As we gain increased understanding of genetic disorders and gene regulation more focus has turned towards complex interactions. Combinations of genes or gene and environmental factors have been suggested to explain the missing heritability behind complex diseases. Furthermore, gene activation and splicing seem to be governed by a complex machinery of histone modification (HM), transcription factor (TF), and DNA sequence signals. This thesis aimed to apply and develop multivariate machine learning methods for use on such biological problems. Monte Carlo feature selection was combined with rule-based classification to identify interactions between HMs and to study the interplay of factors with importance for asthma and allergy. Firstly, publicly available ChIP-seq data (Paper I) for 38 HMs was studied. We trained a classifier for predicting exon inclusion levels based on the HMs signals. We identified HMs important for splicing and illustrated that splicing could be predicted from the HM patterns. Next, we applied a similar methodology on data from two large birth cohorts describing asthma and allergy in children (Paper II). We identified genetic and environmental factors with importance for allergic diseases which confirmed earlier results and found candidate gene-gene and gene-environment interactions. In order to interpret and present the classifiers we developed Ciruvis, a web-based tool for network visualization of classification rules (Paper III). We applied Ciruvis on classifiers trained on both simulated and real data and compared our tool to another methodology for interaction detection using classification. Finally, we continued the earlier study on epigenetics by analyzing HM and TF signals in genes with or without evidence of bidirectional transcription (Paper IV). We identified several HMs and TFs with different signals between unidirectional and bidirectional genes. Among these, the CTCF TF was shown to have a well-positioned peak 60-80 bp upstream of the transcription start site in unidirectional genes.

Situation recognition is a process with the goal of identifying a priori defined situations in a flow of data and information. The purpose is to aid decision makers with focusing on relevant information by filtering out situations of interest. This is an increasingly important and non trivial problem to solve since the amount of information in various decision making situations constantly grow. Situation recognition thus addresses the information gap, i.e. the problem of finding the correct information at the correct time. Interesting situations may also evolve over time and they may consist of multiple participating objects and their actions. This makes the problem even more complex to solve. This thesis explores situation recognition and provides a conceptualization and a definition of the problem, which allow for situations of partial temporal definition to be described. The thesis then focuses on investigating how Petri nets can be used for recognising situations. Existing Petri net based approaches for recognition have some limitations when it comes to fulfilling requirements that can be put on solutions to the situation recognition problem. An extended Petri net based technique that addresses these limitations is therefore introduced. It is shown that this technique can be as efficient as a rule based techniques using the Rete algorithm with extensions for explicitly representing temporal constraints. Such techniques are known to be efficient; hence, the Petri net based technique is efficient too. The thesis also looks at the problem of learning Petri net situation templates using genetic algorithms. Results points towards complex dynamic genome representations as being more suited for learning complex concepts, since these allow for promising solutions to be found more quickly compared with classical bit string based representations. In conclusion, the extended Petri net based technique is argued to offer a viable approach for situation recognition since it: (1) can achieve good recognition performance, (2) is efficient with respect to time, (3) allows for manually constructed situation templates to be improved and (4) can be used with real world data to find real world situations. / <p>Anders Dahlbom is also affiliated to Skövde Artificial Intelligence Lab (SAIL), Information Fusion Research Program, Högskolan i Skövde</p>

This thesis draws on techniques from Management Science and Artificial Intelligence to explore organisational aspects of asset management in electricity transmission enterprises. In this research, factors that influence policies and practices of asset management within electricity transmission enterprises have been identified, in order to examine their interaction and how they impact the policies, practices and performance of transmission businesses. It has been found that, while there is extensive literature on the economics of transmission regulation and pricing, there is little published research linking the engineering and financial aspects of transmission asset management at a management policy level. To remedy this situation, this investigation has drawn on a wide range of literature, together with expert interviews and personal knowledge of the electricity industry, to construct a conceptual model of asset management with broad applicability across transmission enterprises in different parts of the world. A concise representation of the model has been formulated using a Causal Loop Diagram (CLD). To investigate the interactions between factors of influence it is necessary to implement the model and validate it against known outcomes. However, because of the nature of the data (a mix of numeric and non-numeric data, imprecise, incomplete and often approximate) and complexity and imprecision in the definition of relationships between elements, this problem is intractable to modelling by traditional engineering methodologies. The solution has been to utilise techniques from other disciplines. Two implementations have been explored: a multi-level fuzzy rule-based model and a system dynamics model; they offer different but complementary insights into transmission asset management. Each model shows potential for use by transmission businesses for strategic-level decision support. The research demonstrates the key impact of routine maintenance effectiveness on the condition and performance of transmission system assets. However, performance of the transmission network, is not only related to equipment performance, but is a function of system design and operational aspects, such as loading and load factor. Type and supportiveness of regulation, together with the objectives and corporate culture of the transmission organisation also play roles in promoting various strategies for asset management. The cumulative effect of all these drivers is to produce differences in asset management policies and practices, discernable between individual companies and at a regional level, where similar conditions have applied historically and today.

