Owing to the booming development of the WWW, it creates many new challenges for information filtering. Information Filtering (IF) is an area of research that develops tools for discriminating between relevant and irrelevant information. IF can find good matches between the web pages and the users' information needs. Users first give descriptions about what they need, i.e., user profiles, to start the services. A profile index is built on these profiles. A series of incoming web pages will be put into the matching process. Each incoming web page is represented in the same form of the user profile. In this way, the users who are interested in an incoming web page can be identified by comparing the descriptions of the web page with each user profile. At last, the web page will be recommended to the users whose profiles belong to the filtered results. Therefore, a critical issue of the information filtering service is how to index the user profiles for an efficient matching process. When we index the user profile, we can reduce the costs of storage space and the processing time for modifying the user profiles. In this thesis, first, we propose a count-based tree method, which takes the count of each keyword into consideration, to reduce the large storage spaces as needed by the tree method. Next, three large-itemset-based methods are proposed to reduce the storage space, which are called the count-major large itemset method, the weighted large itemset method and the hybrid method. In these three large-itemset-based methods, we first cluster profiles with similar interests into the same group. Next, for each cluster, we apply the mining association rules techniques to help us to construct the index strategies. We design three methods by using the idea of the Apriori algorithm which is one of well-known approaches in mining association rules. But, we modify the minimum support and the goal in the Apriori algorithm. We may not always output the large itemset Lk. That is, we may only use Lw, where w < k. In summary, the cost of storage spaces of our four methods are less than that of the tree method proposed by Yan and Garcia-Molina. According to our simulation results, each of our four methods may provide the best result when different input data sets. Next, we propose a large-itemset-based approach to the incremental update of the index structure for storing keywords to reduce the update cost. When someone's interests are often changed, we must care about the way how to provide the low update cost of the index structure. We take the weight of each keyword into consideration. That is, each keyword can be distinguished the long-term interest which has weight above the threshold from the short-term interest which has weight below the threshold. Owing to that the probability of modifying the short-term interests is higher than that of modifying the long-term interests, we can update the short-term interests locally. According to our simulation results, our method really can reduce the update cost as needed by Wu and Chen' methods.

<p>The internet itself is continuously changing. Since it was widely introduced, the users’ main focus has shifted from a communication tool, to a huge file sharing pool, to a marketing tool, to an information gathering and information reproducing tool.</p><p>Nowadays the way of sharing files needs to be reconsidered, because more and more private information is shared over the internet. The common file sharing applications are no longer appropriate for today’s purposes. Also the common way of socializing people is not up to date anymore.</p><p>Sharing files already implies a communication between users. But why do we not use the given information of shared data in order to socialize them? This thesis discusses the problematic domains of sharing images and socializing and introduces a tool named Picnic that has been conceived and implemented in order to overcome those problems.</p>

Much research show that strategies of personalization aiming to meet customers' needs and behaviors are an important aspect to influence the competitiveness within an industry. However there is a gap of studies on how airlines use customer involvement for creating personalized strategies in the field of e-commerce while enhancing the competitive advantage. This study is aims to describe how airlines can create personalized e-commerce strategies by considering the customer involvement in order to enhance their competitiveness. In conjunction with this first assumption the study aims to demonstrate how airlines could increase their customer loyalty by implementing the concept of customer relationship management (CRM) within their e-commerce strategies. A theoretical framework has been developed derived from a review of the existing literature in order to describe how airlines can use personalization as a strategy in e-commerce to develop a competitive advantage.  A qualitative multiple-case study has been chosen for this study. Three companies have been studied within the European airline industry and have been led by a deductive approach which will aim to investigate the theoretical framework.The analysis of empirical findings revealed some relevant conclusions, which can bring value to the research area, and also to the practice. The main findings show that airlines are involving customers into their e-commerce strategies in various forms and different levels regarding a specific project. One of the major result of this thesis is that airlines seem to focus on point of contacts with customers external to the e-commerce platform.

