<p>This study is based on three models, Markov model, Hidden Markov model and the Radial basis function neural network. A number of work has been done before about application of these three models to the stock market. Though, individual researchers have developed their own techniques to design and test the Radial basis function neural network. This paper aims to show the different ways and precision of applying these three models to predict price processes of the stock market. By comparing the same group of data, authors get different results. Based on Markov model, authors find a tendency of stock market in future and, the Hidden Markov model behaves better in the financial market. When the fluctuation of the stock price index is not drastic, the Radial basis function neural network has a nice prediction.</p>

Graphical techniques for modeling the dependencies of randomvariables have been explored in a variety of different areas includingstatistics, statistical physics, artificial intelligence, speech recognition, image processing, and genetics.Formalisms for manipulating these models have been developedrelatively independently in these research communities. In this paper weexplore hidden Markov models (HMMs) and related structures within the general framework of probabilistic independencenetworks (PINs). The paper contains a self-contained review of the basic principles of PINs.It is shown that the well-known forward-backward (F-B) and Viterbialgorithms for HMMs are special cases of more general inference algorithms forarbitrary PINs. Furthermore, the existence of inference and estimationalgorithms for more general graphical models provides a set of analysistools for HMM practitioners who wish to explore a richer class of HMMstructures.Examples of relatively complex models to handle sensorfusion and coarticulationin speech recognitionare introduced and treated within the graphical model framework toillustrate the advantages of the general approach.

We present a framework for learning in hidden Markov models with distributed state representations. Within this framework, we derive a learning algorithm based on the Expectation--Maximization (EM) procedure for maximum likelihood estimation. Analogous to the standard Baum-Welch update rules, the M-step of our algorithm is exact and can be solved analytically. However, due to the combinatorial nature of the hidden state representation, the exact E-step is intractable. A simple and tractable mean field approximation is derived. Empirical results on a set of problems suggest that both the mean field approximation and Gibbs sampling are viable alternatives to the computationally expensive exact algorithm.

My doctoral research has focused on the development of surface plasmon resonance (SPR) reflectance imaging technique to detect near-field transport properties such as concentration, temperature, and salinity in micro/nano fluidic phenomena in label-free, real-time, and full-field manner. A label-free visualization technique based on surface plasmon resonance (SPR) reflectance sensing is presented for real-time and full-field mapping of microscale concentration and temperature fields. The key idea is that the SPR reflectance sensitivity varies with the refractive index of the near-wall region of the test mixture fluid. The Fresnel equation, based on Kretschmann’s theory, correlates the SPR reflectance with the refractive index of the test medium, and then, the refractive index correlates with the mixture concentration or temperature. The basic operation principle is summarized and the laboratory-developed SPR imaging/analyzing system is described with the measurement sensitivity, uncertainties and detection limitations of the implemented SPR reflectance imaging. Total five proposed uses of SPR reflectance imaging technique are presented: (1) micromixing concentration field development of ethanol penetrating into water contained in a micro-channel, (2) full-field detection of the near-wall salinity profiles for convective/diffusion of saline droplet into water, (3) full-field and real-time surface plasmon resonance imaging thermometry, (4) correlation of near-field refractive index of nanofluids with surface plasmon resonance reflectance, and (5) unveiling hidden complex cavities formed during nanocrystalline self-assembly.

Hidden Markov Models (HMMs) are, undoubtedly, the most employed core technique for Automatic Speech Recognition (ASR). Nevertheless, we are still far from achieving high‐performance ASR systems. Some alternative approaches, most of them based on Artificial Neural Networks (ANNs), were proposed during the late eighties and early nineties. Some of them tackled the ASR problem using predictive ANNs, while others proposed hybrid HMM/ANN systems. However, despite some achievements, nowadays, the dependency on Hidden Markov Models is a fact. During the last decade, however, a new tool appeared in the field of machine learning that has proved to be able to cope with hard classification problems in several fields of application: the Support Vector Machines (SVMs). The SVMs are effective discriminative classifiers with several outstanding characteristics, namely: their solution is that with maximum margin; they are capable to deal with samples of a very higher dimensionality; and their convergence to the minimum of the associated cost function is guaranteed. In this work a novel approach based upon probabilistic kernels in support vector machines have been attempted for speech data classification. The classification accuracy in case of support vector classification depends upon the kernel function used which in turn depends upon the data set in hand. But still as of now there is no way to know a priori which kernel will give us best results The kernel used in this work tries to normalize the time dimension by fitting a probability distribution over individual data points which normalizes the time dimension inherent to speech signals which facilitates the use of support vector machines since it acts on static data only. The divergence between these probability distributions fitted over individual speech utterances is used to form the kernel matrix. Vowel Classification, Isolated Word Recognition (Digit Recognition), have been attempted and results are compared with state of art systems.

