Master of Science / Department of Computing and Information Sciences / Eugene Vasserman / The Medical Device Coordination Framework (MDCF) is an open source Medical Application Platform (MAP) that facilitates interoperability between heterogeneous medical devices. The MDCF is designed to be an open test bed for the conceptual architecture described by the Integrated Clinical Environment (ICE) interoperability standard. In contrast to existing medical device connectivity features that only provide data logging and display capabilities, a MAP such as the MDCF also allows medical devices to be controlled by apps. MAPs are predicted to enable many improvements to health care, however they also create new risks to patient safety and privacy that need to be addressed. As a result, MAPs such as the MDCF and other ICE-like systems require the integration of security features. This thesis lays the groundwork for a comprehensive security architecture within the MDCF. Specifically, we address the need for access control, device certification, communication security, and device authentication. We begin by describing a system for ensuring the trustworthiness of medical devices connecting to the MDCF. To demonstrate trustworthiness of a device, we use a chain of cryptographic certificates which uniquely identify that device and may also serve as non- forgeable proof of regulatory approval, safety testing, or compliance testing. Next, we cover the creation and integration of a pluggable, flexible authentication system into the MDCF, and evaluate the performance of proof-of-concept device authentication providers. We also discuss the design and implementation of a communication security system in the MDCF, which enables the creation and use of communication security providers which can provide data confidentiality, integrity, and authenticity. We conclude this work by presenting the requirements and a high level design for a Role-Based Access Control (RBAC) system within the MDCF.

Doctor of Philosophy / Department of Computing and Information Sciences / Scott A. DeLoach / The Organization-Based Multi-Agent Systems (OMAS) paradigm is an approach to address the challenges posed by complex systems. The complexity of these systems, the changing environment where the systems are deployed, and satisfying higher user expectations are some of current requirements when designing OMAS. For the agents in an OMAS to pursue the achievement of a common goal or task, a certain level of coordination and collaboration occurs among them. An objective in this coordination is to make the decision of who does what. Several solutions have been proposed to answer this task allocation question. The majority of the solutions proposed fall in the categories of marked-based approaches, reactive systems, or game theory approaches. A common fact among these solutions is the system information sharing among agents, which is used only to keep the participant agent informed about other agents activities and mission status. To further exploit and take advantage of this system information shared among agents, a framework is proposed to use this information to answer the question who does what, and reduce the communication among agents. DTAACS-OK is a distributed knowledge-based framework that addresses the Single Agent Task Allocation Problem (SAT-AP) and the Multiple Agent Task Allocation Problem (MAT-AP) in cooperative OMAS. The allocation of tasks is based on an identical organization knowledge posses by all agents in the organization. DTAACS-OK di ers with current solutions in that (a) it is not a marked-based approach where task are auctioned among agents, or (b) it is not based on agents behaviour, where the action or lack of action of an agent cause the reaction of other agents in the organization.

Master of Science / Department of Computing and Information Sciences / Mitchell L. Neilsen / The mobile world is rapidly changing: Smartphones have gone from portable messaging and email devices to streaming-video machines that surf the Web at blazing speed. Now-a-days a smartphone can provide computing capabilities, wireless communication capabilities, run software and perform other tasks just like any traditional computer. These amazing features of a smartphone and Open Source Android market helped in the development of this project. The purpose of this project is to develop an Android application for controlling various elements of user environment. User Controlled Environment is an Android application for home. The environment consists of smart lights, an Android mobile devices for playing music and a display. The application sends the user’s preferred settings to the environment and the respective settings are applied. The preferences are displayed on the screen. The user will be able to view and adjust a variety of environmental preferences. The preferences include the light’s color, light intensity, and the music. When the user exits the application the environment goes to a default state. The users can set preferences which include moods, seven colors of light, three levels of light intensity and songs that the users can select.

