Debugging is a challenging and time-consuming process in software life-cycle. The focus of the thesis is to improve the accuracy of existing fault localization (FL) techniques. We experimented with several source code line level features such as line commit size, line recency, and line length to arrive at a new fault localization technique. Based on our experiments, we propose a novel enhanced cost-aware fault localization (ECFL) technique by combining line length with the existing selected baseline fault localization techniques. ECFL improves the accuracy of DStar (Baseline 1), CombineFastestFL (Baseline 2), and CombineFL (Baseline 3) by locating 81%, 58%, and 30% more real faults respectively in Top-1 evaluation metric. In comparison with the baseline techniques, ECFL requires a marginal additional time (on an average, 5 seconds per bug) and data while providing a significant improvement in accuracy. The source code line features also improve the baseline fault localization techniques when ''learning to rank'' SVM machine learning approach is used to combine the features. We also provide an infrastructure to facilitate future research on combining new source code line features with other fault localization techniques. / Master of Science / Software debugging involves locating and fixing faults (or bugs) in software. It is a challenging and time-consuming process in software life-cycle. Fault localization (FL) techniques help software developers to locate faults by providing a ranked set of program elements. The focus of the thesis is to improve the accuracy of existing fault localization techniques. We experimented with several source code line level features such as line commit size, line recency, and line length to arrive at a new fault localization technique. Based on our experiments, we propose a novel enhanced cost-aware fault localization (ECFL) technique by combining line length with the existing selected baseline fault localization techniques. ECFL improves the accuracy of DStar (Baseline 1), CombineFastestFL (Baseline 2), and CombineFL (Baseline 3) by locating 81%, 58%, and 30% more real faults respectively in Top-1 evaluation metric. In comparison with the baseline techniques, ECFL requires a marginal additional time (on an average, 5 seconds per bug) and data while providing a significant improvement in accuracy. The source code line features also improve the baseline fault localization techniques when machine learning approach is used to combine the features. We also provide an infrastructure to facilitate future research on combining new source code line features with other fault localization techniques.

Defected wheel are one the major reasons endangered state of railroad vehicles safety statue, due to vehicle derailment and worsen the quality of freight and passenger transportation. Therefore, timely defect detection for monitoring and detecting the state of defects is highly critical. This thesis presents a passive non-contact acoustic structural health monitoring approach using ultrasonic acoustic emissions (UAE) to detect certain defects on different structures, as well as, classifying the type of the defect on them. The acoustic emission signals used in this study are in the ultrasonic range (18-120 kHz), which is significantly higher than the majority of the research in this area thus far. For the proposed method, an impulse excitation, such as a hammer strike, is applied to the structure. In addition, ultrasound techniques have higher sensitivity to both surface and subsurface defects, which make the defect detection more accurate. Three structures considered for this study are: 1) a longitudinal beam, 2) a lifting weight, 3) an actual rail-wheel. A longitudinal beam was used at the first step for a better understanding of physics of the ultrasound propagation from the defect, as well, develop a method for extracting the signature response of the defect. Besides, the inherent directionality of the ultrasound microphone increases the signal to noise ratio (SNR) and could be useful in the noisy areas. Next, by considering the ultimate goal of the project, lifting weight was chosen, due to its similarity to the ultimate goal of this project that is a rail-wheel. A detection method and metric were developed by using the lifting weight and two type of synthetic defects were classified on this structure. Also, by using same extracted features, the same types of defects were detected and classified on an actual rail-wheel. / Master of Science

