve como objetivo determinar o papel da avaliação baseada em critérios na formação da alfabetização funcional dos estudantes em ciência da computação e construir um modelo de avaliação baseada em critérios no desenvolvimento da alfabetização funcional além de demonstrar a eficácia dos métodos de formação dos alunos. 'alfabetização funcional em ciência da computação. Os principais métodos de pesquisa foram a metodologia de avaliação baseada em critérios desenvolvida pelos autores e a metodologia de avaliação formativa. Foram analisados alguns elementos da metodologia de avaliação formativa. Para o desenvolvimento adicional da metodologia para a formação da alfabetização funcional dos estudantes em ciência da computação, foi construído um modelo de avaliação baseada em critérios. A introdução da avaliação baseada em critérios permitirá mudar para uma avaliação formativa destinada ao desenvolvimento da competência do estudante. A avaliação, composta por critérios que o estudante entende, o estimula e torna o processo de aprendizado lógico. Com base nos experimentos práticos e na avaliação proposta com base em critérios, foi comprovada a eficácia de métodos para a formação da alfabetização funcional dos estudantes em ciência da computação. [ABSTRACT FROM AUTHOR]

关键词： Electrical engineering   Computer science  

摘要： Next generation wireless systems are widely expected to connect a massive number of low-energy wireless devices that form the fabric of smart technologies and cyberphysical systems. The resultant "Internet of Things" is expected to be one of the major drivers of technological growth over the next decade. However, its design presents new challenges to wireless engineers. One of the key challenges is how to enable a massive number of low-energy wireless devices to sporadically access the scarce spectrum with minimal coordination and channel estimation overheads. In this dissertation, we focus on this fundamental question. We describe a mathematical framework to study this problem and show how this framework relates to a seemingly unrelated problem in theoretical computer science called group testing. Inspired by this connection, we develop novel group testing codes that are optimal under constraints on the number of tests each item can participate in, or the number of items each test can include. We show that these codes can be translated to innovative energy-efficient random-access protocols which embrace sensor collisions and thus eliminate the complexity and energy cost of collision resolution and retransmissions. We then extend our results in a number of different directions. We incorporate the channel noise in our model and develop both combinatorial and statistical group testing codes that are resilient to errors. We further consider the asynchronous case by removing the assumption that the users' transmissions are aligned, which may not necessarily hold in practical applications. We let the users' transmissions within a frame be received with arbitrary and unknown delays at the receiver. We show that such delays can be accommodated by designing cyclic group testing codes without introducing any additional penalty in transmission length with respect to the synchronous case. We further contribute to the group testing literature by introducing an order-optimal exp