Master of Science / Department of Computing and Information Sciences / Daniel Andresen / This report considers a native Android application called “KSUSoy YieldCalc” that assists in yield estimation of soybean before harvest following conventional approach. Android is one of the most popular installed base of any mobile platform, powering many mobile devices in more than 190 countries for users of diverse economic backgrounds, making it more popular than iOS devices (Android developers n.d.).The project “KSUSoy YieldCalc” adopted the Android platform as its base to serve farmers, agronomists, and consultants and deliver performance to save time and enhance farmers’ their confidence. The native application uses “conventional approach” of estimation of yield for calculations and eliminates the need for having Internet connection to access, thereby increasing the application’s flexibility. The project utilized Android Software Development Kit (SDK) as its development platform with extensive Java and Extensible Markup Language (XML) coding. The Department of Agronomy at Kansas State University (KSU) tested the application with promising results. Dr. Ignacio Ciampitti of the Department of Agronomy at KSU currently demonstrates the application to farmers. User feedback has been very satisfactory to date.

Master of Science / Department of Computing and Information Sciences / Daniel A. Andresen / Present times most of the applications allow users to set profiles and activate them manually. Also, there is no such application to alert the calling person automatically whenever we are busy in a meeting. Auto Profile is an application which addresses these issues. Using this application, a custom profile can be created and activated. The user of this application can create any number of custom profiles with their own timings depending on their requirements from the available set of ringer modes such as vibration, silent, normal modes. All these settings will be changed automatically when they define a profile and set it as active. Every profile can be scheduled according to their wish. Also, users can define how often and how long they would like to activate the selected profile. Auto Profile Switcher will recognize their profile based on the current time and it will apply the right profile at the right time according to the given settings. For instance, they can save a home profile and a work profile, and the application will automatically switch from one to other automatically, without any user action. It's also possible for the users to apply manually saved profile wherever they are. Also, the application allows users not to bother about the calls from important persons when they are in some important meeting or busy. Automatically profile is activated as scheduled and sends a respective alert to the caller according to the profile activated. Auto Profile application is user-friendly and easy to use. The proposed application is developed for the android platform, which is used to create custom profiles and make sure that the tasks are executed as scheduled. This relieves users as they do not need to bother about the calls when they are in some important meeting or busy. As this is a mobile application users can easily organize their action wherever they are with ease.

Doctor of Philosophy / Department of Computing and Information Sciences / Doina Caragea / Recent advancements in biological sciences have resulted in the availability of large amounts of sequence data (DNA and protein sequences). Biological sequence data can be annotated using machine learning techniques, but most learning algorithms require data to be represented by a vector of features. In the absence of biologically informative features, k-mers generated using a sliding window-based approach are commonly used to represent biological sequences. A larger k value typically results in better features; however, the number of k-mer features is exponential in k, and many k-mers are not informative. Feature selection is widely used to reduce the dimensionality of the input feature space. Most feature selection techniques use feature-class dependency scores to rank the features. However, when the amount of available labeled data is small, feature selection techniques may not accurately capture feature-class dependency scores. Therefore, instead of working with all k-mers, this dissertation proposes the construction of a reduced set of informative k-mers that can be used to represent biological sequences. This work resulted in three novel unsupervised approaches to construct features: 1. Burrows Wheeler Transform-based approach, that uses the sorted permutations of a given sequence to construct sequential features (subsequences) that occur multiple times in a given sequence. 2. Community detection-based approach, that uses a community detection algorithm to group similar subsequences into communities and refines the communities to form motifs (group of similar subsequences). Motifs obtained using the community detection-based approach satisfy the ZOMOPS constraint (Zero, One or Multiple Occurrences of a Motif Per Sequence). All possible unique subsequences of the obtained motifs are then used as features to represent the sequences. 3. Hybrid-based approach, that combines the Burrows Wheeler Transform-based approach and the community detection-based approach to allow certain mismatches to the features constructed using the Burrows Wheeler Transform-based approach. To evaluate the predictive power of the features constructed using the proposed approaches, experiments were conducted in three learning scenarios: supervised, semi-supervised, and domain adaptation for both nucleotide and protein sequence classification problems. The performance of classifiers learned using features generated with the proposed approaches was compared with the performance of the classifiers learned using k-mers (with feature selection) and feature hashing (another unsupervised dimensionality reduction technique). Experimental results from the three learning scenarios showed that features constructed with the proposed approaches were typically more informative than k-mers and feature hashing.

