This work focuses on automated classification of railway ties based on their condition using aerial imagery. Four approaches are explored and compared to achieve this goal - handcrafted features, HOG features, transfer learning and proposed CNN architecture. Mean test accuracy per class and Quadratic Weighted Kappa score are used as performance metrics, particularly suited for the ordered classification in this work. Transfer learning approach outperforms the handcrafted features and HOG features by a significant margin. The proposed CNN architecture caters to the unique nature of the railway tie images and their defects. The performance of this approach is superior to the handcrafted and HOG features. It also shows a significant reduction in the number of parameters as compared to the transfer learning approach. Data augmentation boosts the performance of all approaches. The problem of label noise is also analyzed. The techniques proposed in this work will help in reducing the time, cost and dependency on experts involved in traditional railway tie inspections and will facilitate efficient documentation and planning for maintenance of railway ties. / Master of Science

Forty-three Pueblo III sandals and sandal fragments from four sites in Tsegi Canyon, Arizona, were analyzed to develop tie terminology, identify possible correlation of tie system to sandal construction, and identify characteristics for future tie research. Each piece of sandal tie cordage was analyzed for tie system type, tie attachment method, and tie cordage structure. Three overall tie systems were found: toe-heel tie systems (nine sandals), criss-cross tie system (one sandal), and side loop tie systems (fifteen sandals) Five additional sandals (narrow, two-warp plainweave construction) had possible toe-heel tie systems. Tie systems of six sandals were not distinguishable between toe-heel and criss-cross. Five sandals had unidentifiable tie fragments, and two sandals had no tie information. Terminology had to be developed for several tie components. Toe loops were found with a single loop (either intact or tied over the toe) and multiple loops (overlapping, adjacent, or parallel). Side loops were overlapping (continuous, mirror image, or reversing), adjacent, and parallel. Ankle/heel loops included single, double, and double intertwined horizontal loops (back of heel), and two vertical loops (side of heel). Loop variations for two-warp plainweave were through weft, around weft, flat loop, and upright loop. Numerous attachment methods were used for the toe, ankle/heel, and side loops, including sole knots and running stitches. Tie materials included four constructions of leaf: unprocessed leaf, S-twist leaf, Z-twist leaf, and plied leaf (Z-2S, S-2Z, Z-3S). Twelve constructions of yarn were identified, including two- and three-ply yarns of various twist combinations, re-plied yarn, and splicing. Toe-heel and criss-cross tie system sandals tended to have only one cordage type used on an individual sandal; side loop system sandals tended to have several cordage types used on an individual sandal. A comparison of tie system and sandal construction methods revealed that coarse-braided and two-warp plainweave sandals more frequently used the toe-heel system. Fine-braided and four-and five-warp plainweave sandals used the toe-heel and side loop systems nearly evenly. Markers for future tie researchers to examine include number of toe loops, types and attachment methods of toe and side loops, and style variations and attachments of ankle/heel loops.

No description available.

The purpose of this study is: 1) to explore the ways in which the Internet may affect individuals’ political diversity in different strengths of social relationships; 2) to identify which of strong-tie diversity and weak-tie diversity contributes more to political tolerance; and 3) to investigate the extent to which tolerant people are different from the less tolerant in their participation decisions when exposed to political diversity. In order to examine the contribution of Internet news use to political heterogeneity, the current study examines the moderation of the negative influence of politically selective exposure on the Internet on political diversity in social networks by Internet news use. To identify the better contributor to tolerance, the two diversities are compared. To assess the consequence of exposure to political difference for political participation for tolerant and less tolerant people, the present study examines any moderating effect of tolerance between political network heterogeneity and participation. It also observes the moderating effect in different tie strengths. This study utilizes data obtained from the U.S. Citizenship, Involvement, Democracy (CID) survey conducted by a collaboration of Center for Democracy and Civil Society at Georgetown University and the European Social Survey. The sample is 1,001 adults aged 18 and over and representative of the contiguous United States. The dataset contains items concerning Internet use, informal social networks, the composition and diversity of ties and associations, democratic values and tolerance under the primary themes of democracy, social capital and civic engagement. The data are analyzed by hierarchical and OLS regression. According to the findings, Internet news use contributes to individuals’ overall political diversity by reducing the negative influence of the selective exposure occurring from online interaction with homogeneous people. When examined in different strengths of interpersonal relationships, selective exposure discourages strong-tie diversity while encouraging weak-tie diversity. Internet news use positively affects strong-tie diversity but had no influence on weak-tie diversity. Weak-tie diversity is found to be a better contributor to political tolerance. Politically tolerant individuals tend to be discouraged for political participation when exposed to difference in their social relationships. Therefore, while political tolerance may increase overall political diversity, it may as well threaten the balance between deliberation and participation. Closer interpersonal associations are not found to reduce the demobilizing effect of exposure to difference for tolerant individuals. / text