This PhD by project is an examination of process-based architectural design. It offers an examination of one approach to undertaking process-based experimentation in architecture - based on reflection of my own practice and body of work - through which I have been able to consider a complex array of questions and issues that are associated with working in this way. By

Thesis (M.Sc. (Hons.)) -- University of Western Sydney, 2001. / "A thesis submitted for the degree of Master of Science (Honours) - Computing and Information Technology at University of Western Sydney" Bibliography : leaves 93-98.

Univ., Diss.--Bern, 2004.

This research project aims to develop a framework to manage uncertainty in cloud manufacturing for small and medium enterprises (SMEs). The framework includes a cloud manufacturing taxonomy; guidance to deal with uncertainty in cloud manufacturing, by providing a process to identify uncertainties; a detailed step-by-step approach to managing the uncertainties; a list of uncertainties; and response strategies to security and privacy uncertainties in cloud manufacturing. Additionally, an online assessment tool has been developed to implement the uncertainty management framework into a real life context. To fulfil the aim and objectives of the research, a comprehensive literature review was performed in order to understand the research aspects. Next, an uncertainty management technique was applied to identify, assess, and control uncertainties in cloud manufacturing. Two well-known approaches were used in the evaluation of the uncertainties in this research: Simple Multi-Attribute Rating Technique (SMART) to prioritise uncertainties; and a fuzzy rule-based system to quantify security and privacy uncertainties. Finally, the framework was embedded into an online assessment tool and validated through expert opinion and case studies. Results from this research are useful for both academia and industry in understanding aspects of cloud manufacturing. The main contribution is a framework that offers new insights for decisions makers on how to deal with uncertainty at adoption and implementation stages of cloud manufacturing. The research also introduced a novel cloud manufacturing taxonomy, a list of uncertainty factors, an assessment process to prioritise uncertainties and quantify security and privacy related uncertainties, and a knowledge base for providing recommendations and solutions.

Made available in DSpace on 2015-03-03T15:47:46Z (GMT). No. of bitstreams: 1 AndersonPC_DISSERT.pdf: 1402040 bytes, checksum: 960b15bc1392a94fb7ba8ba980e3a0b4 (MD5) Previous issue date: 2012-12-20 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / Atualmente, h? diferentes defini??es de implica??es fuzzy aceitas na literatura. Do ponto de vista te?rico, esta falta de consenso demonstra que h? discord?ncias sobre o real significado de "implica??o l?gica" nos contextos Booleano e fuzzy. Do ponto de vista pr?tico, isso gera d?vidas a respeito de quais "operadores de implica??o" os engenheiros de software devem considerar para implementar um Sistema Baseado em Regras Fuzzy (SBRF). Uma escolha ruim destes operadores pode implicar em SBRF's com menor acur?cia e menos apropriados aos seus dom?nios de aplica??o. Uma forma de contornar esta situa??o e conhecer melhor os conectivos l?gicos fuzzy. Para isso se faz necess?rio saber quais propriedades tais conectivos podem satisfazer. Portanto, a m de corroborar com o significado de implica??o fuzzy e corroborar com a implementa??o de SBRF's mais apropriados, v?rias leis Booleanas t?m sido generalizadas e estudadas como equa??es ou inequa??es nas l?gicas fuzzy. Tais generaliza??es s?o chamadas de leis Boolean-like e elas n?o s?o comumente v?lidas em qualquer sem?ntica fuzzy. Neste cen?rio, esta disserta??o apresenta uma investiga??o sobre as condi??es suficientes e necess?rias nas quais tr?s leis Booleanlike ?like ? y ? I(x, y), I(x, I(y, x)) = 1 e I(x, I(y, z)) = I(I(x, y), I(x, z)) ?? se mant?m v?lidas no contexto fuzzy, considerando seis classes de implica??es fuzzy e implica??es geradas por automorfismos. Al?m disso, ainda no intuito de implementar SBRF's mais apropriados, propomos uma extens?o para os mesmos / There are more than one acceptable fuzzy implication definitions in the current literature dealing with this subject. From a theoretical point of view, this fact demonstrates a lack of consensus regarding logical implication meanings in Boolean and fuzzy contexts. From a practical point of view, this raises questions about the implication operators" that software engineers must consider to implement a Fuzzy Rule Based System (FRBS). A poor choice of these operators generates less appropriate FRBSs with respect to1 their application domain. In order to have a better understanding of logical connectives, it is necessary to know the properties that they can satisfy. Therefore, aiming to corroborate with fuzzy implication meaning and contribute to implementing more appropriate FRBSs to their domain, several Boolean laws have been generalized and studied as equations or inequations in fuzzy logics. Those generalizations are called Booleanlike laws and a lot of them do not remain valid in any fuzzy semantics. Within this context, this dissertation presents the investigation of sucient and necessary conditions under which three Boolean-like laws | y I(x; y), I(x; I(y; x)) = 1 and I(x; I(y; z)) = I(I(x; y); I(x; z)) | hold for six known classes of fuzzy implications and for implications generated by automorphisms. Moreover, an extension to FRBSs is proposed

