[Abstract]: Web page access prediction gained its importance from the ever increasing number of e-commerce Web information systems and e-businesses. Web page prediction, that involves personalising the Web users’ browsing experiences, assists Web masters in the improvement of the Web site structure and helps Web users in navigating the site and accessing the information they need. The most widely used approach for this purpose is the pattern discovery process of Web usage mining that entails many techniques like Markov model, association rules and clustering. Implementing pattern discovery techniques as such helps predict the next page tobe accessed by theWeb user based on the user’s previous browsing patterns. However, each of the aforementioned techniques has its own limitations, especiallywhen it comes to accuracy and space complexity. This dissertation achieves better accuracy as well as less state space complexity and rules generated by performingthe following combinations. First, we combine low-order Markov model and association rules. Markov model analysis are performed on the data sets. If the Markov model prediction results in a tie or no state, association rules are used for prediction. The outcome of this integration is better accuracy, less Markov model state space complexity and less number of generated rules than using each of the methods individually. Second, we integrate low-order Markov model and clustering. The data sets are clustered and Markov model analysis are performed oneach cluster instead of the whole data sets. The outcome of the integration is better accuracy than the first combination with less state space complexity than higherorder Markov model. The last integration model involves combining all three techniques together: clustering, association rules and low-order Markov model. The data sets are clustered and Markov model analysis are performed on each cluster. If the Markov model prediction results in close accuracies for the same item, association rules are used for prediction. This integration model achievesbetter Web page access prediction accuracy, less Markov model state space complexity and less number of rules generated than the previous two models.

Melbourne, the capital of Victoria Australia leads the world in having the highest quality drinking water. The Victorian State Government has set targets for reducing per capita water consumption by 15%, 25% and 30% by 2010, 2015 and 2020 respectively and has announced stringent water restrictions to curtail water demand. In this resource constraint environment it is opportune to look for alternative sources of water to supplement Melbourne's traditional water supply. In Melbourne, legislation has been changed to make it possible to use rainwater harvested from domestic tanks for non potable purposes. The annual rainfall in Melbourne's metropolitan area varies from 450mm in the West to 850mm in the East to over 1000mm in the North East mountain ranges. The objectives of the current study are to develop a methodology to estimate the optimal size of the rainwater tank at a particular location considering the local rainfall, roof area, demand for water and the reliability of supply (supply security) required; to quantify the rainwater volume that could be harvested at site using domestic rainwater tanks to minimise pressure on the potable water supply secured from traditional catchment sources until the desalination plant is commissioned in 2013; to analyse the efficacy of rainwater tanks to reduce the stormwater runoff and improve the quality of the stormwater that will otherwise flow into urban drains and to estimate the cost effectiveness ratio and payback period of inst alling rainwater tanks. A simple water balance model was developed to calculate the tank size based on daily rainfall, roof area and the expected demand. The concept of 'reliability' was introduced to measure supply security. Rainfall data from 20 rainfall stations scattered around Melbourne were used to determine the variation in the rainwater tank size dependent on the above stated parameters. It was observed that to achieve the same supply reliability (90%) and to meet a specific demand (toilet and garden use), the tank size required in the western side of Melbourne is as high as 7 times as that required in the north-east side. As a result, the

Les travaux présentés se situent dans le cadre d'extraction de connaissance à partir de données. Un contexte d'étude intéressant et d'actualité a été choisi : les sites web adaptatifs. Pour mettre en oeuvre, de manière la plus automatique possible, de tels sites adaptés aux utilisateurs, nous décidons d'apprendre des modèles d'utilisateurs ou, plus précisément, de leurs types de navigations sur un site web donné. Ces modèles sont appris par inférence grammaticale. Les données disponibles liées au contexte du Web sont particulièrement difficiles à récupérer proprement. Nous choisissons de nous focaliser sur les fichiers de logs serveur en supprimant le bruit inhérent à ces derniers. L'inférence grammaticale peut généraliser ses données d'entrée pour obtenir de bons modèles de langages. Nous travaillons sur les mesures de similarité entre langages pour l'évaluation de la qualité des modèles appris. L'introduction d'une mesure euclidienne entre modèles de langages représentés sous forme d'automates permet de pallier les problèmes des métriques existantes. Des résultats théoriques montrent que cette mesure a les propriétés d'une vraie distance. Enfin, nous présentons divers résultats d'expérimentation sur des données du web que nous pré-traitons avant d'apprendre grâce à elles des modèles utilisateurs issus de l'inférence grammaticale stochastique. Les résultats obtenus sont sensiblement meilleurs que ceux présents dans l'état de l'art, notamment sur les tâches de prédiction de nouvelle page dans une navigation utilisateur.