This study is based on three models, Markov model, Hidden Markov model and the Radial basis function neural network. A number of work has been done before about application of these three models to the stock market. Though, individual researchers have developed their own techniques to design and test the Radial basis function neural network. This paper aims to show the different ways and precision of applying these three models to predict price processes of the stock market. By comparing the same group of data, authors get different results. Based on Markov model, authors find a tendency of stock market in future and, the Hidden Markov model behaves better in the financial market. When the fluctuation of the stock price index is not drastic, the Radial basis function neural network has a nice prediction.

Det övergripande syftet med min C-uppsats är att granska lärarens bedömning och perceptio-ner för elever i behov av särskilt stöd och vidare belysa vilka möjligheter och begränsningar de anser sig ha för att kunna ta hänsyn till elever i behov av särskilt stöd samt få en syn på de skolsituationer som barn och elever med dolda funktionshinder och i behov av särskilt stöd kan befinna sig i. För att få svar på mina frågor har sex lärare inklusive en special lärare intervjuats och samti-digt diskuterades de psykologiska och biologiska faktorer kring barns och ungdomars utveck-ling. Den historiska återblicken ger oss en uppfattning på hur begreppet ”en skola för alla” har utvecklats inom loppet av tiden och vilka syn på barn i behov av särskilt stöd har pedagogerna idag. De centrala frågorna rörde sig om pedagogernas uppfattning om barn med koncentra-tionssvårigheter och deras syn på diagnostisering av barn med problem. En inkluderande inte-grering i jämförelse med segregering diskuterades också samt hur pedagogerna kan hjälpa dessa barn. Eleverna i skolan är olika och deras olikheter måste mötas med omtanke. För att uppfylla de-ras behov krävs kunniga och kompetenta personal i skolan. Skolan måste ha en fungerande och tillfredställande elevvård för att kunna nå skolans mål. Skolornas neddragningar på grund av ekonomiska problem gör att barn med dolda funktionshinder misslyckas allt oftare i da-gens skola. Dessa orsakar att barn får ett dåligt självförtroende med upprepade misslyckande och försämrar deras problem. Allmänt finns det en del olika faktorer som ligger bakom barn med koncentrations svårighe-ter. De biologisk och ärftliga faktorer samt tillväxtmiljön och deras samhällsställning kan ge-nerellt nämnas. Brist på tid, ekonomi och kunskap i skolorna är det en barriär för att kunna hjälpa barn med svårigheter. / The comprehensive purpose with this paper is to have a look at teacher’s assessment and per-ception of pupils with special educational needs. And further illustrate which possibilities and restrictions they believe to have, to take children with special educational needs into consid-eration and get a view of school situations who children with hidden functional disability and with special educations needs are at the present. To get answer to my questions, six teachers inclusive a special teacher for pupils with im-paired disabilities have been interviewed and discussed the psychological and biological fac-tors around children’s development. The historical review gave us a perceptive on how defini-tion of “school for all” has been developed during the time and what is teacher’s opinion on children with special needs today. The central questions were concentrating on teachers un-derstanding of children with concentration difficulty and their opinion on diagnostic of chil-dren with problems. An “including integration” compared to segregating been discussed as well and finally discussed how teachers can help these children. Pupils in school are not comparable and these differences must meets carefully. To meet chil-dren’s requirements schools have need of personnel’s competence and proficiency. Schools required having functioning and satisfactory pupil welfare to achieve the aim. Lowering of school resources due to economical problems do that child with hidden functional disability fails more often in schools these days. These effects cause that children get a horrific self-confidence and worsen their problems. Generally, there are different factors behind the con-centrations difficulty. The biological and hereditary factors as well as home environment and their class society can points out in general. Lacking of time, economy and knowledge stops teachers to helping children in school.

The aim with this study was to increase the knowledge about as well as shed light on hidden refugees’ living conditions by compiling research about hidden refugees’ living conditions and complete this research by doing a field survey with people with different knowledge and experience of hidden refugees’ living conditions. The result was analyzed on the basis of Nussbaum’s list of ten human abilities and the convention of human rights. The study showed that hidden refugees' living conditions are experienced as very critical, and characterized by lack of human rights as well as means to utilize and develop their abilities. The conclusion of the study is that hidden refugees’ living conditions could not be considered humane and that there is a need for ethics when treating hidden refugees.

