This study compared the vocabularies created by various domain experts and the source documents selected by them to create the vocabulary. The results indicate that there is similarity among the vocabularies created and the source documents selected. Also, the relationship between the overlap scores of vocabularies created and overlap scores of source documents selected was tested and it was observed that no significant relation exists between them. In addition, the variability of the overlap scores of the vocabularies generated automatically to the variability of the overlap scores of those produced manually by domain experts was evaluated. The results suggested that these vocabularies are significantly different from each other. / Master of Science

Cheetah (Acinonyx jubatus) persist under continued conservation threat in small populations mostly in protected areas in an historically reduced geographic range. Home range, a useful trait for threat assessment, species reintroduction, and population estimation, is plastic in cheetah with sizes ranging from 40 km2 to over 1000 km2 depending on location. Previous home range estimates for cheetah used the minimum convex polygon (MCP), assuming asymptotic home ranges and MCP insensitivity to sample size. They reported metrics of home range size and overlap based on only outline methods. I use 6 481 observations of 240 female and 315 male cheetah from > 60 matrilines over 25 years in the Serengeti Plains to investigate lifetime, core, yearly, and seasonal range size with kernel density estimation. I investigate autocorrelation using time to statistical independence of locations. I confront the assumption of asymptotic home ranges by testing the traditional and multiscaled home range predictions and provide a novel method for determining kernel asymptotes. I challenge the notion of Serengeti cheetah as a migratory carnivore with analyses of site fidelity and objectively defined core ranges. I assess year to year and seasonal location shifts, showing that yearly shifting lessens as females age. I provide quantitative evidence for philopatry in female- and juvenile dispersal in male cheetah of the Serengeti Plains. I use simple overlap metrics to show that overlap in lifetime and core ranges is greater in related than unrelated female pairs. I use multi-response permutation procedures (MRPP) to show that overlap in unrelated female pairs varies with season. I use correlation of utilization distributions to show that avoidance is apparent only in unrelated pairs of females. My results call into question previous MCP estimates of cheetah home range size, and provide guidance for future sampling of cheetah locations. My home range results will guide management of this imperiled species and my methodological findings may be general and applicable to a wide range of taxa. / Master of Science

No description available.

In this thesis, we address two major problems in Generative Adversarial Networks (GAN), an important sub-field in deep learning. The first problem that we address is the instability in the training process that happens in many real-world problems and the second problem that we address is the lack of a good evaluation metric for the performance of GAN algorithms. To understand and address the first problem, three approaches are developed. Namely, we introduce randomness to the training process; we investigate various normalization methods; most importantly we develop a better parameter initialization strategy to help stabilize training. In the randomness techniques part of the thesis, we developed two randomness approaches, namely the addition of gradient noise and the batch random flipping of the results from the discrimination section of a GAN. In the normalization part of the thesis, we compared the performances of the z-score transform, the min-max normalization, affine transformations and batch normalization. In the most novel and important part of this thesis, we developed techniques to initialize the GAN generator section with parameters that can produce a uniform distribution on the range of the training data. As far as we are aware, this seemingly simple idea has not yet appeared in the extant literature, and the empirical results we obtain on 2-dimensional synthetic data show marked improvement. As to better evaluation metrics, we demonstrate a simple yet effective way to evaluate the effectiveness of the generator using a novel "overlap loss".

In this thesis, we address two major problems in Generative Adversarial Networks (GAN), an important sub-field in deep learning. The first problem that we address is the instability in the training process that happens in many real-world problems and the second problem that we address is the lack of a good evaluation metric for the performance of GAN algorithms. To understand and address the first problem, three approaches are developed. Namely, we introduce randomness to the training process; we investigate various normalization methods; most importantly we develop a better parameter initialization strategy to help stabilize training. In the randomness techniques part of the thesis, we developed two randomness approaches, namely the addition of gradient noise and the batch random flipping of the results from the discrimination section of a GAN. In the normalization part of the thesis, we compared the performances of the z-score transform, the min-max normalization, affine transformations and batch normalization. In the most novel and important part of this thesis, we developed techniques to initialize the GAN generator section with parameters that can produce a uniform distribution on the range of the training data. As far as we are aware, this seemingly simple idea has not yet appeared in the extant literature, and the empirical results we obtain on 2-dimensional synthetic data show marked improvement. As to better evaluation metrics, we demonstrate a simple yet effective way to evaluate the effectiveness of the generator using a novel "overlap loss".