Microfabrication of ceramic features has become a critical issue in realizing the miniaturization of devices. Micromolding and sintering play critical roles in fabricating micron- and submicron-sized ceramic features using nanoparticles. Developed from soft lithography, replica molding has been proven a good method to prepare micron- and submicron-sized features. However, the fidelity of the features can be compromised by incomplete feature cavity filling and feature shrinkage during the forming process. In this study, centrifuge-aided micromolding is developed to prepare micron- and submicron-sized ZnO features. By introducing a centrifugal force, the shear-thinning behavior of the suspensions is utilized, and the cavity filling process and the diffusion of trapped air out of the features are accelerated. The drying shrinkage is decreased by increasing the density of the wet nanoparticle packing from the centrifugal process. The centrifugal force improves the fidelity of all the designed features. ZnO ridges from 0.4 μm to 2 μm size and rods of 1.6 μm size are prepared successfully. The wide applicability of this strategy has been demonstrated by preparing ZrO2 features via the same method. Sintering process has a significant influence on the morphology and microstructural evolution of micron-sized ceramic features. When ceramic features decrease to much smaller sizes, such as in the micron range, the dominating sintering mechanism(s) can be different from those of the bulk at large scales. However, limited effort has been devoted to understanding the sintering behaviors. In this study, the as-prepared micron-sized ZnO ridges and rods were sintered at 950oC for different time in air atmosphere. The sintering process destructs the ZnO features via abnormal grain growth and surface roughening. Destruction prediction of features using sintering time is established based on grain growth. Feature surface roughening is further analyzed with respect to thermodynamic fundamentals. Because of the evaporation tendency during zinc oxide sintering, sintering atmosphere has a significant influence on the sintering behavior and feature fidelity. In this study, micron-sized ZnO ridge features were sintered under air and argon atmospheres. Ridge size, line edge roughness, and grain size were characterized. Quantitative calculation of sintering behaviors was performed in order to obtain fundamental understating of the micron-sized ZnO feature sintering. It is found that oxygen partial pressure is the deciding factor for the ridge feature evolution. ZnO evaporation and defects diffusion are responsible for the ZnO bulk and ridge sintering behavior differences. / Master of Science / In order to produce portable devices with small sizes, novel techniques are required to make small components, which is called microfabrication. Since ceramic materials are widely used in various electronic devices, microfabrication of small ceramic features has become an important issue. When ceramic nanoparticles are used as the raw material, the fabrication of ceramic features mainly consists of two processes: micromolding and sintering, which are the problems that this thesis focuses on. In the micromolding process, the loose nanoparticles are packed to form features with specific shapes. In the sintering process, the nanoparticles in as-prepared features are bonded into a coherent and dense feature. For the micromolding process, a suspension made from the nanoparticles is poured into a mold with as-designed feature shape, and the dry feature is obtained after a drying process. In this study, the factors that will affect the shape of the features are studied. It is found that the major factors include completeness of the filling process and shrinkage during the drying process. By completing the micromolding process in a centrifugal machine, the micromolding process is accelerated, and the shrinkage during the drying process is decreased. Both the two aspects will benefit the feature quality. By using this technique, zinc oxide ridges from 0.4 μm to 2 μm size and rods of 1.6 μm size are fabricated successfully. It is also demonstrated that this technique can be applied to other ceramic materials. Sintering process can convert packed nanoparticles into a coherent object, which can help us to obtain dense ceramic features. However, the sintering process will cause the change in feature shape. For large size ceramic bulks, the sintering theory has been well established to explain these changes. When the size of ceramic materials decreases to very small scale, such as micron size, new sintering theory is needed to explain the change of ceramic features in the sintering process. In this study, micron-sized zinc oxide ridges and rods were sintered at 950oC for different time. It was found that the sintering process will distort the shape of the zinc oxide features. Based on thermodynamic views, the corresponding new theory was established. Because zinc oxide is relatively easy to evaporate during sintering, sintering atmosphere will also affect the shape of the features. In this study, micron-sized zinc oxide ridge features were sintered under air and argon atmospheres. It was found that oxygen content was the major factor that will affect the shape change. The corresponding theory was established to explain the effect of the sintering atmosphere based on thermodynamic views.