Detta arbete visar, delvis var för sig och delvis i kombination, hur regelbaserat och fallbaserat beslutsstöd kan användas vid uppdragsplanering. Uppdragsplanering utförs med hjälp av Mission Support System (MSS). Uppdragsplanering kräver en hel del arbete och en hel del erfarenhet för att den ska bli bra. Detta kan underlättas om man kan få hjälp av olika verktyg som kan dra nytta av uppsatta regler för respektive uppdrag och även dra nytta av tidigare uppdrag av samma karaktär. Sedan tidigare finns det två olika examensarbete som har undersökt respektive del av detta. Målet med detta arbete är delvis att demonstrera metoderna i en prototyp, var för sig och i kombination med varandra, och delvis att försöka besvara frågan om en kombination av metoderna presterar ett bättre beslutsstöd än när metoderna används var för sig. Detta arbete bygger på två tidigare examensarbeten. Metoden som används för att kunna bedöma vilken metod som är att föredra är att man har implementerat både verktygen i en prototyp. I prototypen planerar man sedan ett antal uppdrag och applicerar sedan dem olika metoderna var för sig och även i kombination och utvärderar resultatet. Resultatet pekar på att det ur planeringssynpunkt bör användas en kombination av de två presenterade metoderna. Däremot om man tar in tidsaspekten, så är den erfarenhetsbaserade metoden inte att rekommendera i det utförande som den är i nu. Detta eftersom den tar lång tid att applicera. Tidsåtgången är uppemot 12 timmar. Vilket inte fungerar i verkligheten. / This thesis shows, partly alone, partly in combination, how rule based and case based decision support can be used in mission planning. For mission planning Mission Support System (MSS) is used. Mission planning requires a lot of effort and experience to make a good plan. This can be facilitated if there are tools that can benefit from rules for actual or previously missions of the same character. Thera are two theses that have investigated these different aspects. The goal with this thesis is to partly demonstrate these methods in a prototype, alone and in combination, and partly try to answer the question if a combination of the methods is performing a better decision support than each of them alone. This thesis is based on two previously thesis. The method that is used to to be able to assess which method is preferable is to implement the both tools in a prototype. The prototype is then used to plan a few missions and applying the different methods alone and in combination and evaluate the result. The results indicate that from a planning point, a combination of the two methods should be used. However if one takes the time in consideration, then the case based method is not to recomend in its current status. This because the execution time is long. The execution time is up to 12 hours. Which does not work in reality.

In modern times continuous casting is the preferred way to convert molten steel into solid forms to enable further processing. At Columbus Stainless the continuous casting machine cast slabs of constant thickness with varying width. One important aspect of the continuously cast strand that must be controlled, is the strand width. The strand width exiting from the casting machine, has a direct influence on the product yield which in turn influences the profitability of the company. In general, the strand width control on the austentic and ferritic type steels achieved is excellent with the exception of the 12% chrome non stabilised ferritic steel. This steel type exhibited different strand width changes when a sequence of different heats was cast. The strand width changes corresponded to the different heats in the sequence. Each heat has a unique chemistry and a relationship between the austenite and ferrite fraction at high temperature and the resulting strand width change was explained by Siyasiya[27]. The relationship between the heat composition and width change has in the past resulted in the development of a model that enabled the prediction of the expected width change of a specific heat before it is cast to enable preventative action to be taken. This model has been implemented as an on-line prediction model in the production environment with very encouraging results. This study was initiated because it was uncertain if the implemented model was the most accurate for this application. This study is concerned with the development of more models based on different techniques in an attempt to implement a more accurate model. The data mining techniques used include statistical regression, decision trees and fuzzy logic. The results indicated that the existing model was the most accurate and it could not be improved upon. / Dissertation (MEng (Mechanical Engineering))--University of Pretoria, 2007. / Mechanical and Aeronautical Engineering / MEng / unrestricted