Predicting sales is an important step for many business processes. Several forecasting methods have been applied to uncountable different problems, however with no present research found in the area of renewable energy certificates. Thus, this study aims to examine the possibility of developing a model based on traded volumes of certificates, where a comparison between simpler and more complex models explores the general increased interest in machine learning models. Therefore, five different models are tested with monthly sales data: the statistical model Seasonal Autoregressive Integrated Moving Average, the machine learning models Support Vector Regression and Extreme Gradient Boosting and further the neural networks Long Short-Term Memory and Bidirectional Long Short-Term Memory. Extensive data preparation is operated by taking into account seasonality and trends where data transformations are applied in addition to feature engineering. To evaluate the models, non-aggregated monthly forecasts as well as aggregated predictions of two and three months are examined. The results show that it is feasible to model the sales volumes of renewable energy certificates. As expected, the models generally perform better when evaluated based on aggregated monthly predictions. Also, when considering both evaluation strategies, the Seasonal Autoregressive Integrated Moving Average, Support Vector Regression and Extreme Gradient Boosting are the only models showing better performance compared to a baseline model. The proposed solution to enable smarter and more efficient trading decisions today is a combination of the aggregated two months and quarterly predictions of the Seasonal Autoregressive Integrated Moving Average and Support Vector Regression models. Considering an expected expansion of relevant available data for the company, the recommendation for the future is to specifically further develop the machine learning models with an anticipation of improved performance and valuable feature importance insights.

In representation learning we are interested in how data is represented by different models. Representations from different models are often compared by training a new model on a downstream task using the representations and testing their performance. However, this method is not always applicable and it gives limited insight into the representations. In this thesis, we compare natural image representations from classification models and the generative model BigGAN using two other approaches. The first approach compares the geometric clustering of the representations and the second approach compares if the pairwise similarity between images is similar between different models. All models are large pre-trained models trained on ImageNet and the representations are taken as middle layers of the neural networks. A variety of experiments are performed using these approaches. One of the main results of this thesis shows that the representations of different classes are geometrically separated in all models. The experiments also show that there is no significant geometric difference between representations from training data and representations from validation data. Additionally, it was found that the similarity of representations between different models was approximately the same between the classification models AlexNet and ResNet as well as between the classification models and the BigGAN generator. They were also approximately equally similar to each other as they were to the class embedding of the BigGAN generator. Along with the experiment results, this thesis also provide several suggestions for future work in representation learning since a large number of research questions were explored. / Detta verk studerar representationer från artificiella neuronnät. Representationerna tas som värdena på ett lager i mittendelen av neuronnätet. Eftersom dessa representationer har flera olika användningsområden är syftet att jämföra dem från olika modeller. Ofta jämförs representationer genom att testa hur bra de är som input till en ny modell med ett nytt mål; alltså hur bra representationerna är att använda inom “transfer learning”. Denna metod ger begränsad information om representationerna och är inte alltid applicerbar. Detta verk använder därför två andra tillvägagångssätt för att jämföra representationer. Den första är att jämföra geometriska grupperingar hos olika representationer. Den andra använder ett mått av hur lika olika representationer är. Flera olika experiment utförs med hjälp av dessa tillvägagångssätt. Representationerna kommer frånmodeller som redan tränats på ImageNet. Både klassifikationsmodeller och en generativa modell används med syfte att också jämföra dem med varandra. Det första huvudresultatet från experimenten är att det finns en tydlig geometrisk separation av representationer från olika klasser i modellerna. Experimenten visar också att det inte fanns en tydlig geometrisk separation av representationer från träningsdata och valideringsdata. Ett annat resultat är att representationerna från de olika klassifikationsmodellerna AlexNet och ResNet är ungefär lika lika varandra som mellan klassifikationsmodellerna och generatorn hos den generativa modellen BigGAN. Resultaten visar också att de har en liknande likhet till BigGANs “class embedding”. Fler forskningsfrågor undersöks i andra experiment. Utöver experimenten kommer detta verk med många idéer till framtida forskning.

