Technological innovations

Provision of complete and responsive solution to healthcare services requires a multi-tired health delivery system. One of the aspects of healthcare hierarchy is the need for nursing care of the patient. Nursing care and observation provide basis for nurses to communicate with other aspects of healthcare system. The ability of capturing and managing nursing practice is essential to the quality of human care. The thesis proposes knowledge based decision making and analyzing system for the nurses to capture and manage the nursing practice. Moreover it allows them to monitor nursing care quality, as well as to test an aspect of an electronic healthcare record for recording and reporting nursing practice. The framework used for this system is based on nursing theory and is coupled with the quantitative analysis of qualitative data. It allows us to quantify the qualitative raw natural nursing language data. The results are summarized in the graph that shows the relative importance of those attributes with respect to each other at different instances of nurse-patient encounter. Research has been conducted by the Department of Computer and Electrical Engineering and Computer Science for the College of Nursing.

The purpose of this study was to analyze the biology content, instructional strategies, and assessment methods of 100 biological science websites that were appropriate for Grade 12 educational purposes. For the analysis of each website, an instrument, developed from the National Science Education Standards (NSES) for Grade 12 Life Science coupled with criteria derived from the Web-based Inquiry (WBI) for Learning Science instrument (Bodzin, 2005) and other pertinent published educational literature, was utilized. The analysis focused on elucidating the appropriateness of the biology content, instructional strategies, and assessment tools of selected websites for facilitating the biological science education of Grade 12 students. Frequencies of agreement and disagreement of the content of each selected website with criteria included in the data collection instrument were used for alignment determination of the content of each website with the NSES. Chi-square tests were performed by Microsoft Excel to determine the statistical significance of differences of actual and expected 85% frequencies of alignment of the analyzed website parameters with indicators of alignment to NSES. Chi-square tests indicated that at a 0.05 level of significance there was an overall difference between the actual and expected 85% frequencies of alignment of biology content, instructional strategies and assessment methods with website indicators of alignment with the NSES (p < 0.05). Chi-square tests also indicated that there was a significant difference between the actual and expected frequencies of alignment of analyzed categories (biology content, instructional strategies, and assessment methods) of the sampled websites with website indicators of alignment with the NSES (p < 0.05).

Delay tolerant networks (DTNs) are occasionally-connected networks that may suffer from frequent partitions. DTNs provide service despite long end to end delays or infrequent connectivity. One fundamental problem in DTNs is routing messages from their source to their destination. DTNs differ from the Internet in that disconnections are the norm instead of the exception. Representative DTNs include sensor-based networks using scheduled intermittent connectivity, terrestrial wireless networks that cannot ordinarily maintain end-to-end connectivity, satellite networks with moderate delays and periodic connectivity, underwater acoustic networks with moderate delays and frequent interruptions due to environmental factors, and vehicular networks with cyclic but nondeterministic connectivity. The focus of this dissertation is on routing protocols that send messages in DTNs. When no connected path exists between the source and the destination of the message, other nodes may relay the message to the destination. This dissertation covers routing protocols in DTNs with both deterministic and non-deterministic mobility respectively. In DTNs with deterministic and cyclic mobility, we proposed the first routing protocol that is both scalable and delivery guaranteed. In DTNs with non-deterministic mobility, numerous heuristic protocols are proposed to improve the routing performance. However, none of those can provide a theoretical optimization on a particular performance measurement. In this dissertation, two routing protocols for non-deterministic DTNs are proposed, which minimizes delay and maximizes delivery rate on different scenarios respectively. First, in DTNs with non-deterministic and cyclic mobility, an optimal single-copy forwarding protocol which minimizes delay is proposed.

Many emerging mobile networks aim to provide wireless network services without relying on any infrastructure. The main challenge in these networks comes from their self-organized and distributed nature. There is an inherent reliance on collaboration among the participants in order to achieve the aimed functionalities. Therefore, establishing and quantifying trust, which is the driving force for collaboration, is important for applications in mobile networks. This dissertation focuses on evaluating and quantifying trust to stimulate collaboration in mobile networks, introducing uncertainty concepts and metrics, as well as providing the various analysis and applications of uncertainty-aware reputation systems. Many existing reputation systems sharply divide the trust value into right or wrong, thus ignoring another core dimension of trust: uncertainty. As uncertainty deeply impacts a node's anticipation of others' behavior and decisions during interaction, we include it in the reputation system. Specifically, we use an uncertainty metric to directly reflect a node's confidence in the sufficiency of its past experience, and study how the collection of trust information may affect uncertainty in nodes' opinions. Higher uncertainty leads to higher transaction cost and reduced acceptance of communication. We exploit mobility to efficiently reduce uncertainty and to speed up trust convergence. We also apply the new reputation system to enhance the analysis of the interactions among mobile nodes, and present three sample uncertainty-aware applications. We integrate the uncertainty-aware reputation model with game theory tools, and enhance the analysis on interactions among mobile nodes.