<p>Introduction</p><p>This paper focuses on the Internet and more specifically on how non-profit organizations can use the Internet to enhance their online presence. Our choice of research object is Automation Region (AR), a non-profit cluster organization, whose purpose is to strengthen and make the cluster of companies within industrial automation in Mälardalen visible. This cluster consists of business-to-business companies in the automation industry and AR was formed to increase the productivity, growth and profitability of these companies. The organization has two goals set up, which include: (1) increased exposure for the automation industry as well as the region wherein AR operates and (2) attracting the best competence as well as getting youths and children interested in automation. AR currently has a website (www.automationregion.com) and the overall purpose for it is to support AR's goals. However, we have a difficulty seeing how this website, as it looks and functions today, is to achieve the goals and objectives of the cluster organization and we also find it difficult to see how AR should do in order to expose its website and its brand to its target audience. The two research questions then becomes: How is Automation Region's website being used today? How can Automation Region utilize the Internet to achieve its goals?</p><p>Purpose</p><p>The purpose of this study is to gain a deeper understanding of how cluster organizations like Automation Region may use the Internet to achieve its goals, and by doing this also be able to give recommendations on how Automation Region can benefit from the use of the Internet as their primary channel for goal achievement.</p><p>Method - An Action Research based study</p><p>This paper is based on an action research approach, which contains a real world problem, a framework of ideas, and a set of methodologies. The different means of collecting data used in this paper can be categorized in four separate categories: interviews, internet analysis tool, benchmarking, and other empirical data collected.</p><p>Framework of ideas</p><p>Our framework of ideas consists of an introduction to the term clusters, which then explains the different actors on the cluster stage. Then the benefits of clusters are discussed, leading to a further discussion on how non-profit cluster organizations can utilize the possible benefits of a cluster. This leads to a presentation of different ways to improve an organizations online presence.</p><p>Analysis</p><p>The analysis is divided into two parts. The first part focuses on the results of the website analysis, where different reports from the web analysis program are analyzed. This will show how the website currently is being used and should lead to conclusions on how this usage can be improved. The second part focuses on the available Internet tools and how AR can utilize these to reach their explicit goals for the cluster organization.</p><p>Conclusions</p><p>The so called internal visitors are plentiful on AR's website site and we can conclude that the site is being used in an Intranet like way. Another conclusion is that AR should be able to increase the probability of achieving their goals by further enhancing the use of the Internet tools they currently have (website and e-mail), and also by expanding their online presence with the use of several other potential Internet tools.</p>

<p>The purpose of this paper is to examine the adverbs <em>mainly</em>, <em>mostly</em>, <em>chiefly</em> and <em>principally</em> for their level of synonymity. Dictionaries commonly lists synonyms to their main entries without providing information about commonality, contexts or level of formality or even sample sentences. More often than not this is done in a circular fashion, one main entry is listed as synonym to another which in turn is listed as a synonym to the first. This is confusing to persons not knowing what word to use and when to use it. The adverbs <em>mainly</em>, <em>mostly</em>, <em>chiefly</em> and <em>principally</em> are listed as synonyms to each other in several dictionaries in this circular fashion. By using corpus research I have shown that these adverbs are not the absolute synonyms dictionaries indicate, but differ greatly in usage, context and level of formality. I have investigated real samples of how <em>mainly</em>, <em>mostly</em>, <em>chiefly</em> and <em>principally</em> are used in a British newspaper. From the results I have obtained, I have analyzed the usage of the adverbs and suggested new and improved definitions.</p>

The purpose of this study was to determine the factors that influence why students attend or do not attend the fitness and recreation centers provided to them on campus. With the obesity epidemic growing substantially over the past several decades, it has consequentially affected a vast majority of the nation, including the college-age population. In response to this, the use of fitness and recreation centers on college campuses and universities has become a focal point for administrators and recreation center managers as a focus on student recruitment and retention. The study utilized an online questionnaire which was adapted and revised from a prior survey and corresponding research that assessed fitness and recreation center usage within the college population (Zizzi, Ayers, Watson & Keeler, 2004). The survey used for collection of data in this study assessed factors related to college students’ usage or non-usage of campus recreational facilities by college students. It was distributed to a random sample of 1,100 undergraduate students at Indiana University of Pennsylvania through their student e-mail account. Results from the data collected in this study revealed that users were more likely to be male, Caucasian, non-smokers, and have no health concerns, which is representative of prior research on fitness and recreation center participation. Factors related to usage of the recreational facilities included maintaining one’s current weight as well as weight loss. Factors that influenced occasional non-usage in the user group were lack of time and energy, too much schoolwork, and inadequate facility operational hours. A factor related to non-usage included proximity of the recreational facility in relation to the students’ living arrangements. University administrators and fitness and recreation center managers need to be conscious of the factors that influence usage for the purpose of encouraging student recruitment and retention. Higher levels of usage will increase physical activity and influence health and wellness in the college population. / Dr. Robert Kostelnik Dr. Robert Alman Dr. Joshua Castle

