• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 505
  • 208
  • 197
  • 163
  • 27
  • Tagged with
  • 1181
  • 774
  • 700
  • 437
  • 437
  • 401
  • 401
  • 398
  • 398
  • 116
  • 115
  • 103
  • 88
  • 86
  • 81
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
251

Approximate inference of Bayesian networks through edge deletion

Thornton, Julie Ann January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / William Hsu / Bayesian networks are graphical models whose nodes represent random variables and whose edges represent conditional dependence between variables. Each node in a Bayesian network is equipped with a conditional probability function that expresses the likelihood that the node will take on different values given the values of its parents. A common task for a Bayesian network is to perform inference by computing the marginal probabilities of each possible value for each node. In this thesis, I introduce three new algorithms for approximate inference of Bayesian networks that use edge deletion techniques. The first reduces a network to its maximal weight spanning tree using the Kullback-Leibler information divergence as edge weights, and then runs Pearl’s algorithm on the resulting tree. Because Pearl’s algorithm can perform inference on a tree in linear time, as opposed to the exponential running time of all general exact inference algorithms, this reduction results in a tremendous speedup in inference. The second algorithm applies triangulation pre-processing rules that are guaranteed to be optimal if the original graph has a treewidth of four or less, and then deletes edges from the network and continues applying rules so that the resulting triangulated graph will have a maximum clique size of no more than five. The junction tree exact inference algorithm can then be run on the reduced triangulated graph. While the junction tree algorithm has an exponential worst-case running time in the size of the maximum clique in the triangulated graph, placing a bound on the clique size effectively places a polynomial time bound on the inference procedure. The third algorithm deletes edges from a triangulation of the original network until the maximum clique size in the triangulated graph is below a desired bound. Again, the junction tree algorithm can then be run on the resulting triangulated graph, and the bound on the maximum clique size will also polynomially bound the inference time. When tested for efficiency and accuracy on common Bayesian networks, these three algorithms perform up to 10,000 times faster than current exact and approximate techniques while achieving error values close to those of sampling techniques.
252

Performance evaluation of J2EE & .NET web services interacting with a .NET client

Nakka, Raja Sanjeev Kumar January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / Daniel A. Andresen / The objective of this project is to build an e-commerce website using ASP.NET as well as J2EE technologies. Visual Studio 2005 provides a great IDE to build seamless front end layer for the websites with minimum effort. Hence, it is used to build the presentation layer. The business logic layer is developed in ASP.NET 2.0 and J2EE, exposing the functions as web services. Two similar clients have been developed in ASP.NET. These two clients invoke the .NET and J2EE web services. The two similar websites are subjected to testing for correctness and performance. Stress and load testing is performed thoroughly on the website. A detailed analysis of the results is done using the response times and throughputs from various tests. This report proves that we can build a website using a presentation tier developed in ASP.NET consuming any of the two technologies’ (ASP.NET & J2EE) web services. This proves to be very useful in scenarios where we need the best overall user experience regardless of the technology, to replace or extend an existing business tier by using a different technology.
253

Classification of emotion using sub audible frequencies in vocal data

Narber, Cody G. January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / David A. Gustafson / Current research involving vocal emotion detection has taken a variety of different approaches, and has found certain acoustic attributes to characterize different emotional states. While there have been improvements in classification over the past few years, computer classification is not nearly as accurate as human classification. This paper proposes the existence of an attribute that has not been examined, which can be used as a measure for detecting emotion in human vocal samples. It is shown that the new infrasonic attribute is significant when examining agitated emotions. Therefore, it can be used to help improve vocal emotion detection.
254

