published_or_final_version / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

This study resulted in the creation of a conceptual framework for ethical academic writing that can be applied to cases of authorship identification. The framework is the culmination of research into various other forensic frameworks and aspects related to cyber forensics, in order to ensure maximum effectiveness of this newly developed methodology. The research shows how synergies between forensic linguistics and electronic forensics (computer forensics) create the conceptual space for a new, interdisciplinary, cyber forensic linguistics, along with forensic auditing procedures and tools for authorship identification. The research also shows that an individual’s unique word pattern usage can be used to determine document authorship, and that in other instances, authorship can be attributed with a significant degree of probability using the identified process. The importance of this fact cannot be understated, because accusations of plagiarism have to be based on facts that will withstand cross examination in a court of law. Therefore, forensic auditing procedures are required when attributing authorship in cases of suspected plagiarism, which is regarded as one of the most serious problems facing any academic institution. This study identifies and characterises various forms of plagiarism as well the responses that can be implemented to prevent and deter it. A number of online and offline tools for the detection and prevention of plagiarism are identified, over and above the more commonly used popular tools that, in the author’s view, are overrated because they are based on mechanistic identification of word similarities in source and target texts, rather than on proper grammatical and semantic principles. Linguistic analysis is a field not well understood and often underestimated. Yet it is a critical field of inquiry in determining specific cases of authorship. The research identifies the various methods of linguistic analysis that could be applied to help establish authorship identity, as well as how they can be applied within a forensic environment. Various software tools that could be used to identify and analyse source documents that were plagiarised are identified and briefly characterised. Concordance, function word analysis and other methods of corpus analysis are explained, along with some of their related software packages. Corpus analysis that in the past would have taken months to perform manually, could now only take a matter of hours using the correct programs, given the availability of computerised analysis tools. This research integrates the strengths of these tools within a structurally sound forensic auditing framework, the result of which is a conceptual framework that encompasses all the pertinent factors and ensures admissibility in a court of law by adhering to strict rules and features that are characteristic of the legal requirements for a forensic investigation. / Thesis (M.Com.)-University of KwaZulu-Natal, Westville, 2009.

Thesis (PhD (Mathematical Sciences))--University of Stellenbosch, 2005. / A great deal of work has been done in the area of off-line signature verification over the past two decades. Off-line systems are of interest in scenarios where only hard copies of signatures are available, especially where a large number of documents need to be authenticated. This dissertation is inspired by, amongst other things, the potential financial benefits that the automatic clearing of cheques will have for the banking industry.

Dong, Zhuxin. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 77-80). / Abstracts in English and Chinese. / Abstract --- p.ii / 摘要 --- p.iii / Acknowledgements --- p.iv / Table of Contents --- p.v / List of Figures --- p.vii / List of Tables --- p.ix / Chapter Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Architecture of UDWI --- p.3 / Chapter 1.2 --- Background of IMU Sensor Calibration --- p.5 / Chapter 1.3 --- Organization --- p.7 / Chapter Chapter 2 --- 2D Motion Calibration --- p.10 / Chapter 2.1 --- Experimental Platform --- p.10 / Chapter 2.1.1 --- Transparent Table --- p.10 / Chapter 2.2 --- Matching Algorithm --- p.13 / Chapter 2.2.1 --- Motion Analysis --- p.13 / Chapter 2.2.2 --- Core Algorithm and Matching Criterion --- p.14 / Chapter 2.3 --- Usage of High Speed Camera --- p.17 / Chapter 2.4 --- Functions Realized --- p.17 / Chapter Chapter 3 --- Usage of Camera Calibration --- p.21 / Chapter 3.1 --- Introduction to Camera Calibration --- p.21 / Chapter 3.1.1 --- Related Coordinate Frames --- p.21 / Chapter 3.1.2 --- Pin-Hole Model --- p.24 / Chapter 3.2 --- Calibration for Nonlinear Model --- p.27 / Chapter 3.3 --- Implementation of Process to Calibrate Camera --- p.28 / Chapter 3.3.1 --- Image Capture --- p.28 / Chapter 3.3.2 --- Define World Frame and Extract Corners --- p.28 / Chapter 3.3.3 --- Main Calibration --- p.30 / Chapter 3.4 --- Calibration Results of High Speed Camera --- p.33 / Chapter 3.4.1 --- Lens Selection --- p.33 / Chapter 3.4.2 --- Property of High Speed Camera --- p.34 / Chapter Chapter 4 --- 3D Attitude Calibration --- p.36 / Chapter 4.1 --- The Necessity of Attitude Calibration --- p.36 / Chapter 4.2 --- Stereo Vision and 3D Reconstruction --- p.37 / Chapter 4.2.1 --- Physical Meaning and Mathematical Model Proof --- p.37 / Chapter 4.2.2 --- 3D Point Reconstruction --- p.38 / Chapter 4.3 --- Example of 3D Point Reconstruction --- p.40 / Chapter 4.4 --- Idea of Attitude Calibration --- p.42 / Chapter Chapter 5 --- Experimental Results --- p.45 / Chapter 5.1 --- Calculation of Proportional Parameter --- p.45 / Chapter 5.2 --- Accuracy Test of Stroke Reconstruction --- p.46 / Chapter 5.3 --- Writing Experiments of 26 Letters --- p.47 / Chapter 5.3.1 --- Experimental Results of Letter b --- p.48 / Chapter 5.3.2 --- Experimental Results of Letter n with ZVC --- p.51 / Chapter 5.3.3 --- Experimental Results of Letter u --- p.54 / Chapter 5.4 --- Writing of Single Letter s - Multiple Tests --- p.56 / Chapter 5.5 --- Analysis on Resolution Property of Current Vision Algorithm --- p.58 / Chapter 5.5.1 --- Resolution of Current Algorithm --- p.58 / Chapter 5.5.2 --- Tests with Various Filters --- p.59 / Chapter 5.6 --- Calculation of Static Attitude --- p.61 / Chapter Chapter 6 --- Future Work --- p.64 / Chapter 6.1 --- Another Multiple Tests of Letter k --- p.64 / Chapter 6.2 --- Letter Recognition Based on Neural Networks Classification --- p.66 / Chapter Chapter 7 --- Conclusion --- p.69 / Chapter 7.1 --- Calibration ofMAG-μlMU Sensors --- p.69 / Chapter 7.2 --- Calibration of Accelerometers --- p.70 / Chapter 7.3 --- Calibration of Attitude --- p.70 / Chapter 7.4 --- Future Work --- p.71 / Appendix A The Experimental Results of Writing English Letters --- p.72

published_or_final_version / English / Master / Master of Philosophy

The purpose of this study was to investigate the relationship of word processor use by foreign college writers and their attitudes toward writing, writing revision practices, writing quality, attitudes toward the use of computers, and time spent on computers. The results indicate that students' attitudes toward writing and their perceptions of computer usefulness significantly affected their writing quality. Students with more positive attitudes toward writing and the usefulness of computers tended to produce better quality writing. In addition, the findings indicate that students' writing revision practices significantly affected their attitudes toward writing. Students who revised their writing more frequently tended to have better attitudes toward writing than those who did not. In contrast, students' levels of computer anxiety, computer confidence, computer liking and their writing revision practices did not significantly affect the quality of their writing. Furthermore, the amount of time that students spent on computers did not significantly affect their attitudes toward using computers in writing.

University of Macau / Faculty of Science and Technology. / Department of Computer and Information Science