This study argues that a firm's product location choice may be a function of the firm's way of splitting the product market (i.e., divisionalization) and the concern for product cannibalization at the division and the firm levels. The focus of this study is at the division level and a division's new product location choice vis-à-vis its own products (intra-divisional new product distance), the products of a rival division of competing firms (inter-firm divisional new product distance), and the products of a sister division of the same firm (intra-firm divisional new product distance). The hypotheses were tested using data on the U.S. automobile industry between 1979 and 1999. The results show that a focal division with a high level of inter-firm divisional domain overlap with a rival division, relative to the focal division's own domain, is more likely to locate its new product (here new car model) closer to that rival's existing car models. And it was also found that divisional density affects a division's new product location choice. But this study didn't find any significant role of divisional status on new product location choice. And contrary to our expectation, the results of intra-firm divisional domain overlap and new product location choice suggest that inter-divisional product cannibalization might not be such an important concern when divisions introduce their new products, as we had originally expected. By addressing the firm's competitive engagement in the context of a division's new product location choice, this study expands the basic logic of market overlap at the firm level into the unit- or division-level, and highlights how a division's new product location choice is affected by intra-firm divisional structural relationship as well as interfirm divisional structural relationship. In so doing, this study hopes to contribute to the literature on divisionalization, new product location choice, competition at the unit-level, and product cannibalization, among others.

Despite being ubiquitous within the Classical style, pedals have received little attention in the scholarly literature. This dissertation adopts a form-functional approach to understanding them with an eye toward distinguishing their normative and non-normative functions. A form-functional approach begins by parsing the musical surface into formal and temporal functions. William Caplin has identified five distinct temporal functions: 1) "before-the-beginning"; 2) "the beginning"; 3) "the middle"; 4) "the end"; and 5) "after-the-end." Appearing mostly in locations 2) and 5), tonic pedals express two primary functions, formal initiation and postcadential closing for a theme. Dominant pedals also express postcadential function, but close less stable areas like transitions. They also appear in location 3), most notably in contrasting middles, where they bring harmonic contrast or uncertainty and anticipate a recapitulation. Because they frequently mediate between the home and subordinate key, I have devised four categories of dominant pedal that bring a sense of clarity to the range of functions dominant pedals can serve. V[supercript d] => V[supercript d] pedals normally express a postcadential standing-on-the-dominant or a contrasting-middle function. V[supercript t] => V[supercript t] pedals express the same functions, but in a more stable way. V[supercript t] => V[supercript d] pedals express a retransition function or a contrasting-middle function that enacts a harmonic transformation. The V[supercript d] => V[supercript t] pedal also expresses a contrasting-middle function, but reverses the direction of the harmonic transformation. Finally, with regard to their non-normative functions, pedals can loosen their formal contexts by generating form-functional overlap and expansion. / text