El aprendizaje autom�atico es un �area de investigaci�on que proporciona algoritmos y t�ecnicas que son capaces de aprender autom�aticamente a partir de experiencias pasadas. Estas t�ecnicas son esenciales en el �area de descubrimiento de conocimiento de bases de datos (KDD), cuya fase principal es t�ÿpicamente conocida como miner�ÿa de datos. El proceso de KDD se puede ver como el aprendizaje de un modelo a partir de datos anteriores (generaci�on del modelo) y la aplicaci�on de este modelo a nuevos datos (utilizaci�on del modelo). La fase de utilizaci�on del modelo es muy importante, porque los usuarios y, muy especialmente, las organizaciones toman las decisiones dependiendo del resultado de los modelos. Por lo general, cada modelo se aprende de forma independiente, intentando obtener el mejor resultado (local). Sin embargo, cuando varios modelos se usan conjuntamente, algunos de ellos pueden depender los unos de los otros (por ejemplo, las salidas de un modelo pueden ser las entradas de otro) y aparecen restricciones. En este escenario, la mejor decisi�on local para cada problema tratado individualmente podr�ÿa no dar el mejor resultado global, o el resultado obtenido podr�ÿa no ser v�alido si no cumple las restricciones del problema. El �area de administraci�on de la relaci�on con los clientes (CRM) ha dado origen a problemas reales donde la miner�ÿa de datos y la optimizaci�on (global) deben ser usadas conjuntamente. Por ejemplo, los problemas de prescripci�on de productos tratan de distinguir u ordenar los productos que ser�an ofrecidos a cada cliente (o sim�etricamente, elegir los clientes a los que se les deber�ÿa de ofrecer los productos). Estas �areas (KDD, CRM) carecen de herramientas para tener una visi�on m�as completa de los problemas y una mejor integraci�on de los modelos de acuerdo a sus interdependencias y las restricciones globales y locales. La aplicaci�on cl�asica de miner�ÿa de datos a problemas de prescripci�on de productos, por lo general, ha / Bella Sanjuán, A. (2012). Model Integration in Data Mining: From Local to Global Decisions [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/16964

Yes / Purpose - This study aims to explore: the actual usage of GAS features among Egyptian external auditors, through the technology acceptance model (TAM); how the conceptual complexity of GAS features impact its actual usage; and what factors influencing the GAS use by Egyptian external auditors. Design/methodology/approach - External audit professionals at twelve international audit firms, including the Big 4 and eight medium-sized firms, in Egypt were surveyed. Findings - The results show that the basic features including database queries, ratio analysis, and audit sampling were higher in GAS use, perceived usefulness, and perceived ease of use among Egyptian external auditors than the advanced features: digital analysis, regression/ANOVA, and data mining classification. The SEM analysis by GAS features suggests that perceived ease of use has a stronger effect on GAS use when the conceptual complexity of GAS features is high. The analysis also support that the use of GAS by Egyptian external auditors is more affected by co-worker, supervisor, or organization through perceived usefulness, but not by job relevance, output quality, and result demonstration. Research limitations/implications - Although Egyptian external auditors participated in this study may limit the extent to which the findings may be generalized, the responses provide an insight into the actual usage of GAS features by external auditors and the impact of conceptual complexity of GAS features, which is consistent with the literature concerning the relatively low level of utilizing the advanced features of GAS by internal auditors, suggesting that the issues revealed should be of concern. Practical implications - The results reported in this paper are useful to audit software developers and audit firms in their understanding of factors influencing GAS usage in a different audit context. Originality/value - The study adds value to prior research by providing context-contingent insight into the application of technology acceptance model in an unexplored audit context.

No / Searching for and mapping the physical extent of unmarked graves using geophysical techniques has proven difficult in many cases. The success of individual geophysical techniques for detecting graves depends on a site-by-site basis. Significantly, detection of graves often results from measured contrasts that are linked to the background soils rather than the type of archaeological feature associated with the grave. It is evident that investigation of buried remains should be considered within a 3D space as the variation in burial environment can be extremely varied through the grave. Within this paper, we demonstrate the need for a multi-method survey strategy to investigate unmarked graves, as applied at a "planned" but unmarked pauper's cemetery. The outcome from this case study provides new insights into the strategy that is required at such sites. Perhaps the most significant conclusion is that unmarked graves are best understood in terms of characterization rather than identification. In this paper, we argue for a methodological approach that, while following the current trends to use multiple techniques, is fundamentally dependent on a structured approach to the analysis of the data. The ramifications of this case study illustrate the necessity of an integrated strategy to provide a more holistic understanding of unmarked graves that may help aid in management of these unseen but important aspects of our heritage. It is concluded that the search for graves is still a current debate and one that will be solved by methodological rather than technique-based arguments.

