The prediction of polypharmacy side effects is crucial to reduce the mortality and morbidity of patients suffering from complex diseases. However, its experimental prediction is unfeasible due to the many possible drug combinations, leaving in silico tools as the most promising way of addressing this problem. This thesis improves the performance and robustness of a state-of-the-art graph convolutional network designed to predict polypharmacy side effects, by feeding it with complexity properties of the drug-protein network. The modifications also involve the creation of a direct pipeline to reproduce the results and test it with different datasets. / För att minska dödligheten och sjukligheten hos patienter som lider av komplexa sjukdomar är det avgörande att kunna förutsäga biverkningar från polyfarmaci. Att experimentellt förutsäga biverkningarna är dock ogenomförbart på grund av det stora antalet möjliga läkemedelskombinationer, vilket lämnar in silico-verktyg som det mest lovande sättet att lösa detta problem. Detta arbete förbättrar prestandan och robustheten av ett av det senaste grafiska faltningsnätverken som är utformat för att förutsäga biverkningar från polyfarmaci, genom att mata det med läkemedel-protein-nätverkets komplexitetsegenskaper. Ändringarna involverar också skapandet av en direkt pipeline för att återge resultaten och testa den med olika dataset.

[pt] Com o crescente emprego de inteligência artificial na moderação de conteúdo do Instagram, a empresa é cada vez mais criticada pelos efeitos discriminatórios de seus sistemas algorítmicos, que parecem revelar uma hierarquia de humanidades dentro da rede social. Ainda que a autoexposição dos usuários seja a fonte primária de dados usados pelo Instagram em sua geração de lucros, alguns corpos detêm plena e irrestrita visibilidade enquanto outros são desproporcionalmente ocultados pela moderação automatizada. A moderação algorítmica na rede social integra um novo tipo de governo das condutas, baseado em dados e correlações estatísticas, que assume traços específicos com relação a usuárias mulheres. Trata-se de um poder que, ocultado sob discursos tecnoutópicos e corporativistas do Instagram, não se reconhece como poder e se torna mais dificilmente detectável. Existe, contudo, a possibilidade de resistência a tal governo das condutas. Enquanto grande parte dos trabalhos acadêmicos sobre resistência online é focada na resistência ora a violências interpessoais entre usuários ora a práticas e discursos sexistas, é escassa a literatura sobre a resistência ao poder dos sistemas algorítmicos de moderação de conteúdo. Esta pesquisa pretende contribuir com o debate sobre o tema, explorando como técnica de resistência ou autodefesa digital, a autoexposição das usuárias do Instagram voltada à denúncia do poder algorítmico da rede social, em especial aos silenciamentos e discriminações por ele perpetuados. / [en] With the increasing use of artificial intelligence in Instagram s content moderation, the company is increasingly criticized for the discriminatory effects of its algorithmic systems, which seem to reveal a hierarchy of humanities within the social network. Even though users self-exposure is the primary source of data used by Instagram in its profit generation, some bodies hold full and unrestricted visibility while others are disproportionately hidden by automated moderation. Algorithmic moderation on the social network integrates a new kind of government of conducts, based on data and statistical correlations, that takes on specific traits with regard to female users. This is a power that, hidden under Instagram s technoutopian and corporatist discourses, does not recognize itself as power and becomes more difficult to detect. There is, however, the possibility of resistance to such government of conduct. While much of the academic work on online resistance is focused on resistance either to interpersonal violence between users or to sexist practices and discourses, the literature on resistance to the power of algorithmic content moderation systems is scarce. This research intends to contribute to the debate on the theme, exploring as a resistance technique or digital selfdefense, the self-exposure of female Instagram users aimed at denouncing the algorithmic power of the social network, especially the silences and discriminations perpetuated by it.

<p>This thesis concerns itself with drawing out some interesting connections between the fields of group theory and formal language theory. Given a group with a finite set of generators, it is natural to consider the set of generators and their inverses as an alphabet. We can then consider formal languages such that every group element has at least one representative in the language. We examine what the structure of the language can tell us about group theoretic properties, focusing on the word problem, automatic structures on groups, and generalizations of automatic structures. Finally we prove new results concerning applications of languages of nested words for studying the word problem.</p> / Master of Science (MSc)