When a company evaluates a customer for being a potential prospect, one of the key questions to answer is whether the customer will generate profit in the long run. A possible step to answer this question is to predict the likelihood of the customer returning to the company again after the initial purchase. The aim of this master thesis is to investigate the possibility of using machine learning techniques to predict the likelihood of a new customer returning for a second purchase within a certain time frame. To investigate to what degree machine learning techniques can be used to predict probability of return, a number of di↵erent model setups of Logistic Lasso, Support Vector Machine and Extreme Gradient Boosting are tested. Model development is performed to ensure well-calibrated probability predictions and to possibly overcome the diculty followed from an imbalanced ratio of returning and non-returning customers. Throughout the thesis work, a number of actions are taken in order to account for data protection. One such action is to add noise to the response feature, ensuring that the true fraction of returning and non-returning customers cannot be derived. To further guarantee data protection, axes values of evaluation plots are removed and evaluation metrics are scaled. Nevertheless, it is perfectly possible to select the superior model out of all investigated models. The results obtained show that the best performing model is a Platt calibrated Extreme Gradient Boosting model, which has much higher performance than the other models with regards to considered evaluation metrics, while also providing predicted probabilities of high quality. Further, the results indicate that the setups investigated to account for imbalanced data do not improve model performance. The main con- clusion is that it is possible to obtain probability predictions of high quality for new customers returning to a company for a second purchase within a certain time frame, using machine learning techniques. This provides a powerful tool for a company when evaluating potential prospects.

Organizations with multiple operating requirements require support functions to assist in execution of strategic goals. This effort, in turn, requires management of engineering activities in control of projects and in sustaining facilities. High level strategies include employing engineering support that consists of a project management function encompassing technical and managerial disciplines. The architecture/engineering, and construction office (AEC) is the subject of this research. Engineering and construction oriented organizations have experienced challenges to their abilities to learn and grow. This has potential detrimental implications for these organizations if support functions cannot keep pace with changing objectives and strategy. The competitive nature and low industry margins as well as uniqueness of projects as challenges facing engineering and construction. The differentiated nature of projects tasks also creates a need for temporary and dedicated modes of operation and thereby tends to promote highly dispersed management practices that do not dovetail very well with other organizational processes. Organizational learning is a means to enhance and support knowledge management for improving performance. The problem addressed through this research is the gap between desired and achieved individual and group learning by members of the AEC, and the members' abilities to distinguish between the need for adaptive learning or innovation. This research addresses learning by individuals and groups, and the strategies employed through an empirical study (survey). A conceptual model for organizational learning contributions by individuals and groups is presented and tested for confirmation of exploitive or explorative learning strategies for individuals, and directions composed of depth and breadth of learning. Strategies for groups are tested for internal or external search orientations and directions toward the single or multi-discipline unit. The survey is analyzed by method of principal components extraction and further interpreted to reveal factors that are correlated by Pearson product moment coefficients and tested for significance for potential relationships to factors for outcomes. Correlation across dependent variables prevented interpretation of the most significant factors for group learning strategies. However, results provide possible support for direction in supporting processes that promote networking among individuals and group structures that recognize the dual nature of knowledge - that required for technical competency and that required for success in the organization. Recommendations for practitioners include adjustments to knowledge acquisition direction, promoting external collaboration among firms, and provision of dual succession pathways through technical expertise or organizational processes for senior staff.