Σκοπός: 1) Η μελέτη της επίπτωσης της στυτικής δυσλειτουργίας σε ασθενείς με σύνδρομο επικάλυψης ( ΣΑΑΥ και ΧΑΠ ) και η συγκρισή της με την επίπτωση της στυτικής δυσλειτουργίας σε ασθενείς που παρουσίαζαν μόνο ΣΑΑΥ, και 2) η εκτίμηση της βελτίωσης της στυτικής δυσλειτουργίας (ED) σε άνδρες με σύνδρομο επικάλυψης (αποφρακτική άπνοια του ύπνου και χρόνια αποφρακτική πνευμονοπάθεια) υπό θεραπεία με cPAP (συνεχής θετική πίεση αεραγωγών) και βρογχοδιασταλτικά. Υλικό και μέθοδος : Μελετήθηκαν 48 άνδρες με μέσο όρο ηλικίας τα 52.8 + 10 χρόνια με αποφρακτική άπνοια του ύπνου (ΣΑΑΥ) και χρόνια αποφρακτική πνευμονοπάθεια (ΧΑΠ) και παράλληλα στυτική δυσλειτουργία ED. Επίσης 30 άνδρες με στυτική δυσλειτουργία και ΣΑΑΥ χωρίς αποφρακτική νόσο. Έλαβαν την κατάλληλη θεραπεία για αποφρακτική πνευμονοπάθεια (cPAP και βρογχοδιασταλτικά) για 6 μήνες και έπειτα το επίπεδο της στυτικής λειτουργίας (EF) επανεκτιμήθηκε. Η ED θεωρήθηκε ότι βελτιώθηκε εάν ο βαθμός βελτίωσης (score) αυξήθηκε για τουλάχιστον 5 μονάδες σε σύγκριση με αυτό της βασικής τιμής. Οι καθοριστικοί παράγοντες για βελτίωση της ED επίσης εκτιμήθηκαν, καθώς επίσης και ο βαθμός ικανοποίησης του κάθε ασθενή κατά τη διάρκεια της θεραπείας Αποτελέσματα : Η επίπτωση της ED σε ασθενείς με σύνδρομο επικάλυψης ήταν 81,08%. Η επίπτωση της ED σε ασθενείς πάσχοντες από ΣΑΑΥ και ED ήταν 29,17%. Η ED βελτιώθηκε σε 12 ασθενείς (25%), αλλά μόνο τα 2 /3 αυτών ήταν ικανοποιημένοι με τον βαθμό βελτίωσης μετά από θεραπεία. Η βελτίωση της ED συσχετίστηκε θετικά με την ηλικία, δείκτη άπνοιας / υπόπνοιας και αρνητικά με την διάρκεια της ED. Το score της ED, ο κορεσμός της αιμοσφαιρίνης σε Ο2 κατά τη διάρκεια της νύχτας και το BMI δεν συσχετίστηκαν σημαντικά με το βαθμό βελτίωσης της EF. Συμπεράσματα : 1) Η επίπτωση της ED σε ασθενείς με σύνδρομο επικάλυψης είναι μεγαλύτερη συγκριτικά με αυτή των ασθενών που έπασχαν μόνο από ΣΑΑΥ και συγκριτικά με το γενικό πληθυσμό. 2) Η συνήθης θεραπεία σε ασθενείς με ΣΑΑΥ και ΧΑΠ με βρογχοδιασταλτικά και cPAP έχει θετικά αποτελέσματα στην παράλληλη ED στην μειονότητα των ασθενών αυτών. Από τους άνδρες που βελτιώθηκαν, 1/3 από αυτούς δεν ήταν ικανοποιημένοι με το αποτέλεσμα αυτής της θεραπείας. Διαφορετική θεραπευτική αντιμετώπιση ή συνδυασμός αγωγής θα πρέπει πιθανά να χορηγηθεί στους ασθενείς αυτούς. / Aim: 1) To determine the prevalence of erectile dysfunction in patients with overlap syndrome (obstructive sleep apnea and chronic obstructive pulmonary disease) compared with patients suffered from sleep apnea syndrome and erectile dysfunction and 2)To assess the improvement of concurrent erectile dysfunction (ED) in these patients treated with continuous positive airway pressure (cPAP) and bronchodilators Material and Methods: We evaluated 48 men suffering from OSA, COPD and concurrent ED, and 30 men suffered from OSA and ED. They were treated with conventional therapy for pulmonary obstructive disease (cPAP and bronchodilators) for 6 months and then their erectile function status was reassessed. ED was considered as improved, if ED intensity score increased for at least five points compared to that of baseline. Results: The prevalence of ED in patients with overlap syndrome was 81,08%. The prevalence of ED in men with OSA was 29,17%. ED was improved in 12 patients (25%), but only two thirds of them were satisfied with the grade of improvement after treatment. ED improvement was related positively with age and apnea/hypopnea index and negatively with ED duration. ED intensity score, O2 saturation at night and BMI were not significantly related to the outcome of EF improvement. Conclusions: 1) The prevalence of ED in patients with overlap syndrome was higher compared with the prevalence of ED in men with OSA. 2) Conventional treatment for OSA and COPD, has a positive effect on concurrent ED on the minority of patients. The effect is possibly due to the improvement of respiration during sleep with cPAP and of oxygenation with bronchodilators continuously

A novel design for atmospheric laser radar (lidar) is presented, implementing a digital micromirror device (DMD) for use in (A) aligning transmitter and receiver boresight angles and in (B) field-of-view (FOV) control of such "DMD lidar" instruments. A novel technique is presented to extract the transmitter-receiver overlap-compensation function from ratioing data from different FOVs in the same pointing direction. DMD lidar design considerations and trades are surveyed. Principles of modeling DMD lidar performance are introduced and implemented in a performance-predictive system simulation with data-validated results. Operational capabilities of DMD lidar are demonstrated through a hardware prototype with field measurement examples. Additional capabilities offered by integrating DMD within lidar and other optical systems are presented, including single-pixel Radon-imaging techniques.

This study argues that a firm's product location choice may be a function of the firm's way of splitting the product market (i.e., divisionalization) and the concern for product cannibalization at the division and the firm levels. The focus of this study is at the division level and a division's new product location choice vis-à-vis its own products (intra-divisional new product distance), the products of a rival division of competing firms (inter-firm divisional new product distance), and the products of a sister division of the same firm (intra-firm divisional new product distance). The hypotheses were tested using data on the U.S. automobile industry between 1979 and 1999. The results show that a focal division with a high level of inter-firm divisional domain overlap with a rival division, relative to the focal division's own domain, is more likely to locate its new product (here new car model) closer to that rival's existing car models. And it was also found that divisional density affects a division's new product location choice. But this study didn't find any significant role of divisional status on new product location choice. And contrary to our expectation, the results of intra-firm divisional domain overlap and new product location choice suggest that inter-divisional product cannibalization might not be such an important concern when divisions introduce their new products, as we had originally expected. By addressing the firm's competitive engagement in the context of a division's new product location choice, this study expands the basic logic of market overlap at the firm level into the unit- or division-level, and highlights how a division's new product location choice is affected by intra-firm divisional structural relationship as well as interfirm divisional structural relationship. In so doing, this study hopes to contribute to the literature on divisionalization, new product location choice, competition at the unit-level, and product cannibalization, among others.