La mise en oeuvre exacte et précise du repère international de référence terrestre (ITRF) est une exigence fondamentale pour le développement des Sciences du Système Terre. La réalisation du référentiel mondial, en fait, concerne directement de nombreux domaines allant de la détermination précise des orbites des satellites, à la calibration des altimètres, à l'évaluation des étalonnages absolus d'antennes satellites pour le Global Navigation Satellite System (GNSS) et la validation des corrections du vecteur du centre de masse pour les véhicules spatiaux portant à bord des rétro-réflecteurs pour la technique de télémétrie laser sur satellite (SLR). En conséquence, toutes les études portant sur les mouvements de la surface de la Terre, y compris les océans et les calottes glaciaires, dépendent étroitement de la disponibilité d'un repère de référence fiable qui est fondamental pour référencer les mesures pertinentes. La réalisation de l'ITRF doit alors être périodiquement mise à jour, afin d'intégrer des nouvelles observations et progrès dans les procédures d'analyse des données et/ou des méthodes de combinaison. Toutes les nouvelles stratégies de calcul doivent viser l'amélioration de la réalisation des paramètres physiques du repère, à savoir l'origine et l'échelle, sur lesquels se fondent de façon critique un grand nombre d'études scientifiques et d'applications civiles. Ce travail se concentre sur le potentiel de combiner les observations GNSS et SLR par leur liens à bord de satellites GPS / GLONASS. En fait, les satellites GNSS équipés de rétro-réflecteurs peuvent être observés par les stations SLR, ce qui permet de déterminer les orbites des satellites à travers les deux signaux : optiques et à micro-ondes. En principe, la connexion inter-technique si réalisée pourrait être exploitée pour le calcul de l'ITRF en place des liens terrestres actuellement utilisés. Ces derniers sont connus pour être aujourd'hui un facteur limitant de la précision du repère en raison de leur distribution inhomogène et de leurs divergences avec les estimations de la géodésie spatiale en conséquence des erreurs systématiques dans les observations. Dans cette étude, la force du lien alternatif en orbite a été soigneusement analysée afin d'évaluer les performances de l'approche de combinaison sélectionnée dans les conditions opérationnelles disponibles. L'investigation porte sur la caractérisation de la précision, de la fiabilité et de la pertinence des paramètres combinés du repère de référence. / The accurate and precise implementation of the International Terrestrial Reference Frame (ITRF) is a fundamental requirement for the development of Earth System Sciences. The actual realization of the reference frame, in fact, directly impacts a number of different tasks ranging from precise satellite orbit determination to altimeter calibration, satellite antenna offset assessment for Global Navigation Satellite System (GNSS) and validation of center of mass corrections for spacecrafts carrying on board retro-reflectors for Satellite Laser Ranging (SLR). As a consequence, all the studies investigating motions of the Earth’s surface, including oceans and ice-sheets, strictly depend on the availability of a reliable TRF that is fundamental for geo-referencing the relevant measurements. ITRF realizations must then be periodically updated, in order to account for newly acquired observations and for upgrades in data analysis procedures and/or combination methods. Any innovative computation strategy should ameliorate the realization of the frame physical parameters, namely the origin and the scale, upon which a number of scientific applications critically rely. This work addresses the potential of combining GNSS and SLR observations via their co-location on board GPS/GLONASS satellites. GNSS vehicles equipped with retro-reflector arrays can be tracked by SLR ground stations, which allows determining the spacecraft orbits by means of both optical and microwave signals. In principle, the inter-technique connection so achieved could be exploited for the computation of the ITRF in place of terrestrial ties. These lasts are known to be currently a limiting factor of the frame accuracy because of their inhomogeneous distribution and of their discrepancies with space geodesy estimates due to technique systematic errors. In this study, the strength of the alternative link in orbit has been thoroughly investigated in order to evaluate the performances of the selected space tie approach under the available operational conditions. The analysis focuses on the characterization of the precision, the accuracy and the pertinence of the combined frame parameters.