Master of Science / Department of Computing and Information Sciences / Scott A. DeLoach / A geographic information system (GIS) allows an individual to map, model, query, and analyze large quantities of data from a database according to their spatial locations. This project uses the ArcGis Java software Development Kit (SDK) to visualize, manipulate, and comprehend large amounts of publicly available information relevant to a spatial location. The application developed uses a graphical user interface to examine the public data of Riley County, Kansas. The user is able to load shapefiles through the interface and then examine the many spatial locations. By examining a spatial location the user is able to view the associated attribute information, manipulate it, and add additional attributes. Beyond viewing information at selected geometric locations, a user can also query the layer(s) to return the spatial locations that fit the query. These abilities can allow a user to understand and visualize patterns that they would not have been able to easily see from looking at the raw data. Increasing users' understanding of the environment they are working with improves their likelihood of success in their desired objectives.

Master of Science / Department of Computing and Information Sciences / Gurdip Singh / TinyOS is a component based operating system written in nesC programming language. TinyOS provides interfaces and components for common low level abstractions such as packet communication, routing and sensing for node level sensor network application programming. This project aims to provide high level abstractions to users by providing the notion of a virtual node, which represents a set of physical nodes, allowing users to specify global scenarios, and a mechanism to decompose a high level global scenario into local node level scenarios for each of the individual sensor nodes. A global scenario with virtual components, provided by the user, is first converted into a global scenario by eliminating the virtual components from the model by using a mapping information provided the user and replacing these virtual components by their respective physical components. Appropriate algorithm components and the automatically generated adapter components for these algorithm components are then plugged-in to implement inter-node interactions. This global scenario is then converted to the node level local scenarios by introducing the automatically generated proxy components for the remote components and connecting these proxy components using the RMI layer. The Cadena model is modified to include the attribute location for the components to identify the remote components. The make files are then generated for these local scenarios and are ready to be deployed on the physical motes. The framework provides a GUI tool which is used to visualize the data of the sensor network in both simulation and deployment. The framework provides the user with commands that can be issued to the network from the Cadena component model as a set of interfaces to the components and a python script is used to capture this information in an xml file. The Cadena model is modified to include the attribute observable to the interfaces to identify them as the GUI commands. The GUI loads this XML file and the topology file for the actual deployment, can issue commands to the network and displays the results to the user. The GUI tool also enhances the Tossim simulator to model the external effects over the sensor network and to place the motes based on the topology information using the Tython environment.

Master of Science / Department of Computing and Information Sciences / William H. Hsu / This thesis discusses the background and methodologies necessary for constructing features in order to discover hidden links in relational data. Specifically, we consider the problems of predicting, classifying and annotating friends relations in friends networks, based upon features constructed from network structure and user profile data. I first document a data model for the blog service LiveJournal, and define a set of machine learning problems such as predicting existing links and estimating inter-pair distance. Next, I explain how the problem of classifying a user pair in a social networks, as directly connected or not, poses the problem of selecting and constructing relevant features. In order to construct these features, a genetic programming approach is used to construct multiple symbol trees with base features as their leaves; in this manner, the genetic program selects and constructs features that many not have been considered, but possess better predictive properties than the base features. In order to extract certain graph features from the relatively large social network, a new shortest path search algorithm is presented which computes and operates on a Euclidean embedding of the network. Finally, I present classification results and discuss the properties of the frequently constructed features in order to gain insight on hidden relations that exists in this domain.

Master of Science / Department of Computing and Information Sciences / Doina Caragea / In recent years, we have witnessed a significant increase in the number, size and diversity of the available data sources in many application domains. Data sources in a particular domain are autonomously created and maintained, and therefore distributed and semantically heterogeneous. In this thesis, we focused on the problem of querying such semantically heterogeneous data sources from a user's perspective. We approach this problem by using the concepts of ontologies and mappings between ontologies. A system for answering queries in a transparent way to the user has been designed and implemented. The main components of this system are an ontology mapping algorithm that maps user ontologies to data source ontologies, and a query processing engine that maps user queries to queries that can be answered by the data sources in the system. We have shown that machine learning algorithms can also be incorporated in the system, thus making it possible to learn machine learning classifiers (in particular, generative models such as Naïve Bayes) from distributed, semantically heterogeneous data sources. Because many data sources today are relational in nature, in this work we have dealt specifically with relational data sources, as opposed to flat files, XML or object oriented data sources. However, our system can be easily extended to other types of data sources.