A service to automate the task assignment process in YAWL

Samanthula, Krishna Nagarjun Reddy January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / Gurdip Singh / Developing an optimal working environment and managing the of work load in an efficient manner are the major challenges for most businesses today. So, the importance of the workflow and workflow management in an organization is unquestionable. Many organizations use sophisticated systems to organize the workflows. One such workflow system based on a concise and powerful modeling language called “Yet Another Workflow Language” is YAWL. YAWL handles complex data, transformations, integration with organizational resources and Web Service integration. Workflow comprises of three main perspectives: control-flow, data and the resources. In Yawl, the control-flow and the data-flow are tightly coupled within the workflow enactment engine. But the resource perspective is provided by a discrete custom service called Resource Service. Administrative tools are provided using which the administrator has to manually select the resource (referred as participant) which needs to perform a particular task of the workflow. This project aims at developing a service which can automate the assignment of the tasks to the participants by using the Resource service which provides number of interfaces that expose the full functionality of the service. The application of this project with respect to Healthcare domain is presented. Healthcare domain is the one of the most demanding and yet critical business process. Hospitals face increasing pressure to both improve the quality of the services delivered to patients and to reduce costs .Hence there is significant demand on hospitals in regard to how the organization, execution, and monitoring of work processes is performed. Workflow Management Systems like YAWL offers a potential solution as they support processes by managing the flow of work.
255

Adhoc routing based data collection application in wireless sensor networks

Pinjala, Mallikarjuna Rao January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / Gurdip Singh / Ad hoc based routing protocol is a reactive protocol to route messages between mobile nodes. It allows nodes to pass messages through their neighbors to nodes which they cannot directly communicate. It uses Route Request (RREQ) and Route Reply (RREP) messages for communication. Wireless sensor networks consist of tiny sensor motes with capabilities of sensing, computation and wireless communication. This project aims to implement data collector application to collect the temperature data from the set of wireless sensor devices located within a building, which will help in gathering the information by finding the route with minimum number of hops to reach destination and generates low message traffic by not encouraging the duplicate message within the network. Using this application, wireless devices can communicate effectively to provide the network information to the user. This system consists of a mobile wireless sensor device called base station which is connected to a PC to communicate and is the root of the network. It also consists of set of client sensor devices which are present in different parts of the building. This project has been evaluated by determining how well the ad hoc protocol performs by measuring the number of messages and time consumed in learning about the complete topology. This application will eventually find the path with minimum number of hops. Simple Network Management Protocol (SNMP) is also used to monitor the sensor nodes remotely. This project was developed using nesC and C programming languages with TinyOS and UNIX based operating systems. It has been tested with a sufficient number of motes and evaluated based on the number of messages generated and number of hops traveled for each route request.
256

A development environment and static analyses for GUARDOL - a language for the specification of high assurance guards

Dodds, Josiah January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / John M. Hatcliff / There are a number of network situations where different networks have different security policies and still need to share information. While it is important to allow some data to flow between the two networks, it is just as important that they don't share any data that violates the respective security policies of the networks. Constraints on data sharing are often phrased in terms of classification levels of data (e.g. top secret, secret, public). They might also be stated in terms of the contents of the data (e.g. are there military base names, is the location correct). The software and hardware that works to solve these problems is called Cross Domain Solutions (CDS). There are a variety of hardware platforms capable of implementing CDS. These platforms are all configured in different ways and they are often proprietary. Not only are there a number of platforms on the market, many are difficult to understand, verify, or even specify. The Guardol project provides an open, non-proprietary, and domain-specific language for specifying CDS security policies and implementing CDS. Guardol is designed to be easy to understand and verify. This thesis describes the design and implementation of primary Guardol components. It includes a description of the Eclipse GUI plug-ins that have been developed for the project as well as a description of new formal analyses and translations that have been developed for the language. The translation is used to plug into external tools for model checking and the analyses help to make the translation clean and efficient. The analyses are also useful tools to help make the use of Guardol easier for developers.
257

Tag recommendation using Latent Dirichlet Allocation.