Background: The increasing reliance on Artificial Intelligence (AI) in personalized marketing has reshaped consumer interactions in the digital era. With technological advancements, there is a growing need to explore how AI-driven personalization influences consumer behavior, particularly in satisfaction, loyalty, and ethical considerations. Purpose: This thesis investigates the impact of AI-driven personalized marketing on consumer perceptions, attitudes, and behaviors. It aims to understand how trust and ethical considerations such as data privacy and algorithmic bias influence consumer responses and engagement with personalized marketing. Method: Employing a quantitative approach, this study integrates quantitative analysis from surveys conducted on 100 participants. This method provides a comprehensive understanding of the implications of AI-driven personalization. Statistical tools like SPSS were used for data analysis, ensuring rigorous examination of the collected data. Conclusion: The findings reveal a nuanced response to personalized marketing. While AI-driven personalization can enhance consumer engagement and satisfaction, transparency and ethical considerations are critical in securing consumer trust and loyalty. The study underscores the importance of ethical marketing practices and the need for continuous adaptation to technological advancements to align with consumer expectations and ethical standards. This research contributes to academic discussions on personalized marketing and offers strategic insights for integrating technological advancements with consumer-centric approaches in marketing practices.

International Telemetering Conference Proceedings / October 17-20, 1994 / Town & Country Hotel and Conference Center, San Diego, California / Use of top-down design principles and standard interface techniques provides the basis for a global telemetry data collection, analysis, and satellite control network with a high degree of survivability via use of distributed architecture. Use of Commercial Off-The-Shelf (COTS) hardware and software minimizes costs and provides for easy expansion and adaption to new satellite constellations. Adaptive techniques and low cost multiplexers provide for graceful system wide degradation and flexible data distribution.

The delegating of thought, memory and action outside of the human body, inseparable from the process of individuation and identity formation, and the following implications for music establish an underlying theme of this text. It is a reflection on the process of "supplementation," of prosthetization or exteriorization in the recent and contemporary milieu of music making, in which nothing is any longer immediately at hand, where everything is found mediated and instrumentalized, technicized, unbalanced.

This diploma thesis deals with the development of computational and algorithmic thinking in a non-formal education setting, specifically in the conditions of a summer camp activity, which is one of the forms of non-formal education. The thesis verifies by the pedagogical experiment whether it is possible to develop the computational thinking of camp participants, when unplugged activities are included in the programme content. The theoretical part defines the concepts of computational and algorithmic thinking. It also includes property of unplugged methodology. The practical part evaluates the pedagogical field experiment for which 3 sets of unplugged activities and 2 sets of test tools were created. The experiment took place during two summer camp for children from 9 to 18 years old, who were divided into two age groups for testing purposes. In pursuance of the experiment it was found out that the conditions of summer camp activity allow the inclusion of unplugged activities, which represent a method enabling the development of the computational thinking of the camp participants. Though it emerged that these activities have a bearing only on the participants who are more active during these activities and who are more interested in them. KEYWORDS unplugged activities, computational thinking,...

Cílem této diplomové práce je představení zcela nového způsobu vytváření i distribuování zpráv a obsahu, kdy se do tohoto jinak tradičně "lidského odvětví" začínají v poslední době promítat automatizované a chytré počítačové systémy využívající umělou inteligenci. Hodlám vysvětlit, co tento pojem znamená s ohledem na mediální prostor, představit jeho poměrně krátkou historii a popsat, jak dochází k jeho zavádění v žurnalistické praxi. Jelikož se jedná o stále vyvíjející se část možná budoucí novinářské práce, která je zatím v začátcích, nepracují s ní zatím ve větším tuzemská média, i když postupně začínají experimentovat například se strojovou žurnalistikou, tedy předstupněm plně automatizované umělé inteligence. Výzkum je proto zaměřen na vývoj a používání umělé inteligence tiskovými agenturami Reuters a AP, které s těmito nástroji už začaly pracovat a na jejichž případu hodlám ukázat možný budoucí nástup i praxi. V závěru zhodnotím největší výhody i nejdiskutovanější etické problémy fungování umělé inteligence v žurnalistice.

