Mud Lake is a wildlife refuge located in southeastern Idaho just north of Bear Lake that traps sediment from Bear River water flowing into Bear Lake.Very few water quality and sediment observations, if any, exist spatially in Mud Lake. Spatial patterns of sediment deposition may affect Mud Lake flows and habitat; prediction of those patterns should help refuge managers predict water quality constituents and spatial distribution of fine sediment.This will help sustain the purposes of Mud Lake as a habitat and migratory station for species. The main objective of the research is the development of Multivariate Relevant Vector Machine (MVRVM) to predict suspended fine sediment and water quality constituents, and to provide an understanding for the practical problem of determining the amount of data required for the MVRVM. MVRVM isa statistical learning algorithm that is based on Bayes theory.It has been widely used to predict patterns in hydrological systems and other fields. This research represents the first known attempt to use a MVRVM approach to predict transport of very fine sediment andwater quality constituents in a complex natural system. The results demonstrate the ability of the MVRVM to capture and predict the underlying patterns in data.Also careful construction of the experimental design for data collection can lead the Relevant Vectors (RVs is a subset of training observation which carries significant information that is used for prediction) to show locations of significant patterns. The predictions of water quality constituents will be of potential value to US Fish and Wildlife refuge managers in making decisions for operation and management in the case of Mud Lake based on their objectives, and will lead the way for scientists to expand the use of the MVRVM for modeling of suspended fine sediment and water quality in complex natural systems.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第17927号 / 情博第509号 / 新制||情||90(附属図書館) / 30747 / 京都大学大学院情報学研究科社会情報学専攻 / (主査)教授 吉川 正俊, 教授 石田 亨, 教授 田中 克己 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DGAM

京都大学 / 0048 / 新制・課程博士 / 博士(経済学) / 甲第21291号 / 経博第579号 / 新制||経||286(附属図書館) / 京都大学大学院経済学研究科経済学専攻 / (主査)教授 徳賀 芳弘, 准教授 草野 真樹, 教授 藤井 秀樹 / 学位規則第4条第1項該当 / Doctor of Economics / Kyoto University / DFAM

Morality has been a subject of study for centuries, though there is still much that is not understood about the factors that dictate moral decision making and moral identity. This study examines the relationship between the Big Five Personality traits (agreeableness, openness, neuroticism, conscientiousness, and extraversion) as well as empathy and moral decision making. However, we use a slightly different methodology than previous studies on morality and personality by incorporating more relevant moral dilemmas related to the ongoing COVID-19 pandemic in addition to the purely hypothetical dilemmas used in previous research (referred to as standard dilemmas). Additionally, rather than considering only the response to the dilemma as variable, this study also explored the relationship between these personality traits and the guilt felt by the decision made as well as how difficult participants felt it was for them to make the decision. The results suggest that there is a relationship between certain aspects of personality and the guilt felt by moral decisions as well as how difficult one finds making a decision. Additionally, the results imply that the relevance of the dilemma does significantly impact moral dilemma decisions and the feeling associated with such decisions.