Subscription-based services are becoming more popular in today’s society. Therefore, any company that engages in the subscription-based business needs to understand the user behavior and minimize the number of users canceling their subscription, i.e. minimize churn. According to marketing metrics, the probability of selling to an existing user is markedly higher than selling to a brand new user. Nonetheless, it is of great importance that more focus is directed towards preventing users from leaving the service, in other words preventing user churn. To be able to prevent user churn the company needs to identify the users in the risk zone of churning. Therefore, this thesis project will treat this as a classification problem. The objective of the thesis project was to develop a statistical model to predict churn for a subscription-based service. Various statistical methods were used in order to identify patterns in user behavior using activity and engagement data including variables describing recency, frequency, and volume. The best performing statistical model for predicting churn was achieved by the Random Forest algorithm. The selected model is able to separate the two classes of churning users and the non-churning users with 73% probability and has a fairly low missclassification rate of 35%. The results show that it is possible to predict user churn using statistical models. Although, there are indications that it is difficult for the model to generalize a specific behavioral pattern for user churn. This is understandable since human behavior is hard to predict. The results show that variables describing how frequent the user is interacting with the service are explaining the most whether a user is likely to churn or not. / Prenumerationstjänster blir alltmer populära i dagens samhälle. Därför är det viktigt för ett företag med en prenumerationsbaserad verksamhet att ha en god förståelse för sina användares beteendemönster på tjänsten, samt att de minskar antalet användare som avslutar sin prenumeration. Enligt marknads-föringsstatistik är sannolikheten att sälja till en redan existerande användare betydligt högre än att sälja till en helt ny. Av den anledningen, är det viktigt att ett stort fokus riktas mot att förebygga att användare lämnar tjänsten. För att förebygga att användare lämnar tjänsten måste företaget identifiera vilka användare som är i riskzonen att lämna. Därför har detta examensarbete behandlats som ett klassifikations problem. Syftet med arbetet var att utveckla en statistisk modell för att förutspå vilka användare som sannolikt kommer att lämna prenumerationstjänsten inom nästa månad. Olika statistiska metoder har prövats för att identifiera användares beteendemönster i aktivitet- och engagemangsdata, data som inkluderar variabler som beskriver senaste interaktion, frekvens och volym. Bäst prestanda för att förutspå om en användare kommer att lämna tjänsten gavs av Random Forest algoritmen. Den valda modellen kan separera de två klasserna av användare som lämnar tjänsten och de användare som stannar med 73% sannolikhet och har en relativt låg missfrekvens på 35%. Resultatet av arbetet visar att det går att förutspå vilka användare som befinner sig i riskzonen för att lämna tjänsten med hjälp av statistiska modeller, även om det är svårt för modellen att generalisera ett specifikt beteendemönster för de olika grupperna. Detta är dock förståeligt då det är mänskligt beteende som modellen försöker att förutspå. Resultatet av arbetet pekar mot att variabler som beskriver frekvensen av användandet av tjänsten beskriver mer om en användare är påväg att lämna tjänsten än variabler som beskriver användarens aktivitet i volym.

M. Tech. Business Information Systems / A Virtual Learning Environment (VLE) enables teaching and learning pedagogy that allows students to study without geographical barriers and time constraints. VLE promotes innovations in institutions of higher learning and encourages lecturers and students to move away from the face-to-face learning method to virtual learning pedagogy. The ability to learn anywhere, any time (which is what VLE is all about) was found to be the most appealing. Furthermore, VLE enables institutions of higher learning to enrol large numbers of students without having to worry about the size of the classroom. Global learning, as well as collaboration between leaners and lecturers is encouraged and supported through VLEs. Although institutions of higher learning spend huge amounts of money on technologies such as VLE, most VLEs are still not contextualized to cater for the needs of students in developing countries such as South Africa. This leads to the VLEs in developing countries not being effectively utilised. This study aims at designing a contextualized VL model that suits South African institutions of higher learning. The study identifies factors necessary for contextualizing VLE to fit the student's perspective in a developing country (in this case, South Africa).

