Thesis (Ph. D.)--University of Hawaii at Manoa, 1995. / Includes bibliographical references (leaves 220-227). / Microfiche. / xiii, 227 leaves, bound ill. (some col.) 29 cm

Thesis (M.S.)--Rochester Institute of Technology, 1988. / Includes bibliographical references.

Thesis (MTech (Information Technology))--Peninsula Technikon, 2004. / Word processed copy. Summary in English. Includes bibliographical references (leaves 117-120). Also available online.

Thesis (M.S.)--Worcester Polytechnic Institute. / Keywords: loss; multicast. Includes bibliographical references (p. 78-80).

With the increasing availability of large amount of data in a wide range of applications, no matter for industry or academia, it becomes crucial to understand the nature of complex raw data, in order to gain more values from data engineering. Although many problems have been successfully solved by some mature machine learning techniques, the problem of learning from imbalanced data continues to be one of the challenges in the field of data engineering and machine learning, which attracted growing attention in recent years due to its complexity. In this thesis, we focus on four aspects of imbalanced data learning and propose solutions to the key problems. The first aspect is about ensemble methods for imbalanced data classification. Ensemble methods, e.g. bagging and boosting, have the advantages to cure imbalanced data by integrated with sampling methods. However, there are still problems in the integration. One problem is that undersampling and oversampling are complementary each other and the sampling ratio is crucial to the classification performance. This thesis introduces a new method HSBagging which is based on bagging with hybrid sampling. Experiments show that HSBagging outperforms other state-of-the-art bagging method on imbalanced data. Another problem is about the integration of boosting and sampling for imbalanced data classification. The classifier weights of existing AdaBoost-based methods are inconsistent with the objective of class imbalance classification. In this thesis, we propose a novel boosting optimization framework GOBoost. This framework can be applied to any boosting-based method for class imbalance classification by simply replacing the calculation of classifier weights. Experiments show that the GOBoost-based methods significantly outperform the corresponding boosting-based methods. The second aspect is about online learning for imbalanced data stream with concept drift. In the online learning scenario, if the data stream is imbalanced, it will be difficult to detect concept drifts and adapt the online learner to them. The ensemble classifier weights are hard to adjust to achieve the balance between the stability and adaptability. Besides, the classier built on samples in size-fixed chunk, which may be highly imbalanced, is unstable in the ensemble. In this thesis, we propose Adaptive Chunk-based Dynamic Weighted Majority (ACDWM) to dynamically weigh the individual classifiers according to their performance on the current data chunk. Meanwhile, the chunk size is adaptively selected by statistical hypothesis tests. Experiments on both synthetic and real datasets with concept drift show that ACDWM outperforms both of the state-of-the-art chunk-based and online methods. In addition to imbalanced data classification, the third aspect is about clustering on imbalanced data. This thesis studies the key problem of imbalanced data clustering called uniform effect within the k-means-type framework, where the clustering results tend to be balanced. Thus, this thesis introduces a new method called Self-adaptive Multi-prototype-based Competitive Learning (SMCL) for imbalanced clusters. It uses multiple subclusters to represent each cluster with an automatic adjustment of the number of subclusters. Then, the subclusters are merged into the final clusters based on a novel separation measure. Experimental results show the efficacy of SMCL for imbalanced clusters and the superiorities against its competitors. Rather than a specific algorithm for imbalanced data learning, the final aspect is about a measure of class imbalanced dataset for classification. Recent studies have shown that imbalance ratio is not the only cause of the performance loss of a classifier in imbalanced data classification. To the best of our knowledge, there is no any measurement about the extent of influence of class imbalance on the classification performance of imbalanced data. Accordingly, this thesis proposes a data measure called Bayes Imbalance Impact Index (B1³) to reflect the extent of influence purely by the factor of imbalance for the whole dataset. As a result we can therefore use B1³ to judge whether it is worth using imbalance recovery methods like sampling or cost-sensitive methods to recover the performance loss of a classifier. The experiments show that B1³ is highly consistent with improvement of F1score made by the imbalance recovery methods on both synthetic and real benchmark datasets. Two ensemble frameworks for imbalanced data classification are proposed for sampling rate selection and boosting weight optimization, respectively. 2. •A chunk-based online learning algorithm is proposed to dynamically adjust the ensemble classifiers and select the chunk size for imbalanced data stream with concept drift. 3. •A multi-prototype competitive learning algorithm is proposed for clustering on imbalanced data. 4. •A measure of imbalanced data is proposed to evaluate how the classification performance of a dataset is influenced by the factor of imbalance.