The role of social networking has widely been embraced in an entrepreneurial context due to its nature of providing easy access to information, support and other complementary resources needed for the creation and development of entrepreneurial venture (Omta et al., 2001; Johannisson,1990 cited in Song et al., 2017). The entrepreneurs’ network is affiliated with the study of social relations that can influence the creation and development of new business by shedding light on the functionality of social ties in the diffusion of resources that are vital for the establishment of firms (Greve, 1995). Therefore, studying the field of entrepreneurship in the context of social network can offer a fruitful perspective on entrepreneurship (Greve, 1995).Despite the vast and diverse research in the field of networking and entrepreneurship, there are still some gaps concerning what is actually going on in a network and the understanding about network operation, their nature and role in influencing business performance still limited to fairly broader and theoretical perspective (Jack, 2005). So, in line with these views, the purpose of this study is to explore and enhance the understanding of the networking in an entrepreneurial context by shedding light on the role of strong and weak social ties. The foundation of this study relies on Granovetter’s (1973, 1985) strong and weak tie hypothesis and Jack (2005) work, that by highlighting the characteristics, formation, and outcome of strong and weak social relationships contribute to building the concept of entrepreneur’s social network.A multiple case study of five entrepreneurs, operating their networks and businesses in Linköping, Sweden, has been conducted to achieve the purpose of this study. This research work uses a qualitative approach and is based on the semi-structured interviews, which allows a deeper insight into the studied area through analysis of multiple cases.The result of this study indicates that both the strong and weak social ties, in which an entrepreneur is bonded, are used to a varying level in terms of information flow. Strong relationships, that relies on frequent interaction, hold great importance in providing quality information as compared to weak relationships but simultaneously accountable for providing information that is useless and not up-to date. On, the other hand weak-ties contacts are dispersed in a social network and rarely interact with each other. But these weak connections are also important and play their role once get activated by entrepreneurs. Considering the information flow weaker relationships also occurred to be useful however the relatively small size of the town can mean that similar information is repeated by several connections. Since, the aspect of individual knowledge and experience help in the formation of new relationships but the result of this study highlights that, a personal motivation and drive towards networking also plays a crucial role in the formation of both strong and weak ties.The study findings can benefit the researchers and the entrepreneurs in developing their knowledge about networking and role of social ties in providing access to information and resources necessary for the creation and the development of venture. However, the strong connections hold a dominant position concerning the information and the resource flow but weak connections are also crucial and provide support in areas that strong connections lack. Thus, the entrepreneur that keeps the balance between both his/her strong and weak connections can efficiently reap the benefit of social networking.

No description available.

The presence of both narcissism and social capital in the digital social networking site Facebook is altering the dynamic of relationship development. Therefore, the central research question directing this study is as follows: What role do narcissistic personality traits play in bridging social capital on Facebook? From this, two sub-questions are raised: 1) What are the motives for users to express narcissistic traits through Facebook? 2) How is narcissistic-motivated activity influencing social capital development through Facebook? Guiding this thesis are concepts of social tie relationships (as they apply to narcissism), and digital social networks. An exploratory study of in-depth interviews was utilized to conduct such research. A key objective in this thesis is understanding motive and purpose for generating social capital in an online environment. The findings from this study suggest Facebook is a facilitator for the expression of narcissistic traits. As a result, this is influencing the disconnected and questionable value of digital social capital.

The presence of both narcissism and social capital in the digital social networking site Facebook is altering the dynamic of relationship development. Therefore, the central research question directing this study is as follows: What role do narcissistic personality traits play in bridging social capital on Facebook? From this, two sub-questions are raised: 1) What are the motives for users to express narcissistic traits through Facebook? 2) How is narcissistic-motivated activity influencing social capital development through Facebook? Guiding this thesis are concepts of social tie relationships (as they apply to narcissism), and digital social networks. An exploratory study of in-depth interviews was utilized to conduct such research. A key objective in this thesis is understanding motive and purpose for generating social capital in an online environment. The findings from this study suggest Facebook is a facilitator for the expression of narcissistic traits. As a result, this is influencing the disconnected and questionable value of digital social capital.

The presence of both narcissism and social capital in the digital social networking site Facebook is altering the dynamic of relationship development. Therefore, the central research question directing this study is as follows: What role do narcissistic personality traits play in bridging social capital on Facebook? From this, two sub-questions are raised: 1) What are the motives for users to express narcissistic traits through Facebook? 2) How is narcissistic-motivated activity influencing social capital development through Facebook? Guiding this thesis are concepts of social tie relationships (as they apply to narcissism), and digital social networks. An exploratory study of in-depth interviews was utilized to conduct such research. A key objective in this thesis is understanding motive and purpose for generating social capital in an online environment. The findings from this study suggest Facebook is a facilitator for the expression of narcissistic traits. As a result, this is influencing the disconnected and questionable value of digital social capital.