Σκοπός της παρούσας μεταπτυχιακής εργασίας είναι η μελέτη και η αξιολόγηση των δυνατοτήτων προηγμένης σημασιολογικής αναζήτησης (advanced semantic search) πάνω σε περιεχόμενο που προέρχεται από το Διαδικτύο. Στα πλαίσια της εργασίας, σχεδιασθηκε και υλοποιήθηκε υποσύστημα, το οποίο ενσωματώθηκε και αξιολογήθηκε πάνω στο μηχανισμό peRSSonal ο οποίος ανακτά, επεξεργάζεται και παρουσιάζει στους χρήστες του άρθρα και υλικό από διάφορες ειδησεογραφικές πύλες (news portals) του Διαδικτύου, προσαρμόζοντάς τα στις προσωπικές επιλογές του χρήστη. Η αναζήτηση παραμετροποιείται με στοιχεία όπως: χρονικό πλαίσιο δημοσίευσης τού υλικού (διάστημα από – έως), κατηγορία (πρότυπες κατηγορίες του συστήματος peRSSonal), φυσική γλώσσα στην οποία είναι γραμμένο καθώς και δυνατότητα για στατικό ή σημασιολογικό (εννοιολογικό) ταίριασμα (semantic matching) με τα άρθρα της βάσης. Αρχικά, από την επερώτηση (query) του χρήστη δημιουργείται ένα σύνολο ριζών (stems) των λέξεων οι οποίες δόθηκαν. Η εξαγωγή των ριζών εκτελείται με υποβοήθηση από stemming αλγορίθμο για την αγγλική γλώσσα, ενώ ο σχεδιασμός του συστήματος προβλέπει και τη μελλοντική υποστήριξη διαφορετικών φυσικών γλωσσών καταβάλλοντας μικρό κόπο. Για τις λεκτικές ρίζες που προκύπτουν, εντοπίζονται σχετικές τους και ταυτόχρονα με τη διαδικασία αυτή διενεργείται αναζήτηση στη βάση δεδομένων για κωδικολέξεις (keywords) με βάση την κατηγορία του άρθρου, ούτως ώστε να εμπλουτιστεί το ερώτημα του χρήστη με επιπλέον πληροφορία, καθιστώντας πιο επιτυχημένη και στοχευμένη την αναζήτηση στην πληθώρα των άρθρων που υπάρχουν αποθηκευμένα στη βάση δεδομένων. Για αυτές τις κωδικολέξεις υπολογίζονται συντελεστές – βάρη που θα προσδιορίζουν τη συνάφειά τους με την επερώτηση του χρήστη. Ανάλογα με τον τύπο της αναζήτησης, στατική η σημασιολογική, το υποσύστημα αναζήτησης συγκρίνει την επερώτηση του χρήστη με τα αποθηκευμένα άρθρα και για κάθε ένα από αυτά, ο αλγόριθμος υπολογίζει το βαθμό συνάφειάς του με την επερώτηση. Τα άρθρα που επιλέγονται τελικά είναι αυτά που ξεπερνούν ένα κατώφλι συνάφειας, το οποίο τα κατατάσσει εννοιολογικά πιο κοντά στην επερώτηση του χρήστη. Σημαντικό σημείο στο στάδιο αυτό, είναι η δυνατότητα, για τους εγγεγραμένους χρήστες της Δικτυακής Πύλης, να εκτελείται περεταίρω φιλτράρισμα πάνω στο πρωτογενές αποτέλεσμα, βάσει των προσωπικών τους επιλογών καθώς και πληροφορίας που προέρχεται από τη βάση δεδομένων και που διαμορφώνεται δυναμικά από την παρατήρηση της γενικής συμπεριφοράς των χρηστών κατά την πλοήγηση τους μέσα στον σύστημα (χρόνος παραμονής στα άρθρα, άρθρα που δεν προτιμώνται, συχνότητα επιλογής άρθρων από μια δεδομένη θεματική ενότητα κλπ). Σκοπός είναι η εξαγωγή πιο στοχευμένου συνόλου άρθρων που ικανοποιεί τελικά περισσότερο τον χρήστη. Τέλος, για την βελτίωση της απόδοσης του συστήματος, σχεδιάστηκε και υλοποιήθηκε αλγόριθμος που εκτελεί caching στα αποτελέσματα των επερωτήσεων. Με τον τρόπο αυτό, κάθε νέα αναζήτηση θα λαμβάνει πολύ πιο γρήγορα τα cached αποτελέσματα προγενέστερων παρόμοιων αναζητήσεων, ξοδεύοντας το χρόνο στα πιο πρόσφατα άρθρα. Το caching εκτελείται δυναμικά, τροποποιώντας σε κάθε επερώτηση που υποβάλλεται τα αντίστοιχα cached αποτελέσματα και μεταβάλλοντας τις προτεραιότητές τους και τα βάρη τους, ώστε να οδηγεί την έξοδο ολοένα και πιο κοντά στα επιθυμητά άρθρα και παραμένοντας πιο κοντά στο εξελισόμενο προφίλ και στις προτιμήσεις του χρήστη. Μέσα από την εργασία, προέκυψαν αποτελέσματα που έχουν να κάνουν με σύγκριση αλγορίθμων σε όλα τα παραπάνω στάδια του μηχανισμού αλλά και ανταπόκριση του μηχανισμού στις ανάγκες του χρήστη. / The scope of the present MSc Thesis is the study and the evaluation of the features provided by an advanced