Made available in DSpace on 2016-04-29T14:23:12Z (GMT). No. of bitstreams: 1 Deyse Cristiani da Silva.pdf: 462376 bytes, checksum: 42e31059316d9a3a06e3c6a18116adef (MD5) Previous issue date: 2012-10-18 / The development of this work, we intended to raise an overview of distance education, considering its historical aspects, its specificities, legislation and the conditions of its implementation in our country, especially in the Pontifical Catholic universities in Brazil. It is conceived education as forming the human being, beyond the prospect of training human resources in order to not only contemplate solutions to meet the material needs of the current population, but contributes to human emancipation as a prerequisite to citizenship. Distance education breaks the relationship with space / time, it has excelled in higher education institutions, and is realized through mediated communication, the media and presented in the model of distance education presented in this text. The point of the research was to investigate the distance learning of the Pontifical Catholic University of Brazil, based on the model of distance education presented in Chapter 3, as well as verification that the models applied in educational institutions are faithful to the views and humanistic characteristics that proposes to make in their Institutional Development Projects. The methodology uses the method of document analysis, descriptive qualitative research. The visions, missions and goals defined by each institution of higher education are analyzed in this research. The applicability exerted by each, and what was explained in Institutional Development Projects is actually exercised especially in Distance Education. The authors of this work are that based Ricardo Rossato, José Manuel Moran, Marcos Formiga and Marco Silva. The first chapter explains the history of the University: emergence, consolidation and updates, in the second chapter, the concepts are structured, stories, laws and models of distance education and, finally, in the last chapter, the visions are explained, and features applicability of Distance Education at the Pontifical Catholic universities in Brazil. The conclusion of the study aimed to differentiate the visions and conceptions according to Institutional Development Project of PUCs in Brazil, based on the model of Distance Education foregoing assessing whether distance learning is being applied targeting the human and not the number of students / O desenvolvimento deste trabalho, pretendeu-se levantar uma panorama sobre a educação à distância, considerando seus aspectos históricos, suas especificidades, legislação e as condições de sua implementação em nosso país, especialmente nas Pontifícias Universidades Católicas no Brasil. Concebe-se a educação como formadora do ser humano, para além da perspectiva de capacitação de recursos humanos, de modo à contemplar não apenas soluções para atender as atuais necessidades materiais da população, mas que contribua para a emancipação humana como um dos requisitos para o exercício da cidadania. A educação a distância rompe com a relação espaço/tempo, que tem se destacado nas instituições de ensino superior, e se concretiza por intermédio da comunicação mediada, pela mídia e apresentada no modelo de Educação a Distância apresentado nesse texto. A questão da pesquisa consisti uma investigação do ensino a distância das Pontifícias Universidades Católicas do Brasil, partindo do modelo de Educação a Distância apresentado no Capitulo 3, bem como na verificação se os modelos aplicados nas instituições de ensino são fiéis às visões e características humanísticas que propõem realizar em seus Projetos de Desenvolvimento Institucional. A metodologia aplicada utiliza o método de análise documental, de pesquisa qualitativa descritiva. As visões, missões e objetivos definidos por cada Instituição de ensino Superior são analisadas nessa pesquisa. A aplicabilidade exercida por cada uma, e se o que foi explicitado nos Projetos de Desenvolvimento Institucionais é de fato exercido em especial na Educação a Distância. Os autores que embasaram este trabalho são Ricardo Rossato, José Manuel Moran, Marcos Formiga e Marco Silva. O primeiro capitulo explicita a História da Universidade: surgimento, consolidação e as atualidades, no segundo capitulo, estão estruturados os conceitos, histórias, legislação e modelos da Educação a Distância e, por fim, no ultimo capitulo, são explicitadas as visões, características e aplicabilidade da Educação a Distância nas Pontifícias Universidades Católicas no Brasil. A conclusão do trabalho visou diferenciar as visões e concepções de acordo com Projeto de Desenvolvimento Institucional das PUCs no Brasil, partindo do modelo de Educação a Distância exposto anteriormente, avaliando se o ensino a distância está sendo aplicado visando o humano e não a quantidade de alunos

The process of designing a floorplan is highly iterative and requires extensive human labor. Currently, there are a number of computer programs that aid humans in floorplan design. These programs, however, are limited in their inability to fully automate the creative process. Such automation would allow a professional to quickly generate many possible floorplan solutions, greatly expediting the process. However, automating this creative process is very difficult because of the many implicit and explicit rules a model must learn in order create viable floorplans. In this paper, we propose a method of floorplan generation using two machine learning models: a sequential model that generates rooms within the floorplan, and a graph-based model that finds adjacencies between generated rooms. Each of these models can be altered such that they are each capable of producing a floorplan independently; however, we find that the combination of these models outperforms each of its pieces, as well as a statistic-based approach.