In recent years, multi-touch has been heralded as a revolution in humancomputer interaction. Multi-touch provides features such as gestural interaction, tangible interfaces, pen-based computing, and interface customization – features embraced by an increasingly tech-savvy public. However, multi-touch platforms have not been adopted as "everyday" computer interaction devices; that is, multi-touch has not been applied to general-purpose computing. The questions this thesis seeks to address are: Will the general public adopt these systems as their chief interaction paradigm? Can multi-touch provide such a compelling platform that it displaces the desktop mouse and keyboard? Is multi-touch truly the next revolution in human-computer interaction? As a first step toward answering these questions, we observe that generalpurpose computing relies on text input, and ask: "Can multi-touch, without a text entry peripheral, provide a platform for efficient text entry? And, by extension, is such a platform viable for general-purpose computing?" We investigate these questions through four user studies that collected objective and subjective data for text entry and word processing tasks. The first of these studies establishes a benchmark for text entry performance on a multi-touch platform, across a variety of input modes. The second study attempts to improve this performance by iv examining an alternate input technique. The third and fourth studies include mousestyle interaction for formatting rich-text on a multi-touch platform, in the context of a word processing task. These studies establish a foundation for future efforts in general-purpose computing on a multi-touch platform. Furthermore, this work details deficiencies in tactile feedback with modern multi-touch platforms, and describes an exploration of audible feedback. Finally, the thesis conveys a vision for a general-purpose multi-touch platform, its design and rationale.

Typescript (photocopy). / Digitized by Kansas Correctional Industries

Thesis (MScIng)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: Data capturing equipment are an expensive part of testing the coverage of a deployed or planned wireless service. This thesis presents the development of such a data capturing system that make use of 1800MHz GSM base stations as transmitters and a mobile phone connected to a laptop as receiver. The measurements taken, are then verified against know propagation models. Datavaslegging toerusting wat gebruik word om die dekking van draadlose stelsels te toets is baie duur en moeilik bekombaar. Hierdie tesis beskryf die ontwikkeling van so ’n datavaslegger wat baie goedkoper is en maklik gebruik kan word. Dit maak gebruik van ’n sellulêr foon en GPS gekoppel aan ’n skootrekenaar, wat die ontvanger is. Cell C basis staties word gebruik as die senders. Die data wat gemeet is word dan geverifieer deur gebruik te maak van bestaande radio frekwensie voortplanting modelle.

This thesis is concerned with the problem of processing data from Wireless Sensor Networks (WSNs) to meet the requirements of emergency responders (e.g. Fire and Rescue Services). A WSN typically consists of spatially distributed sensor nodes to cooperatively monitor the physical or environmental conditions. Sensor data about the physical or environmental conditions can then be used as part of the input to predict, detect, and monitor emergencies. Although WSNs have demonstrated their great potential in facilitating Emergency Response, sensor data cannot be interpreted directly due to its large volume, noise, and redundancy. In addition, emergency responders are not interested in raw data, they are interested in the meaning it conveys. This thesis presents research on processing and combining data from multiple types of sensors, and combining sensor data with other relevant data, for the purpose of obtaining data of greater quality and information of greater relevance to emergency responders. The current theory and practice in Emergency Response and the existing technology aids were reviewed to identify the requirements from both application and technology perspectives (Chapter 2). The detailed process of information extraction from sensor data and sensor data fusion techniques were reviewed to identify what constitutes suitable sensor data fusion techniques and challenges presented in sensor data processing (Chapter 3). A study of Incident Commanders' requirements utilised a goal-driven task analysis method to identify gaps in current means of obtaining relevant information during response to fire emergencies and a list of opportunities for WSN technology to fill those gaps (Chapter 4). A high-level Emergency Information Management System Architecture was proposed, including the main components that are needed, the interaction between components, and system function specification at different incident stages (Chapter 5). A set of state-awareness rules was proposed, and integrated with Kalman Filter to improve the performance of filtering. The proposed data pre-processing approach achieved both improved outlier removal and quick detection of real events (Chapter 6). A data storage mechanism was proposed to support timely response to queries regardless of the increase in volume of data (Chapter 7). What can be considered as “meaning” (e.g. events) for emergency responders were identified and a generic emergency event detection model was proposed to identify patterns presenting in sensor data and associate patterns with events (Chapter 8). In conclusion, the added benefits that the technical work can provide to the current Emergency Response is discussed and specific contributions and future work are highlighted (Chapter 9).

The main goal of video coding algorithms is to achieve high compression efficiency while maintaining quality of the compressed signal at the highest level. Human visual system is the ultimate receiver of compressed signal and final judge of its quality. This dissertation presents work towards optimal video compression algorithm that is based on the characteristics of our visual system. Modeling phenomena such as backward temporal masking and motion masking we developed algorithms that are implemented in the state-of- the-art video encoders. Result of using our algorithms is visually lossless compression with improved efficiency, as verified by standard subjective quality and psychophysical tests. Savings in bitrate compared to the High Efficiency Video Coding / H.265 reference implementation are up to 45%. / Includes bibliography. / Dissertation (Ph.D.)--Florida Atlantic University, 2014. / FAU Electronic Theses and Dissertations Collection