Master of Landscape Architecture / Department of Landscape Architecture/Regional and Community Planning / Anne E. Beamish / The purpose of this research was to develop a mobile sensory garden to help children with Sensory Integration Disorders. Sensory Integration (SI) is how an individual processes sensory information. Issues with sensory integration can lead to behavioral disorders, which can cause difficulties with social-emotional skills, gross motor skills, fine motor skills, play skills, and self-help skills. The goal of the project was to develop a mobile sensory garden to help alleviate or reduce the negative aspects of Sensory Integration Disorders. The research involved qualitative research, projective design, and evaluative research on a mobile sensory cart that was used in an elementary school in Topeka, Kansas. Methods included: interviews, design/build, and observation. The current sensory room was inventoried in Williams Science & Fine Arts Elementary Magnet School in Topeka, Kansas, and the children who used were observed. The faculty at the school, including the principal, teachers whose students use the current sensory room, counselors, and other support faculty, were interviewed. Once the observations were analyzed, a mobile sensory garden cart was designed and built. The new cart was brought to the elementary school and placed in the sensory room. A second phase of observation assessed how the children interacted with the new mobile sensory garden cart. During the second phase of observation the children showed intense interest in the mobile sensory garden cart. Some of this can be attributed to it being a new item in the sensory room. Some of the most popular items included the zen garden, wood slices on the wooden dowel, and the plants. In the second week children transitioned to using both the sensory garden cart and the original items in the sensory room. When an item on the sensory garden cart best suited their needs, that item was chosen and when their needs were best suited by other items in the room then that item was chosen. Overall the mobile sensory garden cart was beneficial to the school and the children. It can be easily replicated and modified to fit the individual needs of the children and school.

No description available.

There is a significant need to identify approaches for classifying chemical sensor array data with high success rates that would enhance sensor detection capabilities. The present study attempts to fill this need by investigating six machine learning methods to classify a dataset collected using a chemical sensor array: K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Classification and Regression Trees (CART), Random Forest (RF), Naïve Bayes Classifier (NB), and Principal Component Regression (PCR). A total of 10 predictors that are associated with the response from 10 sensor channels are used to train and test the classifiers. A training dataset of 4 classes containing 136 samples is used to build the classifiers, and a dataset of 4 classes with 56 samples is used for testing. The results generated with the six different methods are compared and discussed. The RF, CART, and KNN are found to have success rates greater than 90%, and to outperform the other methods.

In recent years, power demands around the world and particularly in North America increased rapidly due to increase in customer’s demand, while the development in transmission system is rather slow. This stresses the present transmission system and voltage stability becomes an important issue in this regard. Pattern recognition in conjunction with voltage stability analysis could be an effective tool to solve this problem In this thesis, a methodology to detect the voltage stability ahead of time is presented. Dynamic simulation software PSS/E is used to simulate voltage stable and unstable cases, these cases are used to train and test the pattern recognition algorithms. Statistical and algorithmic pattern recognition methods are used. The proposed method is tested on IEEE 39 bus system. Finally, the pattern recognition models to predict the voltage stability of the system are developed.