Choubey, Rahul January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / Doina Caragea / The vast amount of data present on the internet calls for ways to label and organize this data according to specific categories, in order to facilitate search and browsing activities. This can be easily accomplished by making use of folksonomies and user provided tags. However, it can be difficult for users to provide meaningful tags. Tag recommendation systems can guide the users towards informative tags for online resources such as websites, pictures, etc. The aim of this thesis is to build a system for recommending tags to URLs available through a bookmark sharing service, called BibSonomy. We assume that the URLs for which we recommend tags do not have any prior tags assigned to them. Two approaches are proposed to address the tagging problem, both of them based on Latent Dirichlet Allocation (LDA) Blei et al. [2003]. LDA is a generative and probabilistic topic model which aims to infer the hidden topical structure in a collection of documents. According to LDA, documents can be seen as mixtures of topics, while topics can be seen as mixtures of words (in our case, tags). The first approach that we propose, called topic words based approach, recommends the top words in the top topics representing a resource as tags for that particular resource. The second approach, called topic distance based approach, uses the tags of the most similar training resources (identified using the KL-divergence Kullback and Liebler [1951]) to recommend tags for a test untagged resource. The dataset used in this work was made available through the ECML/PKDD Discovery Challenge 2009. We construct the documents that are provided as input to LDA in two ways, thus producing two different datasets. In the first dataset, we use only the description and the tags (when available) corresponding to a URL. In the second dataset, we crawl the URL content and use it to construct the document. Experimental results show that the LDA approach is not very effective at recommending tags for new untagged resources. However, using the resource content gives better results than using the description only. Furthermore, the topic distance based approach is better than the topic words based approach, when only the descriptions are used to construct documents, while the topic words based approach works better when the contents are used to construct documents.
258

Converged stepped spillway models in OpenFOAM

Sweeney, Brian P. January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / Mitchell L. Neilsen / The United States Department of Agriculture (USDA) is currently researching the effectiveness of various earth dam designs and their ability to prevent erosion. This report utilizes experimental results from the USDA experimental hydraulic engineering research unit to develop computational fluid dynamics models using OpenFOAM. Several variations of smooth and stepped dam models are created and analyzed with OpenFOAM on multiple cores using Message Passing Interface. In this report, seven dam designs are analyzed to extract flow velocities and pressures and animations. This data and OpenFOAM models are helpful for determining potential erosion conditions.
259

SOS - an android application for emergencies

Suryawanshi, Akash January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / Daniel A. Andresen / The aim of the project is to develop an Android application that lets its users to send notifications in case of an emergency or a panic situation. The users can send multiple text messages and emails on the press of a single button. The phone numbers, email ids and the contents of the text and email messages can be set from within the application. The text messages and emails sent, along with the content, also have the last known location of the user. This is very helpful in tracking the whereabouts of the person. The user can also call 911 directly from within the application, if the nature of the situation demands it. Additionally the user of the application may allow the app to track their location. If this option is selected, the application fetches the device’s location at about every 15 minutes and stores it in a database. This information is very useful and can be used in a variety of ways. One such use of the location data is from within the Android app where the user can view a map that shows their location history over a period of time for a particular day.
260

Elderly support - android application for fall detection and tracking

Rudraraju, Tejitha January 1900 (has links)
Master of Science / Department of Computing and Information Sciences / Mitchell L. Neilsen / The purpose of the project is to develop an Android application that is capable of detecting possible falls for the elderly. With the advancement of wireless communications, the world has become smarter and there has been increase in use of smart phones. Android, being an open source platform has made it simple for every individual to develop their own applications, which in turn can be used on Android devices. Falls among the elderly are a serious concern for both families as well as medical professionals, since falls are considered to be the eighth leading cause of death in the United States. Untreated fall injuries in adults 65 or older can result in serious risks and health complications, since 20% of falls require immediate medical attention and about one-tenth of the falls result in fractures. Thus, as mentioned, fall detection is a critical event requiring quick and accurate response, especially for elderly people living by themselves. This is the motivation behind developing an elderly support, which detects a fall and alerts the caretaker regarding the information. The elderly support is about tracking the person and notifying the caretaker if there is an occurrence. One of the main features of this Android application is that it notifies the caretaker with alert messages which contain all the necessary information. The alert messages contain useful information about the people in danger, such as his/her geo location and also corresponding directions on a map. In occasions of false alerts, the supervised person is given the ability to estimate the value of importance of a possible alert and to stop it before proceeding with further steps. This project is geared towards supporting the elderly. The mobile application is capable of detecting possible falls and through a user-friendly interface that can be used to alert relatives, doctors, and other people who take care of the elderly.

Page generated in 0.0249 seconds