教学课程资源库 更多>>

关键词： Medicine   Bioinformatics   Computer science  

摘要： Life's most valuable asset is health. Continuously understanding the state of our health and modeling how it evolves is essential if we wish to improve it. Given the opportunity that people live with more data about their life today than any other time in history, the challenge rests in interweaving this data with the growing body of knowledge to compute and model the health state of an individual continually. This dissertation presents an approach to build a personal model and dynamically estimate the health state of an individual by fusing multi-modal data and domain knowledge. The system is stitched together from four essential abstraction elements: 1. the events in our life, 2. the layers of our biological systems (from molecular to an organism), 3. the functional utilities that arise from biological underpinnings, and 4. how we interact with these utilities in the reality of daily life. Connecting these four elements via graph network blocks forms the backbone by which we instantiate a digital twin of an individual. Edges and nodes in this graph structure are then regularly updated with learning techniques as data is continuously digested. Experiments demonstrate the use of dense and heterogeneous real-world data from a variety of personal and environmental sensors to monitor individual cardiovascular health state. State estimation and individual modeling is the fundamental basis for many revolutionary efforts in health. Continuous access to the individual's state allows a departure from disease-oriented approaches, to a total health continuum paradigm. Precision in predicting health requires a detailed understanding of an individual and their trajectory. By encasing this state estimation within a navigational approach, a user can use a systematic guidance framework to plan actions for each moment to transition their current state towards a desired one. This work concludes by presenting this framework of combining the health state and personal graph model to pe

关键词： English literature   Computer science  

摘要： Nick Montfort’s book of computational poetry, The Truelist, allows the reader to understand the code written by Montfort to produce the book as both “bricoleur” and “bricolage.” At the core of my examination, I will ask: how does computational poetry, specifically that of The Truelist, provide us with a new definition of what it means to embody a coded space, specifically a space that, according to Montfort, is constantly redefining itself for new readers? To focus specifically on examining the literariness of the text, I will primarily use Jacques Derrida’s concept of the “bricolage” from “Structure, Sign, and Play in the Discourse of the Human Sciences” (specifically focusing on the output of the code) in an attempt to understand how Montfort’s poetry-producing computations redefine poetry’s embodiment of space and place, and how the code itself becomes a bricoleur, particularly with regard to associative imagery and how this computational “communication” forbids the reader to engage passively with the text and its potential variations.

关键词： Computer Science  

摘要： Large scale data is often stored in distributed storage systems. Until not too long ago, data scientists transferred all the data from data servers to a single server to be analyzed—known as monolithic data analysis. However, it is not always justified to collect all data into one machine when the data is too large to fit on one machine, the network is constrained, or the control domain of the data is limited to multiple private networks. In such situations, it is more plausible to push the executions of the learning algorithm to the data servers that hold the data. Motivated by this, this thesis aims in analyzing the feasibility and the performance of distributed machine learning experiments where a coordinator program delegates learning tasks to multiple data servers. In this context, individual data servers can each learn their own models or contribute to the learning of a global model using the subset of the data locally stored and the coordinator program can merge the results. One of the challenges of this setup is that data servers usually carry a prearranged partition of the data which could potentially be skewed in terms of data distribution. This might pose problems in the learning process as each server could work on data that is skewed. Therefore, we explore the effects of the data skewness on the accuracy of two sets of learning algorithms, i.e., subagging and iterative methods. In subagging models, each data server builds its own base model and the coordinator votes new queries against all data servers to make predictions. In iterative methods, there is a global model that each data server improves by running iterative learning algorithms against their local datasets. We further propose that servers can share a fraction of their data to balance each other’s distribution.