Yes / In recent years, automatic facial age estimation has gained popularity due to its numerous applications. Much work has been done on frontal images and lately, minimal estimation errors have been achieved on most of the benchmark databases. However, in reality, images obtained in unconstrained environments are not always frontal. For instance, when conducting a demographic study or crowd analysis, one may get profile images of the face. To the best of our knowledge, no attempt has been made to estimate ages from the side-view of face images. Here we exploit this by using a pre-trained deep residual neural network (ResNet) to extract features. We then utilize a sparse partial least squares regression approach to estimate ages. Despite having less information as compared to frontal images, our results show that the extracted deep features achieve a promising performance.

The purpose of this project was to expand the applications of time series prediction and action recognition for use with motion capture data and football plays. Both the motion capture data and football play trajectories were represented in the form of multidimensional time series. Each point of interest on the human body or football players path, was represented in two or three time series, one for each dimension of motion recorded in the data. By formulating a phase space reconstruction of the data, the remainder of each input time series was predicted utilizing kernel regression. This process was applied to the first portion of a play. Utilizing features from the theory of chaotic systems and specialized geometric features, the specific type of motion for the motion capture data or the type of play for the football data was identified by using the features with various classifiers. The chaotic features used included the maximum Lyapunov exponent, the correlation integral, and the correlation dimension. The variance and mean were also utilized in conjunction with the chaotic features. The geometric features used were the minimum, maximum, mean, and median of the x, y, and z axis time series, as well as various angles and measures of the trajectory as a whole. The accuracy of the features and classifiers was compared and combinations of features were analyzed. The novelty of this work lies in the method to recognize types of actions from a prediction made from only a short, initial portion of an action utilizing various sets of features and classifiers.

Electronic learning (e-learning) is an increasingly popular approach used to 1rain. knowledge and skills in work organizations. A 2006 market report estimated that organizations increased the use of e-learning by 8% from 2005 to 2006. Scientific research on how to best design e-learning to maximize learner outcomes is needed. Some attention is being directed at computer-based games due to their perceived motivational properties. By using a computer-based game and a non-game control group, this study examined differences in learning and satisfaction with training, and it also examined individual differences in goal orientation and self-efficacy. None of the hypotheses were supported. Instead, participants in the control group spent significantly more time in training and scored higher on a post-training performance goal orientation measure than participants in the game condition Moreover, when time spent in training was controlled participants in the control group performed significantly better on the post-training performance measure than did participants in the game condition. In addition consistent with research on individual differences, learning goal orientation (but not performance goal orientation) and self-efficacy were positively related to GPA. The results of this study suggest that more attention needs to be paid to the design of games used in training and on understanding how specific game features impact learning, motivation, and satisfaction.

Although digital media has been exploited to improve digital libraries, social networking sites, and book promotion for adult and child stakeholders, but encouraging children who have the choice to either read from a book or on a screen remains limited worldwide, including Jordan. This interest has meant that data about children¿s reading habits were needed, and the present study was intended as a contribution towards this aim. Interviews were conducted with Jordanian writers, publishers, child specialists, and various children¿s cultural centres. The managers and personnel unanimously showed that Jordanian children are not good readers and that a limited number of books are published for children as there are actual boundaries preventing Jordanian writers from publishing books. In particular, subjecting the typical sorts of children¿s websites ¿ 'Club Penguin', 'PBS Kids', 'A Story before Bed', 'Baraem', 'Storyline Online', and 'Raneen' ¿ to evaluation showed that 'Club Penguin' got the highest rank among the other websites in terms of multimodal features, usability, and language, while 'PBS Kids' got the highest rank regarding interactivity, and 'A Story before Bed' got the highest rank in reading activities. Although it was realised that most children were satisfied with the aspects of usability and ease of use rather than the structure or the aesthetic of the website, and were more attracted to the websites that provide multimodal features such as special characters, narration, gesture, and interactivity. The targeted websites¿ parameters obtained from the survey were used as guidance in the design structure of the KITABAK website, as a virtual reading environment for children¿s reading practices. The evaluation results that were obtained showed that there is a significant correlation towards encouraging children¿s reading habits and reading from printed books accompanying the website; girls showed more interest in reading iv than boys; and there is an obvious willingness for the adaptation of the website as a part of the Jordanian school curriculum. In addition, the KITABAK website was accepted significantly more than 'Club Penguin', mainly because the KITABAK website has facilities, games and reading activities. Also, results showed that children who were subjected to testing the KITABAK website for a one-week period proved to accept the website significantly more than those who were subjected to testing it once. / Applied Science University