semantic search over digital content which comes from the Internet. For the purposes of our work, we designed and implemented a module (subsystem), which was embedded and evaluated on the PeRSSonal news portal. The PeRSSonal news portal retrieves, processes and presents to the end user articles and other content from major News Portals of the Internet by adapting on the user’s personal preferences and profile. For the search procedure, parameters such as the date interval, the thematic category and the article’s language are used. Furthermore it is possible to use static or dynamic (semantic) matching with the articles of the database. In the first phase of the procedure, from the query that the user submitted we create a set of keywords, which are the stemmed words of the words described in the initial query. The extraction stemmed words is executed by an algorithm which implements the Porter Stemmer technique. The system currently supports the English language in the search procedure but its modular architecture allows for the support of other languages as well with little effort. For the keywords produced with this procedure, we locate their synonyms and in the same time a search in the database is conducted in order to find other keywords based on the thematic category of the submitted query. This second set of keywords enriches the first set thus making the search more focused on the thematic category the user chose. For these keywords that enrich the initial query, weights are computed based on their relation with the keywords of the initial query. Based on the type of the search (static or semantic), the search subsystem compares the enriched set of keywords with the articles stored in the database and for each one of these articles which match to the keywords of the query, a degree of relevance is computed. The articles that are selected to be in the final result are the ones that surpass a specific threshold of relevance which semantically brings them close to the user query. A significant point during this phase of the procedure is the possibility to execute for the registered users of the PeRSSonal portal a more detailed filtering on the primal result based on their personal preferences and data that is produced dynamically by observing their behavior (time they spend on the articles, not preferred articles, frequency of selecting a specific thematic category) in the system, during the sessions in that. The goal is the creation of a more focused result on the end user which satisfies him more. In the final phase of the algorithm, and in order to optimize the algorithm’s performance, we design and implement an algorithm which uses cache memory in the form of a database table and runs on the server machine. For each query that is submitted to the system, we store the retrieved results in this table and in the future queries, prior to triggering the search procedure, we compare the queries with the cached ones.In this way, every new search that already has a match in the cache table will consume much less time to execute as it will search only for articles which are not found in the cache. The caching algorithm is executed dynamically by modifying for every submitted query the cached results and by changing their priorities and their relevance weights in order to include in the output the desired articles and to stay closer to the user’s profile and preferences. From the experimental results of this work we had the chance to draw useful conclusions by the comparison of different algorithmic approaches for all the stages of the mechanism and by the response and performance of the algorithm as faced by the end user.