The National Library of Medicine (NLM), an institute in the National Institutes of Health (NIH), maintains a collection of 17,000 digitized spine X-ray images obtained from the second National Health and Nutrition Examination Survey (NHANES II). Research effort has been devoted to develop a web-accessible retrieval system that allows retrieval of images from the NHANES II database on relevant and frequently found pathologies. A comprehensive and successful image retrieval system requires effective image representation and matching methods, relevance feedback algorithms to incorporate user opinions, and efficient indexing schemes for fast access to image databases. This dissertation studies and develops approaches for all of the above areas within the context of content-Based Image Retrieval (CBIR) of spine X-ray images from the NHANES II collection. Shape is an important characteristic for describing pertinent pathologies in various types of medical images, including spine X-ray images. Retrieving images with shapes similar to a specific user query can be useful for finding pathologies exhibited in images in large survey collections. In this work, vertebral outlines are extracted for image retrieval using shape matching methods to detect the presence of anterior osteophytes. The Multiple Open Triangle (MOT) shape representation method is proposed for partial shape matching (PSM), and a Corner-Guided Dynamic Programming (DP) strategy is developed to search partial intervals for matching comparison based on a 9-point model marked by a board-certified radiologist. The MOT method demonstrates higher retrieval accuracy compared to other approaches and the retrieval speed is improved significantly through the use of Corner-Guided DP. Computer-calculated low-level image features fall short when imitating high-level human visual perception. Relevance Feedback (RF) attempts to bridge the gap between the two by analyzing and employing user feedback. The need for overcoming this gap is more evident in medical image retrieval. Existing RF approaches are analyzed and a weight-updating formula for RF is developed. A hybrid retrieval approach is proposed that utilizes both CBIR with RF and RF history. This hybrid approach uses short-term memory to store the feedback history, which contributes to the retrieval results and helps select images for user feedback. An approximate 20% average increase in retrieval recall percentage is achieved within two RF iterations. Efficient indexing methods are desired for fast database access. An agglomerative clustering algorithm is adopted to pre-index the database based on pre-calculated pair-wise distances between indexed parts. Retrieval with this pre-indexing procedure is shown to offer faster retrieval and maintain a comparable recall percentage.

Community engagement and relevance are topics prominently discussed in the museum field. Conversations about public history and social justice, however, are less common. Combining these two ideas and thinking broadly about how museums, particularly historic houses, can stay relevant in their community by adopting a community-centered mission, this thesis uses Casey Farm as a case study. By conducting interviews with the site managers and market manager alongside surveying market vendors and visitors, this thesis compares the museum’s perspective of their relevance with the lived experiences of visitors. Ultimately arguing that historic houses should prioritize community interests when creating programming to retain audiences to the museum. Studying how Casey Farm partners with Coastal Growers’ Market to increase community relevance by aligning with the goals of the Food Movement, this example can encourage other historic house museums to use the resources at their site and seek out possible partnerships in their community.

Product Costing in the Swedish Engineering Industry - a survey  This thesis is about product costing in the Swedish engineering industry. The thesis is empirical and consists of two studies, one descriptive and one explanatory study. The purpose of the descriptive study has been to make a survey of which product costing methods are used, which elements of product costing that are considered to be problematic and what companies spend time on related to product costing. The purpose of the explanatory study has been to compare and analyze the correspondence between literature and practice regarding design aspects of product costing, by using a contingency theoretical approach. The research method that has been used to collect data has been an e-mail survey. In the 90s, a dissertation was written by Urban Ask and Christian Ax in the area, but since then no major survey has been made either in Sweden or in any other country. Since the 1990s, the production environment has been developed and new production strategies have been implemented, which has contributed to new perspectives on product costing. The results of the descriptive study show that the recommendations given in "Enhetliga principer för självkostnadsberäkningar" (EP) in 1936 still have a large impact on Swedish product costing practices. On the other hand, a trend can be noted which indicates that companies choose to work with new product costing systems such as target costing, activity based costing, value stream costing and kaizen costing. With regard to the allocation of indirect costs, the results show that the companies focus on ensuring that the allocation takes place in an appropriate manner. The outcome of the results shows that the distribution of indirect costs is differentiated, in the majority of cases a number of allocation bases, overhead components and cost centers are used. The product costing elements that are perceived to be most problematic to perform in practice are to: determine causal relationships, determine what deviations from standards are due to and calculate costs for future products. Overall the findings in the descriptive study indicate on a technically speaking well-developed Swedish costing practice. The explanatory study lends support to the hypotheses that the design of cost systems depends on contingency factors. Strongest support for the hypotheses was found between two factors (production strategies and company size) and cost system aspects. Weaker support was found between six other factors (manufacturing complexity, number of products, kind of product, the degree of automation, high- vs low-tech companies and the degree of competition) and cost system aspects. Overall the findings in the explanatory study indicate that companies do not shape their product costing according to the recommendations made in the literature.