No description available.

Problem: Today's knowledge society needs companies, personnel and employees to constantly work with both competence and knowledge development. E-learning is a learning method that can be used in businesses in order to develop and change the work process. E-learning has been shown to lead to cost efficiency, delivery efficiency, free study pace, personal responsibility and savings in the education process. As more and more companies use e-learning, it is essential to investigate the effects of e-learning on the business and its competence development. Purpose: The purpose of the study is to investigate the effects that e-learning has in a business and on staff who invest and use e-learning platforms as a pedagogical training method. The study will also examine the expectations and goals that companies set in connection with the investment in their e-learning. The study will further describe the effects achieved and the non-achieved effects that have arisen in connection with e-learning compared to previous educational methods. Method: This dissertation is based on a qualitative case study where semi-structured interviews are conducted with about ten workers from Postnord who have in some way been a part of constructing their application “Postnord Learning” and its use. These interviews should contribute a subjective perspective to the study. The qualitative method is applied in order to give the participants the opportunity to answer the questions without restrictions. The study is based on an inductive approach with the purpose of applying the interviews as the primary focus and the search for other theories as a secondary focus in connection with data presented in previous studies. Conclusion: The effects that e-learning brings to a business are cost and time efficiency, improved competence development and the ability to monitor the levels of commitment to e-learning platforms. The negative effects arising from e-learning are technical barriers, which means that the necessary units must be adjusted for e-learning to function and add operational value. Other effects are decoupling from the platform, lack of motivation for skills development and limited social interactions. / Problem: Dagens kunskapssamhälle kräver att företag, personal och anställda ständigt arbetar med både kompetens och kunskapsutveckling. E-lärande är en lärandemetod som kan användas i verksamheten i syfte att förbättra och förändra arbetsprocessen. E-lärandet har visat sig kunna leda till kostnadseffektivitet, leveranseffektivitet, fri studietakt, eget ansvar och besparingar i utbildningsprocessen. Eftersom att allt fler verksamheter tillämpar e-lärandet är det essentiellt att undersöka vilka effekter e-lärandet har på verksamheten och dess kompetensutveckling.  Syfte: Syftet med avhandlingen är att undersöka vilka effekter som e-lärandet tillför i en verksamhet och på personal som investerar och nyttjar e-lärandeplattformar som en pedagogisk utbildningsmetod. Studien kommer även att undersöka de förväntningar och målsom verksamheter sätter i samband med investeringen i sitt e-lärande. Studien kommer vidare skildra de uppnådda effekterna respektive de ej uppnådda effekterna som uppstått i samband med e-lärande jämfört med tidigare utbildningsmetoder.  Metod: Denna avhandling bygger på en kvalitativ fallstudie där semistrukturerade intervjuer utförs om tio arbetare från Postnord som på något sätt har varit en del av att konstruera sin applikation ”Postnord Learning” och dess användning. Dessa intervjuer ska bidra med ett subjektivt perspektiv till studien. Den kvalitativa metoden tillämpas i syfte att ge deltagarna möjlighet att besvara frågorna utan begränsningar. Studien baseras på ett induktivt tillvägagångssätt med ända mål att tillämpa intervjuerna som primärfokus samt sökandet efter andra teorier som sekundärt fokus i samband med data som presenterats i tidigare studier.  Slutsats: Effekterna som e-lärandet tillför i en verksamhet är kostnad- och tidseffektivitet, förbättrad kompetensutveckling och förmågan att övervaka nivåerna av engagemang före-lärande plattformar. Den negativa effekten som härrör från e-lärande är tekniska barriärer, vilket innebär att de nödvändiga enheterna måste justeras för att e-lärande ska fungera och tillföra verksamhetsvärde. Andra effekter är frikoppling av plattformen, bristande motivation för kompetensutveckling och begränsad sociala interaktioner.