The internet itself is continuously changing. Since it was widely introduced, the users’ main focus has shifted from a communication tool, to a huge file sharing pool, to a marketing tool, to an information gathering and information reproducing tool. Nowadays the way of sharing files needs to be reconsidered, because more and more private information is shared over the internet. The common file sharing applications are no longer appropriate for today’s purposes. Also the common way of socializing people is not up to date anymore. Sharing files already implies a communication between users. But why do we not use the given information of shared data in order to socialize them? This thesis discusses the problematic domains of sharing images and socializing and introduces a tool named Picnic that has been conceived and implemented in order to overcome those problems.

A internet introduziu uma nova economia a economia da era digital, em que todos os consumidores e empresas estão em rede. A economia tradicional foi fragmentada, convertendo cada vez mais os mercados de massa em mercados de nicho. E os impactos destas transformações podem ser sentidos nas estruturas organizacionais, no ambiente econômico e nas relações entre as pessoas. Diante destas transformações, as empresas têm repensado os seus modelos de negócios e suas estratégias competitivas. No centro destas transformações se encontra o cliente, antes visto como passivo, hoje é considerado um recurso e competência da companhia. Para atender a este novo fator, a estratégia de personalização tem sido adotada pelas companhias. A personalização visa atender à necessidade dos clientes de personalizar não apenas o produto fisicamente, mas a sua experiência no momento da compra. É neste contexto que este trabalho é desenvolvido, com o objetivo de buscar um melhor entendimento de como a Tecnologia da Informação tem impactado e influenciado os fatores que habilitam a estratégia de personalização. Para atingir o objetivo, inicialmente foi feita uma ampla pesquisa bibliografia para identificar os fatores que influenciam a personalização. Em seguida, através da abordagem de estudos de caso múltiplos foi investigado se estes fatores ocorriam empiricamente e como eles ocorriam. Os resultados encontrados permitiram identificar divergências e concordâncias nas afirmações encontradas na literatura, que ao final fornecem refinamento e a necessidade de uma maior reflexão sobre os fatores que tem influenciado e direcionado a personalização. / The internet introduced a new economy the economy in the digital era, in which all the consumers and the companies are connected. The traditional economy was fragmented, converting the mass markets into niche markets. And these changes are impacting the organizational structures, the economic environment and the human relations. To cope with these transformations, companies have to rethink their business models and their competitive strategies. In the middle of these transformations is the customer, former seeing as passive, now is considered a company resource and a competence to pursue. In order to attend to this new factor, the personalization strategy has being adopted by the companies. The personalization\'s goal is to attend the customer\'s need for personalization, not only physically modifying the product, but creating their own experience in the buying process. This is the context where this research occurs, with the objective of achieving a better understanding about how the Information Technology impacts and influences the personalization strategy enablers. In order to achieve the objective, a broad literature research was initially made to identify the personalization influencing factors. Following, through a multiple case study approach, it was investigated if these factors could be found empirically and how they were occurring. The results allowed the identification of divergences and concordances in the facts found in literature that ultimately provides theory refinement and the necessity of a deeper reflection about the factors that are influencing and driving the personalization.