Organohalide-respiring Bacteria (OHRB) play key roles in the reductive dehalogenation of natural organohalides and anthropogenic chlorinated contaminants. Reductive dehalogenases (RDases) catalyze the cleavage of carbon-halogen bonds, enabling respiratory energy conservation and growth. Large numbers of RDase genes, a majority lacking experimental characterization of function, are found on the genomes of OHRB. In silico genomics tools were employed to identify shared sequence features among RDase genes and proteins, predict RDase functionality, and elucidate RDase evolutionary history. These analyses showed that the RDase superfamily could be divided into proteins exported to the membrane and cytoplasmic proteins, indicating that not all RDases function in respiration. Further, Hidden Markov models (HMMs) and multiple sequence alignments (MSAs) based upon biochemically characterized RDases identified previously uncharacterized members of an RDase superfamily, delineated protein domains and amino acid motifs serving to distinguish RDases from unrelated iron-sulfur proteins. Such conserved and discriminatory features among RDases may facilitate monitoring of organohalide-degrading microbial communities or improve accuracy of genome annotation. Phylogenetic analyses of RDase superfamily sequences provided evidence of convergent evolution and horizontal gene transfer (HGT) across distinct OHRB genera. Yet, the low frequency of RDase transfer outside the genus level and the absence of RDase transfer between phyla indicate that RDases evolve primarily by vertical evolution or HGT is restricted among related OHRB strains. Polyphyletic evolutionary lineages within the RDase superfamily comprise distantly-related RDases, some exhibiting activities towards the same substrates, suggesting a longstanding history of OHRB adaptation to natural organohalides. Similar functional and phylogenetic analyses provided evidence that nitrous oxide (N₂O, a potent greenhouse gas) reductase (nosZ) genes from versatile OHRB members of the Anaeromyxobacter and Desulfomonile genera comprised a nosZ sub-family evolutionarily distinct from nosZ found in non-OHRB denitrifiers. Hence, elucidation of RDase and NosZ sequence diversity may enhance the mitigation of anthropogenic organohalides and greenhouse gases (i.e., N₂O), respectively. The tetrachloroethene-respiring bacterium Geobacter lovleyi strain SZ exhibited genomic features distinguishing it from non-organohalide-respiring members of the Geobacter genus, including a conjugative pilus transfer gene cluster, a chromosomal genomic island harboring two RDase genes, and a diminished set of c-type cytochrome genes. The G. lovleyi strain SZ genome also harbored a 77 kbp plasmid carrying 15 out of the 24 genes involved in biosynthesis of corrinoid, likely related to this strains ability to degrade PCE to cis-DCE in the absence of supplied corrinoid (i.e., vitamin B₁₂). Although corrinoids are essential cofactors to RDases, the strictly organohalide-respiring Dehalococcoides mccartyi strains are corrinoid auxotrophs and depend upon uptake of extracellular corrinoids via Archaeal and Bacterial salvage pathways. A key corrinoid salvage gene in D. mccartyi, cbiZ, occurs at duplicated loci adjacent to RDase genes and appears to have been horizontally-acquired from Archaea. These comparative genome analyses highlight RDase dependencies upon corrinoids and also suggest mobile genomic elements (e.g., plasmids) are associated with organohalide respiration and corrinoid acquisition among OHRB. In summary, analyses of OHRB genomes promise to enable more complete modeling of metabolic and evolutionary processes associated with the turnover of organohalides in anoxic environments. These efforts also expand knowledge of biomarkers for monitoring OHRB activity in anoxic environments, and will improve our understanding of the fate of chlorinated contaminants.

This thesis analyzes the impact of the political decision to broaden the scope of the Government of Canada's 3-year National Homelessness Initiative (Human Resource Development Canada, NHI, 2002) from an urban focus to one that includes smaller communities. This change provided the opportunity to study the phenomenon of homelessness and how rural responses are formed. This author postulates that this focus of attention on an almost invisible phenomenon—rural homelessness—and the accompanying community planning processes funded by the Supportive Community Partnership Initiative (SCPI) will impact local social policy development. A multi-dimensional analytical approach was adopted and considered three components: first, a policy review, a broad look at the policy agenda framework in Canada; then, a case study to illustrate implementation issues related to the National Homelessness Initiative; and, finally, a reflection on current practice in order to realize a holistic critique of public policy. The influence of socio-economic, political, and cultural factors on local planning and capacity building will be highlighted. Various models of governance were adopted across the country and guided the collaborative processes. This thesis presents an in-depth look at the community action plans and activities of the Simcoe County Alliance to End Homelessness (SCATEH) in both the rural and urban settings of Simcoe County. The processes adopted, capacity building components identified, and outcomes over the 7 years covered by the SCPI agreement are examined. The limitations of using participatory local action planning to respond to complex issues such as homelessness are detailed along with a modified community-based policy development model recommended as a learning tool to be used by those volunteers acting as agents of change. It is widely recognized that safe, affordable social housing is a fundamental need, and one that is extremely difficult to meet. The contribution this research makes is to reveal how effective government-community partnerships can be in a rural setting.