Submitted by Sara Ribeiro (sara.ribeiro@ucb.br) on 2017-08-16T20:20:25Z No. of bitstreams: 1 DanielLuizFiusaDissertacao2017.pdf: 1799911 bytes, checksum: d8636107eb168c749d4f0926a8352a2d (MD5) / Approved for entry into archive by Sara Ribeiro (sara.ribeiro@ucb.br) on 2017-08-16T20:20:32Z (GMT) No. of bitstreams: 1 DanielLuizFiusaDissertacao2017.pdf: 1799911 bytes, checksum: d8636107eb168c749d4f0926a8352a2d (MD5) / Made available in DSpace on 2017-08-16T20:20:32Z (GMT). No. of bitstreams: 1 DanielLuizFiusaDissertacao2017.pdf: 1799911 bytes, checksum: d8636107eb168c749d4f0926a8352a2d (MD5) Previous issue date: 2017-03-30 / The present paper uses random utility models from Perloff and Salop to analyze consumer behavior in the credit card market in Brazil, where choices are influenced by specific shocks in consumer preferences, "noises", and from its decisions, to determine the financial margin and the equilibrium price with the competition between two credit cards. The first model proposes that competition occurs with shocks in preferences linked to the amount of benefits that the cards offer to customers. In the second, issuers dispute consumer preference through shocks to the quality of these benefits. In this sense, the cards with the highest score in the relationship program or the discount percentage offered in the next purchases are preferred. Finally, the third model explores the relation of the interest rate of the revolving credit to the annuity price. In case the customer comes to use this facility, it will evaluate as best card what to offer the lowest rate of interest in the revolving credit. The results show the positive correlation of the quantity and quality of the benefits offered with the demand for the card and, consequently, the financial margin established for its issuer. On the other hand, the consumer is less likely to buy the card with the higher interest rate. / O presente trabalho utiliza de modelos de utilidade aleat??ria, com origem em Perloff e Salop, para analisar o comportamento do consumidor no mercado de cart??es de cr??dito no Brasil, no qual as escolhas s??o influenciadas por choques espec??ficos nas prefer??ncias dos consumidores, ???ru??dos???, e a partir das suas decis??es, apurar a margem financeira e o pre??o de equil??brio com a competi????o entre dois cart??es de cr??dito. O primeiro modelo prop??e que a competi????o ocorra com choques nas prefer??ncias vinculados ?? quantidade de benef??cios que os cart??es oferecem aos clientes. No segundo, os emissores disputam a prefer??ncia do consumidor por meio de choques na qualidade desses benef??cios. Nesse sentido, os cart??es com a maior pontua????o no programa de relacionamento ou do percentual de desconto oferecido nas pr??ximas compras s??o os preferidos. Por ??ltimo, o terceiro modelo explora a rela????o da taxa de juros do cr??dito rotativo com o pre??o da anuidade. No caso do cliente vir usar essa facilidade, ele ir?? avaliar como melhor cart??o o que oferecer a menor taxa de juros no cr??dito rotativo. Os resultados obtidos evidenciam a correla????o positiva da quantidade e qualidade dos benef??cios oferecidos com a demanda pelo cart??o e, consequentemente, na margem financeira apurada para o seu emissor. Por outro lado, o consumidor fica menos propenso a adquirir o cart??o com a maior taxa de juros.

Cette thèse s'inscrit dans le cadre de la statistique non paramétrique et porte sur la classification et la discrimination en grande dimension, et plus particulièrement sur la sélection de variables. Une première partie traite de la sélection de variables à travers CART, dans un cadre de régression et de classification binaire. La procédure exhaustive développée s'appuie sur le principe de la sélection de modèle qui permet d'obtenir des inégalités ``oracle'' et de réaliser la sélection de variables par contraste pénalisé. Une seconde partie est motivée par un problème industriel. Il s'agit de déterminer parmi les signaux temporels, mesurés au cours d'essais, ceux capables d'expliquer le ressenti de confort du conducteur, puis d'identifier les pages temporelles responsables de cette pertinence. La démarche adoptée s'articule autour du prétraitement des signaux, de la réduction de la dimension par projection dans une base d'ondelettes commune, et de la sélection de variables en mêlant CART et une stratégie pas à pas. Une dernière partie aborde le thème de la classification de données fonctionnelles au moyen des k-plus proches voisins. La procédure consiste à appliquer les k-plus proches voisins sur les coordonnées de la projection des données dans un espace fini dimensionnel. Cette procédure implique de déterminer simultanément la dimension de l'espace de projection et le nombre de voisins. La version usuelle des k-plus proches voisins et une version légèrement pénalisée sont considérées théoriquement. Un travail sur données réelles et simulées semble montrer qu'un faible terme de pénalité stabilise la sélection en conservant de bonnes performances.

The thesis project describes  the analysis, process and major factors in development of a simple component oriented Web Shop in ASP.NET. It addresses to the concepts used in  the application as  well  as  derivation  of  technical  design  and  development  from  the  concepts  acquired  by studying existing approaches. The report  describes  a  brief  summary of  existing  approaches  and related  technologies.  It also lays  the foundation  of  goal  oriented  approach  by  providing  an  overview  of  component  based  software engineering. The  basic  concepts for modularization  were  barrowed from  entities  identification,  object models and component models pplication’s architecture is set to be a layered app roach combining the software layered  architecture  approach  with  multi  tier  architecture  of  web  applications. Class models explaining the inner structure of each component have been provided and an overview of user interface pages is given to explain the application outer flow. The application sets out to prove the significance of component oriented approach as well as  the  support  provided for it by ASP.Net. The resulting package proves  to  have  scalable  components  that could be scaled for  or  reused in another  application or  in  a later version of the same application. / First and Final Version of our Thesis Report / SoftIn - Introducing methods and tools for software development in small and medium-sized enterprises