Este trabalho tem como objetivo investigar e discutir, a partir da Psicanálise, os fundamentos da constituição da corporeidade. Além disso, pretende-se enfatizar o paradoxo como sua característica marcante: ela é psíquica e somática, intrapsíquica e intersubjetiva. É feita uma analogia entre a corporeidade e a figura de uma casa, ressaltando, novamente o elemento paradoxal: a casa é espaço de recolhimento e, simultaneamente, de abertura ao outro, ao mundo. O processo de personalização, descrito por Donald Winnicott, será aqui concebido como o paradigma da constituição da corporeidade, representado pela imagem do bebê no colo da mãe: sustentado no cuidado dela, o bebê pode elaborar imaginativamente o corpo e suas funções. O resultado disto é o sentimento de habitar o próprio corpo, marcado pela presença do outro. De acordo com Winnicott, a personalização não é algo dado, mas uma conquista do desenvolvimento emocional primitivo. Se este processo não se der de maneira satisfatória, haverá um prejuízo em relação ao estabelecimento da morada no corpo e também na relação do indivíduo com o mundo. A complementaridade fundamental entre o indivíduo e seu meio será ampliada da relação entre mãe e bebê para a relação entre indivíduo e cultura, a fim de examinar certos aspectos do contexto sócio-político na atualidade articulando-os às suas implicações no âmbito dos processos de subjetivação / The aim of this study is to investigate and discuss, from a psychoanalytical perspective, the fundamentals of the constitution of embodiement. Moreover, it intends to emphasize paradox as its hallmark: it is both psychic and somatic, intrapsychic and intersubjective. The study makes an analogy between embodiement and home, emphasizing again the paradoxical aspect: home is at the same time a place of refuge and of opening to others, to the world. The personalization process, described by Donald Winnicott, is understood here as a paradigm of the constitution of embodiement, represented by the image of a baby in its mothers lap: sustained in her care, the baby may imaginatively develop the body and its functions. The result is the feeling of inhabiting the body, marked by the presence of another. According to Winnicott, personalization is not a given, but an achievement of the primitive emocional development. If this process does not occur satisfactorily, there will be a loss in establishing the inhabiting of the body and also in the individuals relationship with the world. This study will expand the fundamental complementarity between the individual and their environment from the mother-baby relationship, to the relationship between the individual and their culture, in order to examine certain aspects of the current socio-political context, articulating them to their implications in view of subjectivation processes

A fim de refletirmos sobre as condições de produção na contemporaneidade, propomos discussões e análises em torno do funcionamento dos recursos de personalização de conteúdos na Internet que impõem filtros aos usuários. Para isso, mobilizamos a teoria de Análise do Discurso de linha francesa, sobretudo a partir dos trabalhos de Michel Pêcheux. Delimitamos a observação do funcionamento de tais recursos no mecanismo de busca do Google, uma vez que ele representa uma das maiores corporações na Internet e detém a maior parte do mercado de buscas. A partir dessa delimitação, selecionamos o corpus de análise oriundo da política de privacidade do buscador, de definições dadas pela empresa sobre o funcionamento de tais recursos e de relatos sobre esse funcionamento postados por usuários e disseminados na rede. Acreditamos que a análise do funcionamento dessas técnicas nos dizem sobre as condições de produção do discurso na contemporaneidade, bem como das materialidades discursivas que lhes são peculiares, compreendendo que é pela influência materialista que se chega, na Análise do Discurso, a noção de materialidade discursiva. Temos, dessa forma, contemplada uma discussão sobre história, sujeito e língua que fundamenta o funcionamento da ideologia. Ainda nesse ínterim, propomos pensar o discurso ressaltando o que concerne à memória, bem como as questões que decorrem ao considerarmos seu funcionamento perpassado pelas Tecnologias de Informação e Comunicação (TIC\'s), o que nos move a refletir sobre o discurso eletrônico e a memória metálica. Tudo isso nos mostra os efeitos discursivos dos filtros que giram em torno do controle na rede, característica que tem marcado as relações de poder no cenário atual. / Aiming to reflect on the conditions of production in contemporary society, we propose discussions and analysis about the functioning of the contents customization features on the Internet that that imposes filters to users. For that, we mobilize the Discourse Analysis Theory of French orientation, especially from Michel Pêcheux. We delimited the observation of the functioning of such features in the Google search engine, once it is one of the largest corporations on the Internet and holds most of the search market. From this definition, we selected the corpus of analysis derived from the privacy policy of the search engine, the definitions given by the company on the operation of such resources and comments about this functioning posted by users in specialized forums. We believe that the observation of the functioning of such techniques tell us about the discourse production conditions in contemporary society, as well as the discursive materiality that are peculiar, understanding that it is the materialistic influence that comes the notion of discursive materiality in Discourse Analysis. We thereby contemplated a discussion of history, subject and language that underpins the functioning of ideology. Yet in the interim, we propose to think the discourse emphasizing that concerns the memory, as well as issues that derive when considering its operation permeated by Information and Communication Technologies (ICTs), which moves us to reflect on the electronic discourse and the metal memory. All this shows the discursive effects of filters that revolve around the control of on the network, a feature that has characterized relations of power in the current scenario.