关键词： Computer science  

摘要： ‘Open up’ and learn more about what it is like to live in a world of security-minded programmers. The problem is insecure software and a lack of secure software application programmers. This study uncovers the etiology of application programmers who take the time and effort to ensure the code is secure. Fourteen informants interviewed provided their lived experiences as secure software programmers. The data collected was thematized using a qualitative hermeneutic phenomenological methodology. The findings show that training, leadership support, and suitability are reified themes. A training platform for programmers to learn how to write secure programming structures in the corporation’s software language is a solution.

关键词： Agriculture engineering   Developmental psychology   Computer science  

摘要： Tassel is the male inflorescences organ of the maize plant that develops atop the plant. Coarse features of tassels, including shape and size, can influence shedding pollen, fertilization, and subsequently grain yield. Therefore, understanding tassel dynamics and characteristics as well as how it evolves during the plant growth can help the plant scientist community to increase the grain yield as a final goal. To do so, first, tassels were investigated in one time points. The tassels were cut in the field and their images were captured in a lightbox. Coarse features were measured using novel image processing approaches. 351 tassels with different genotypes were used for the experiment. Tassel length, first lowest branch length, and angle as well as central spike length were measured by applying image processing and machine learning techniques. Tassels were also classified to open and close structures to obtain accurate predictions for the traits. The results show that R2 values for the tassel length and central spike length were 0.92 and 0.80, respectively. In addition, the R2 values for the first lowest branch length and angle were 0.63 and 0.91, respectively. The R2 values for the first lowest branch length was low compared to others because locating the first lowest branch point and its corresponding branch tip was hard due to branches occlusion. This study was done to create a robust algorithm for tassel phenotyping. Challenges were figured out for better tassel phenotyping in the field. Then, we looked at a diverse panel in the field, using stationary cameras to capture 6 tassels every 10 minutes for 8 hours per day during a month. Traditional approaches for phenotyping anthesis progression are time-consuming, subjective, and labor-intensive and are thus impractical for phenotyping large populations in multiple environments. In this work, we utilize a high throughput phenotyping approach that is based on extracting time-lapse information of anthesis progress fr

关键词： Computer science  

摘要： Mobile Computing is a technology that allows transmission of audio, video, and other types of data via a computer or any other wireless-enabled device without having to be connected to a fixed physical link. Despite increasing usage of mobile computing, exploiting its full potential is difficult due to its inherent problems such as resource scarcity, connection instability, and limited computational power. In particular, the advent of connecting mobile devices to the internet offers the possibility of offloading computation and data intensive tasks from mobile devices to remote cloud servers for efficient execution. This proposed thesis develops an algorithm that uses an objective function to adaptively decide strategies for computational offloading according to changing context information. By following the style of Service-Oriented Architecture (SOA), the proposed framework brings cloud computing to mobile devices for mobile applications to benefit from remote execution of tasks in the cloud. This research discusses the algorithm and framework, along with the results of the experiments with a newly developed system for self-driving vehicles and points out the anticipated advantages of Adaptive Computational Offloading.

关键词： Communication   Social structure   Computer science  

摘要： Information Cascades are propagation patterns in human networks where information such as political propaganda, rumors, or fashion trends, emanate from one or more starting points and are either accepted or rejected by each connected member of a community. The decisions made by individuals either govern a large ripple through the community, or prevent significant changes from occurring, making Information Cascades of avid interest within economics and network science. Since Information Cascades have similar behavior to the well studied field of epidemiology, many Cascades are modeled with disease propagation models like the Susceptibility, Infection, and Recovery (SIR) model. In this thesis we argue that such generic models are insufficient for predicting information spread in human behavioral networks, because they do not represent temporal availability of communication and individual variations in susceptibility to peer-pressure. We propose a new model wherein agents have heterogeneous activity periods and activation thresholds, representing individual susceptibility to peer-pressure. By applying this model to networks with a range of topologies, as well as real-world networks, we are able to more accurately represent human behavior in a variety of communities.