Nous considérons des jeux de congestion sur des graphes. Dans les jeux non-atomiques, nous considérons un ensemble de joueurs infinitésimaux. Chaque joueur veut aller d'un sommet à un autre en choisissant une route de coût minimal. Le coût de chaque route dépend du nombre de joueur la choisissant. Dans les jeux atomiques divisibles, nous considérons un ensemble de joueurs ayant chacun une demande à transférer d'un sommet à un autre, en la subdivisant éventuellement sur plusieurs routes. Dans ces jeux, un équilibre de Nash est atteint lorsque chaque joueur a choisi une stratégie de coût minimal. L'existence d'un équilibre de Nash est assurée sous de faibles hypothèses. Les principaux sujets sont l'unicité, le calcul, l'efficacité et la sensibilité de l'équilibre de Nash. De nombreux résultats sont connus dans le cas où les joueurs sont tous impactés de la même façon par la congestion. Le but de cette thèse est de généraliser ces résultats au cas où les joueurs ont des fonctions de coût différentes. Nous obtenons des résultats sur l'unicité de l'équilibre dans les jeux non-atomiques. Nous donnons deux algorithmes capables de calculer un équilibre dans les jeux non-atomiques lorsque les fonctions de coût sont affines. Nous obtenons une borne sur le prix de l'anarchie pour certains jeux atomiques divisibles et prouvons qu'il n'est pas borné en général, même lorsque les fonctions sont affines. Enfin, nous prouvons des résultats sur la sensibilité de l'équilibre par rapport à la demande dans les jeux atomiques divisibles / We consider congestion games on graphs. In nonatomic games, we are given a set of infinitesimal players. Each player wants to go from one vertex to another by taking a route of minimal cost, the cost of a route depending on the number of players using it. In atomic splittable games, we are given a set of players with a non-negligible demand. Each player wants to ship his demand from one vertex to another by dividing it among different routes. In these games, we reach a Nash equilibrium when every player has chosen a minimal-cost strategy. The existence of a Nash equilibrium is ensured under mild conditions. The main issues are the uniqueness, the computation, the efficiency and the sensitivity of the Nash equilibrium. Many results are known in the specific case where all players are impacted in the same way by the congestion. The goal of this thesis is to generalize these results in the case where we allow player-specific cost functions. We obtain results on uniqueness of the equilibrium in nonatomic games. We give two algorithms able to compute a Nash equilibrium in nonatomic games when the cost functions are affine. We find a bound on the price of anarchy for some atomic splittable games, and prove that it is unbounded in general, even when the cost functions are affine. Finally we find results on the sensitivity of the equilibrium to the demand in atomic splittable games

Nessa tese consideramos problemas de alocação e precificação de itens, onde temos um conjunto de itens e um conjunto de compradores interessados em tais itens. Nosso objetivo é escolher uma alocação de itens a compradores juntamente com uma precificação para tais itens para maximizar o lucro obtido, considerando o valor máximo que um comprador está disposto a pagar por um determinado item. Em particular, focamos em três problemas: o Problema da Compra Máxima, o Problema da Precificação Livre de Inveja e o Leilão de Anúncios de Segundo Preço. O Problema da Compra Máxima e o Problema da Precificação Livre de Inveja modelam o problema que empresas que vendem produtos ou serviços enfrentam na realidade, onde é necessário escolher corretamente os preços dos produtos ou serviços disponíveis para os clientes para obter um lucro interessante. Já o Leilão de Anúncios de Segundo Preço modela o problema enfrentado por empresas donas de ferramentas de busca que desejam vender espaço para anunciantes nos resultados das buscas dos usuários. Ambas as questões, tanto a precificação de produtos e serviços quanto a alocação de anunciantes em resultados de buscas, são de grande relevância econômica e, portanto, são interessantes de serem atacadas dos pontos de vista teórico e prático. Nosso foco nesse trabalho é considerar algoritmos de aproximação e algoritmos de programação inteira mista para os problemas mencionados, apresentando novos resultados superiores àqueles conhecidos previamente na literatura, bem como determinar a complexidade computacional destes problemas ou de alguns de seus casos particulares de interesse. / In this thesis we consider allocation and pricing problems, where we have a set of items and a set of consumers interested in such items. Our objective is to choose an allocation of items to consumers, considering the maximum value a consumer is willing to pay in a specific item. In particular, we focus in three problems: the Max-Buying Problem, the Envy-Free Pricing Problem and the Second-Price Ad Auction. The Max-Buying Problem and the Envy-Free Pricing Problem model a problem faced in reality by companies that sell products or services, where it is necessary to correctly choose the price of the products or services available to clients in order to obtain an interesting profit. The Second-Price Ad Auction models the problem faced by companies that own search engines and desire to sell space for advertisers in the search results of the users. Both questions, the pricing of items and services and the allocation of advertisers in search results are of great economical relevance and, for this, are interesting to be attacked from a theoretical and a practical perspective. Our focus in this work is to consider approximation algorithms and mixed integer programming algorithms for the aforementioned problems, presenting new results superior than the previously known in the literature, as well as to determine the computational complexity of such problems or some of their interesting particular cases.