A very distinguished point of online marketing is that it can collect data about the consumers* shopping processes rather than the shopping results only. That is, it cannot only collect order data but also the browsing and shopping cart data. So far, the browsing records have been used to analyze the Web server traffic. However, regarding the analysis of shopping cart data, it has not been found in any marketing research yet. The purpose of this study is trying to verify the value of shopping cart data by examining whether it can improve the performance of the marketing communication decisions. According to Source-Message-Media-Receiver (SMMR) communication model, there are three important Web marketing communication decisions. These decisions are who are the target customers, what message should be communicated, and how to communicate. For each above marketing communication decision, in order to check whether the data from shopping cart can improve its performance, this research proposed an algorithm that integrates the shopping cart data into each decision process. Three hypotheses have been proposed in terms of the value of each new proposed algorithm. Three experiments have been implemented to test these hypotheses. The results reveal that the proposed algorithms can improve the performance of the marketing communications decisions. However, it is only a starting point to integrate the shopping cart data into the marketing research. As the online shopping becomes more popular, it is worthwhile to put more efforts to understand the details about the value of the online shopping cart data.

This goal of this project is to develop a set of business rules to mitigate risk related to a specific financial decision within the prepaid debit card industry. Under certain circumstances issuers of prepaid debit cards may need to decide if funds on hold can be released early for use by card holders prior to the final transaction settlement. After a brief introduction to the prepaid card industry and the financial risk associated with the early release of funds on hold, the paper presents the motivation to apply the CART (Classification and Regression Trees) method. The paper provides a tutorial of the CART algorithms formally developed by Breiman, Friedman, Olshen and Stone in the monograph Classification and Regression Trees (1984), as well as, a detailed explanation of the R programming code to implement the RPART function. (Therneau 2010) Special attention is given to parameter selection and the process of finding an optimal solution that balances complexity against predictive classification accuracy when measured against an independent data set through a cross validation process. Lastly, the paper presents an analysis of the financial risk mitigation based on the resulting business rules. / text

Stölder och snatteri är ett stort problem för fysiska butiker, dessutom marknadsutvecklingen för e-handel är avgörande för fysiska butikernas framgång och lönsamhet. Allt mer pengar investeras i säkerhetslösningar, väktare och personal Målet med den här kandidat-uppsatsen har varit att utveckla en produkt för ett modern och innovativt smart kundvagn. Den framtagna Smart kundvagnen är ett hjälpmedel som är bättre anpassad efter konsumenternas behov för att skapa trevligare upplevelse hos fysiska butiker och underlättar köpprocessen, vilket i längden kan leda till trygga och återkommande kunder. Genom att använda och utnyttja ny teknik tillsammans internetuppkoppling och nya funktioner till ett konkurrenskraftigt pris finns stora möjligheter att sälja denna produkt. Kundvagnen använder sig av bildanalys och lastcell som möjliggör att kunden inte längre behöver manuellt skanna en vara. istället identifieras varan och registreras när den placeras i kundvagnen. Prototypen har jämförts och slutsats har dragit att produkten är ett bättre alternativ än existerande produkter för ändamålet. / Thefts and shoplifting is a big problem for local stores, additionally the development of e-commerce is a deciding factor for the success of physical stores and their profitability. An increasing amount of money is being invested in security solutions, guards and other related personnel.  The goal of this Bachelors thesis has been to develop a product for a modern and innovative smart shopping cart. The resulting cart is a tool that is more suitable for the benefit of consumers and their need for a more comfortable shopping experience, it has the benefit of streamlining the shopping process which in turn could attract repeat customers.  Through the use of new technology as well as the Internet and new functions at a competitive price point there is a great opportunity to market this product. The shopping cart uses image analysis and a load cell which alleviates the need for customers to manually scan each article for purchase, instead it is identified and registered as it is placed in the shopping cart.  The prototype has been compared to other solutions and conclusions have been drawn as to the optimality of the product fulfill the purpose.