关键词： Clothing   Gallium nitride   Generative adversarial networks   Generators   Task analysis   Visualization   Computer science   Clothes collocation   clothing attributes   fashion data   generative adversarial network (GAN)   image translation  

摘要： The choice of which clothes to wear affects how one is perceived, as well as constitutes an expression of one's personal style. Based on the recent advances in image-to-image translation by the conditional generative adversarial network (cGAN), we propose a new framework with a multidiscriminator by incorporating different types of conditional information into the discriminator of cGAN for clothing matches. In contrast with most extant frameworks under cGAN, with one generator and one discriminator, the proposed framework investigates the potential of utilizing conditional information delivered by multidiscriminators to guide the generator. Under this framework, we propose an Attribute-GAN with two discriminators and a category-attribute GAN (CA-GAN) with three discriminators. In order to evaluate the performance of our proposed models, we built a large-scale data set that consists of 19 081 pairs of collocation clothing images with 90 manually labeled attributes. Experimental results demonstrate that with supervision of the additional attribute discriminator or category discriminator, the quality of the generated clothing images by GANs is consistently improved in comparison with the state-of-the-art methods.

关键词： Indexes   Complexity theory   Learning systems   Training   Benchmark testing   Technological innovation   Computer science   Bayes classifier   class imbalance learning   data complexity   imbalance measure   imbalance recovery methods  

摘要： Recent studies of imbalanced data classification have shown that the imbalance ratio (IR) is not the only cause of performance loss in a classifier, as other data factors, such as small disjuncts, noise, and overlapping, can also make the problem difficult. The relationship between the IR and other data factors has been demonstrated, but to the best of our knowledge, there is no measurement of the extent to which class imbalance influences the classification performance of imbalanced data. In addition, it is also unknown which data factor serves as the main barrier for classification in a data set. In this article, we focus on the Bayes optimal classifier and examine the influence of class imbalance from a theoretical perspective. We propose an instance measure called the Individual Bayes Imbalance Impact Index (IBI3) and a data measure called the Bayes Imbalance Impact Index (BI3). IBI3 and BI3 reflect the extent of influence using only the imbalance factor, in terms of each minority class sample and the whole data set, respectively. Therefore, IBI3 can be used as an instance complexity measure of imbalance and BI3 as a criterion to demonstrate the degree to which imbalance deteriorates the classification of a data set. We can, therefore, use BI3 to access whether it is worth using imbalance recovery methods, such as sampling or cost-sensitive methods, to recover the performance loss of a classifier. The experiments show that IBI3 is highly consistent with the increase of the prediction score obtained by the imbalance recovery methods and that BI3 is highly consistent with the improvement in the F1 score obtained by the imbalance recovery methods on both synthetic and real benchmark data sets.

关键词： Social research   Computer science  

摘要： Critical public data in the United States are vulnerable to theft, creating severe financial and legal implications for payment-card acceptors. When security analysts and managers who work for payment card processing organizations implement strategies to reduce or eliminate payment-card fraud, they protect their organizations, consumers, and the local and national economy. Grounded in Cressey’s fraud theory, the purpose of this qualitative single case study was to explore strategies business owners and card processors use to reduce or eliminate payment-card fraud. The participants were 3 data security analysts and 1 manager working for an international payment card processing organization with 10 years or more experience working with payment card fraud detection in the southeastern United States. The data collection process was face-to-face semistructured interviews and review of company documentation. Within-case analysis, pattern matching, and methodological triangulation were used to identify 4 themes. The key themes related to artificial intelligence, cardholder and acceptor education, enhanced security strategies, and Payment Card Industry Data Security Standard (PCI-DSS) rules and regulations to reduce or end card fraud. The key recommendations are enforcement of stricter PCI-DSS rules and regulations for accepting payment cards at the acceptor and processor levels to reduce the potential for fraud through the use of holograms and card reader clearance between customers. The implications for social change include the potential to reduce costs to consumers, reduce overhead costs for businesses, and provide price reductions for consumers. Additionally, consumers may gain a sense of security when using their payment-card for purchases.