This thesis examined the factors of L1, input frequency and emergent productivity in child L2 acquisition. This thesis is the first study to look at the interplay of L1 and usage factors in children learning a L2. The focus of the thesis was an investigation of these factors in the acquisition of article and auxiliary systems of English, which have been proven to be problematic areas for both L1 and L2 learners. While accounts of L1 transfer in L2 are better developed in generative theory, the roles of input frequency and emergent productivity are better developed in constructivist theory. The thesis assessed these two approaches against the data from L2 children from various L1 backgrounds. The children’s accuracy and error patterns with articles and auxiliaries were investigated. The main findings were as follows. L1 typology facilitated the acquisition of the structure of the NP and VP, but it only extended as far as the awareness of the presence of the functional morpheme (article or auxiliary). L1 transfer effects were observed only in the first 1.5 years of acquisition, which could be due to the unstable L1 knowledge in child L2 learners. The use of articles and auxiliaries was also influenced by their input frequencies and distribution, as more frequent forms were supplied more accurately and were substituted for less frequent forms. Different forms of articles and auxiliaries emerged separately and followed different paths of development. It was argued that they were acquired piecemeal and that productivity with these forms emerged gradually. It was concluded that constructionist theories were better supported by the data, since the findings on input frequency and productivity were not compatible with the generative approach, and L1 transfer was incorporated into the constructionist approach to account for the findings. It was argued that by the onset of acquisition, child L2 learners had established constructions in their L1 that were abstract enough to be transferred to L2 and did not rely on lexically specific information. As all children learned specific morphological forms of L2 piecemeal, in doing so they demonstrated input effects that held across all L1 backgrounds.

<p>Internet banking has an impact on banking performance as well as customer’s daily life. Customers are getting more used to use Internet banking services because the Internet is so popular and common available, it is more convenient and comfortable for customers to use banking services on the Internet. Hereby Internet banking also helps to improve banking service and increasing banks profitability by reducing costs. It also increases the overall value of the banks services by providing convenience, reliable, faster, cheaper services. The purpose of study is to find out which services customer prefers to execute the Internet bank and which services customers prefer to execute through branch office and based on this, what combination of services are desirable for banks customers in Visby. Finding shows that young and educated people are more frequently users of Internet banking for balance inquiry and for viewing the transaction history of his/her account. For some services customers prefer to visit branch office when opining a new account, deposit money, insurances service and loan activities. Banks encourage their customers to use online services but there is need that there are some activities that customer must go to the branch office.</p>

tRNA genes have the fundamental role of translating the genetic code during protein synthesis. Beyond solely a passive decoding role, the tRNA pool exerts selection pressures on the codon usage of organisms and the viruses that infect them because processing codons read by rare tRNAs can be slow or even erroneous. To better understand the interactions of codons and anticodons in eukaryotic species, we first investigated whether tRNAs packaged into HIV-1 particles may relate to the poor codon usage of HIV-1 genes. By comparing the codon usage of HIV-1 genes with that of its human host, we found that tRNAs decoding poorly adapted codons are overrepresented in HIV-1 virions. Because the affinity of Gag-Pol for all tRNAs is non-specific, HIV packaging is most likely passive and reflects the tRNA pool at the time of viral particle formation. Moreover, differences that we found in the codon usage between early and late genes suggest alterations in the tRNA pool are induced late in viral infection. Next, we tested whether a reduced tRNA anticodon pattern, which was called into question by predicted tRNA datasets, is maintained across eukaryotes. tRNA prediction methods are prone to falsely identifying tRNA-derived repetitive sequences as functional tRNA genes. Thus, we proposed and tested a novel approach to identify falsely predicted tRNA genes using phylogenetics. Phylogenetic analysis removed nearly all the genes deviating from the anticodon pattern, therefore the anticodon pattern is reaffirmed across eukaryotes.

tRNA genes have the fundamental role of translating the genetic code during protein synthesis. Beyond solely a passive decoding role, the tRNA pool exerts selection pressures on the codon usage of organisms and the viruses that infect them because processing codons read by rare tRNAs can be slow or even erroneous. To better understand the interactions of codons and anticodons in eukaryotic species, we first investigated whether tRNAs packaged into HIV-1 particles may relate to the poor codon usage of HIV-1 genes. By comparing the codon usage of HIV-1 genes with that of its human host, we found that tRNAs decoding poorly adapted codons are overrepresented in HIV-1 virions. Because the affinity of Gag-Pol for all tRNAs is non-specific, HIV packaging is most likely passive and reflects the tRNA pool at the time of viral particle formation. Moreover, differences that we found in the codon usage between early and late genes suggest alterations in the tRNA pool are induced late in viral infection. Next, we tested whether a reduced tRNA anticodon pattern, which was called into question by predicted tRNA datasets, is maintained across eukaryotes. tRNA prediction methods are prone to falsely identifying tRNA-derived repetitive sequences as functional tRNA genes. Thus, we proposed and tested a novel approach to identify falsely predicted tRNA genes using phylogenetics. Phylogenetic analysis removed nearly all the genes deviating from the anticodon pattern, therefore the anticodon pattern is reaffirmed across eukaryotes.