Submitted by Irene Nascimento (irene.kessia@ufpe.br) on 2015-03-03T19:00:11Z No. of bitstreams: 2 TEXTO FINAL.pdf: 941406 bytes, checksum: 7fa4b2e89b6c1f051c94eeb58042fbdd (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) / Made available in DSpace on 2015-03-03T19:00:11Z (GMT). No. of bitstreams: 2 TEXTO FINAL.pdf: 941406 bytes, checksum: 7fa4b2e89b6c1f051c94eeb58042fbdd (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Previous issue date: 2012-02-06 / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / This dissertation intends to analyze the attribution of the juridical category of “person” to the unborn child in the context of the juridical-philosophical formulation of meanings attributed to the term person, from the Roman tradition until the context of the crisis of legal security by codification in which the dignity is juridically chosen as a value of the human person. In Roman law, the concept of the person (persona) was not endowed from the abstract, how it was attributed by the Pandectistic of the 19th century, but embraced the human being concretely in his multiple dimensions, inclusively considering itself as such the unborn child, to which was guaranteed the rights in view of his birth. In regard of axiom conceptus pro iam nato habetur, based in the roman treatment given to the unborn, emerged two interpretations: one, more faithful to the Justinian tradition, which recognizes the concrete reality of the unborn child and equates him to the already born, and the other, abstract, sustained by the Pandectistic and based on the thoughts of Savigny, which considers the parity between the unborn and the born simply a mere fiction. This latter interpretation influenced the elaboration of many civil codes like the German and the Brazilian Code of 1916, both of which determined that the natural personhood can be attributed only when the child is born alive. The idea of personhood while a simple legal attribution withdraws itself from the philosophical tradition that considers the person in his ontological aspects, a fact that deserves to be recognized as primary in the juridical planning. The principle of human dignity, incorporated in the Brazilian Constitution of 1988 as the foundation of the Republic, serves as an indication of an ‘ethical personalism’ which reflects itself in personal rights, providing elements to enlarge the semantic content of the juridical concept of person, recognizing the personal dignity of the unborn child, as done in the ‘Pact of San Jose, Costa Rica’ (American Convention on Human Rights), to singularize the juridical personhood as a right and the unborn child as a person. Keywords:

Hemsidan handgjordasaker.se är en Internetmarknadsplats som tillhandahåller tjänsten för försäljning av handgjorda produkter. Detta projekt syftar till att ge förslag på hur anpassningar av hemsidans funktionalitet kan göras för att skapa en mer användarcentrerad hemsida. Genom att använda personifieringsprocesser har sidansgränssnitt presenterats på ett sätt som sätter användaren i fokus och med hjälp av utökade funktionaliteter erbjuder användaren möjlighet till att själv påverka och anpassa innehållet på hemsidan. Resultatet presenterades i form av ett ramverk för interaktiondär fokus låg på avdelningar med produkter och användarprofiler. / The webpage handgjordasaker.se is an online marketplace providing service for the saleof handmade products. This project aims to provide suggestions on how adaptations ofthe site's functionality can be done to create a more user-centered webpage. By usingpersonalization processes, sites interfaces is presented in a way that puts the user infocus and with the help of expanded functionality offers the user the opportunity toself-influence and adapt the content on the website. The results were presented in theform of a framework for the interaction where the focus was on products user profiles.