Tinnitus is a conscious perception of phantom sounds in the absence of external acoustic stimuli, and masking is one of the popular ways to treat it. Due to the variation in the perceived tinnitus sound from patient to patient, the usefulness of masking therapy cannot be generalized. Thus, it is important to first determine the feasibility of masking therapy on a particular patient, by quantifying the tinnitus sound, and then generate an appropriate masking signal. This paper aims to achieve this kind of individual profiling by developing interactive software -Tinnitus Analyzer, based on clinical approach. The developed software has been proposed to be used in place of traditional clinical methods and this software (as a part of the future work) will be implemented in the practical scenario involving real tinnitus patients.

Sensors are used to monitor and control the physical environment. A Wireless Sen- sor Network (WSN) is composed of a large number of sensor nodes that are densely deployed either inside the phenomenon or very close to it [18][5]. Sensor nodes measure various parameters of the environment and transmit data collected to one or more sinks, using hop-by-hop communication. Once a sink receives sensed data, it processes and forwards it to the users. Sensors are usually battery powered and it is hard to recharge them. It will take a limited time before they deplete their energy and become unfunctional. Optimizing energy consumption to prolong network lifetime is an important issue in wireless sensor networks. In mobile sensor networks, sensors can self-propel via springs [14], wheels [20], or they can be attached to transporters, such as robots [20] and vehicles [36]. In static sensor networks with uniform deployment (uniform density), sensors closest to the sink will die first, which will cause uneven energy consumption and limitation of network life- time. In the dissertation, the nonuniform density is studied and analyzed so that the energy consumption within the monitored area is balanced and the network lifetime is prolonged. Several mechanisms are proposed to relocate the sensors after the initial deployment to achieve the desired density while minimizing the total moving cost. Using mobile relays for data gathering is another energy efficient approach. Mobile sensors can be used as ferries, which carry data to the sink for static sensors so that expensive multi-hop communication and long distance communication are reduced. In this thesis, we propose a mobile relay based routing protocol that considers both energy efficiency and data delivery delay. It can be applied to both event-based reporting and periodical report applications.

Collaborative filtering (CF), a very successful recommender system, is one of the applications of data mining for incomplete data. The main objective of CF is to make accurate recommendations from highly sparse user rating data. My contributions to this research topic include proposing the frameworks of imputation-boosted collaborative filtering (IBCF) and imputed neighborhood based collaborative filtering (INCF). We also proposed a model-based CF technique, TAN-ELR CF, and two hybrid CF algorithms, sequential mixture CF and joint mixture CF. Empirical results show that our proposed CF algorithms have very good predictive performances. In the investigation of applying imputation techniques in mining incomplete data, we proposed imputation-helped classifiers, and VCI predictors (voting on classifications from imputed learning sets), both of which resulted in significant improvement in classification performance for incomplete data over conventional machine learned classifiers, including kNN, neural network, one rule, decision table, SVM, logistic regression, decision tree (C4.5), random forest, and decision list (PART), and the well known Bagging predictors. The main imputation techniques involved in these algorithms include EM (expectation maximization) and BMI (Bayesian multiple imputation).

Traditional methods such as distance weighing, correlation and data driven methods have been used in the estimation of missing precipitation data. Also common is the use of radar (NEXRAD) data to provide better spatial distribution of precipitation as well as infilling missing rain gage data. Conventional regression models are often used to capture highly variant nonlinear spatial and temporal relationships between NEXRAD and rain gage data. This study aims to understand and model the relationships between radar (NEXRAD) estimated rainfall data and the data measured by conventional rain gages. The study is also an investigation into the use of emerging computational data modeling (inductive) techniques and mathematical programming formulations to develop new optimal functional approximations. Radar based rainfall data and rain gage data are analyzed to understand the spatio-temporal associations, as well as the effect of changes in the length or availability of data on the models. The upper and lower Kissimmee basins of south Florida form the test-bed to evaluate the proposed and developed approaches and also to check the validity and operational applicability of these functional relationships among NEXRAD and rain gage data for infilling of missing data.

The field of Video Transcoding has been evolving throughout the past ten years. The need for transcoding of video files has greatly increased because of the new upcoming standards which are incompatible with old ones. This thesis takes the method of using machine learning for video transcoding mode decisions and discusses ways to improve the process of generating the algorithm for implementation in different video transcoders. The transcoding methods used decrease the complexity in the mode decision inside the video encoder. Also methods which automate and improve results are discussed and implemented in two different sets of transcoders: H.263 to VP6 , and MPEG-2 to H.264. Both of these transcoders have shown a complexity loss of almost 50%. Video transcoding is important because the quantity of video standards have been increasing while devices usually can only decode one specific codec.

H.264/AVC encoder complexity is mainly due to variable size in Intra and Inter frames. This makes H.264/AVC very difficult to implement, especially for real time applications and mobile devices. The current technological challenge is to conserve the compression capacity and quality that H.264 offers but reduce the encoding time and, therefore, the processing complexity. This thesis applies machine learning technique for video encoding mode decisions and investigates ways to improve the process of generating more general low complexity H.264/AVC video encoders. The proposed H.264 encoding method decreases the complexity in the mode decision inside the Inter frames. Results show, at least, a 150% average reduction of complexity and, at most, 0.6 average increases in PSNR for different kinds of videos and formats.