There is scarce research that implement a formal framework when evaluating usability evaluation methods such as heuristicevaluation. This paper aimed to explore and compare the results of a heuristic evaluation performed by end-users and a heuristicevaluation performed by experts. Both heuristic evaluations took place in the context of forestry industry where a mobileapplication developed to give harvest operators performance feedback was evaluated. A thorough literature review for researchregarding evaluation of UEM was a crucial first step. The outcome of this produced an evaluation framework that included threecriteria, Relevance, Frequency and Timeliness. These criteria were used to analyse the results from the heuristics evaluationsperformed by the two groups, using mixed methods. The quantitative analysis concluded that the evaluation performed by theend-users had a higher frequency and relevance value, and that the evaluation performed by the expert group had higher valuefor their solution rate in the timeliness criteria. Furthermore, the qualitative analysis held within the criteria timeliness concludedthat the two groups identified different types of usability problems, confirming previous research performed on different types ofheuristic evaluators.

Neural networks are regarded as state of the art within many areas of machine learning, however due to their growing complexity and size, a question regarding their trustability and understandability has been raised. Thus, neural networks are often being considered a "black-box". This has lead to the emersion of evaluation methods trying to decipher these complex networks. Two of these methods, layer-wise relevance propagation (LRP) and sensitivity analysis (SA), are used to generate heatmaps, which presents pixels in the input image that have an impact on the classification. In this report, the aim is to do a usability-analysis by evaluating and comparing these methods to see how they can be used in order to understand a particular classification. The method used in this report is to iteratively distort image regions that were highlighted as important by the two heatmapping-methods. This lead to the findings that distorting essential features of an image according to the LRP heatmaps lead to a decrease in classification score, while distorting inessential features of an image according to the combination of SA and LRP heatmaps lead to an increase in classification score. The results corresponded well with the theory of the heatmapping-methods and lead to the conclusion that a combination of the two evaluation methods is advocated for, to fully understand a particular classification. / Neurala nätverk betraktas som den senaste tekniken i många områden inom maskininlärning, dock har deras pålitlighet och förståelse ifrågasatts på grund av deras växande komplexitet och storlek. Således, blir neurala nätverk ofta sedda som en "svart låda". Detta har lett till utvecklingen  av evalueringsmetoder som ämnar att tolka dessa komplexa nätverk. Två av dessa metoder, layer-wise relevance propagation (LRP) och sensitivity analysis (SA), används för att generera färgdiagram som visar pixlar i indata-bilden som har en påverkan på klassificeringen. I den här rapporten, är målet att göra en användarbarhets-analys genom att utvärdera och jämföra dessa metoder för att se hur de kan användas för att förstå en specifik klassificering. Metoden som används i denna rapport är att iterativt förvränga bilder genom att följa de två färgdiagrams-metoderna. Detta ledde till insikterna att förvrängning av väsentliga delar av bilden, vilket framgick ur LRP färgdiagrammen, tydligt minskade sannolikheten för klassen. Det framkom även att förvrängning av oväsentliga delar, som framgick genom att kombinera SA och LRP färgdiagrammen, ökade sannolikheten för klassen. Resultaten stämde väl överens med teorin och detta ledde till slutsatsen att en kombination av metoderna rekommenderas för att förstå en specifik klassificering.

The objective of this thesis was to investigate intersectional (categorical/social) invisibility and the extent to which this phenomenon occurs in a comparison of dominant (i.e., Caucasian), and non-dominant (African-American and Latino) social/ethnic groups. It has been found that intersectional invisibility occurs among African-American women with respect to Caucasian men and women, and African American men (Sesko & Biernat, 2010), but little of this research has been done regarding Latinas. Thus, this experiment aims to not only examine whether Latinas are also subject to intersectional invisibility among dominant (i.e., Caucasian) and non-dominant (i.e., African American and/or Latino) groups, but to determine whether the theory can be extended to perceptions between non-dominant groups such as African-Americans and Latinos. Determining whether intersectional invisibility occurs among Latinas, moreover, may provide theoretical and practical insights of what advantages/disadvantages Latinas may particularly endure as members of the rapidly growing Latino population in the U.S.