Current artificial intelligence tools, including machine learning and large language models, display inabilities to interact with medical data in real time and raise privacy concerns related to user data management. This study illustrates the development of a system prototype using LangChain4j, which is an open-source project offering a multitude of AI-tools, including embedding tools, retrieval-augmented generation, and unified API:s for large language model providers. It was utilized to process medical data from a Neo4j database and enabled real-time interaction for that data. All content generation was generated locally to address privacy concerns, while using Apache Kafka for data distribution. The system prototype was evaluated by response time, resource consumption and accuracy assessment. Among the models assessed, LLaMA 3 emerged as the top performer in accuracy, successfully identifying 42.87% of all attributes with a correctness rate of 89.81%. Meanwhile, Phi3 exhibited superior outcomes in both resource consumption and response time. The embedding process, while enabling the selection of visible data, imposed limitations on general usability. In summary, this thesis advances data interaction using AI by developing a prototype that enables real-time interaction with medical data. It achieves high accuracy and efficient resource utilization while addressing limitations in current AI tools related to real-time processing and privacy concerns. / Nuvarande verktyg för artificiell intelligens, inklusive maskininlärning och stora språkmodeller, visar oförmåga att interagera med medicinska data i realtid och väcker integritetsproblem relaterade till hantering av användardata. Denna studie illustrerar utvecklingen av ett systemprototyp med LangChain4j, ett open-source-projekt som erbjuder en mängd AI-verktyg, inklusive inbäddningsverktyg, retrieval-augmented generation och enhetliga API för leverantörer av stora språkmodeller. Det användes för att bearbeta medicinska data från en Neo4j-databas och möjliggjorde realtidsinteraktion för dessa data. All innehållsgenerering skedde lokalt med Apache Kafka för datadistribution. Systemprototypen utvärderades utifrån svarstid, resursförbrukning och noggrannhetsbedömning. Bland de modeller som utvärderades visade sig LLaMA 3 vara den bästa presteraren i noggrannhet, och identifierade framgångsrikt 42,87 % av alla attribut med en korrekthet på 89,81 %. Samtidigt visade Phi3 överlägsna resultat både i resursförbrukning och svarstid. Inbäddningsprocessen, medan den möjliggjorde valet av synliga data, innebar begränsningar för allmän användbarhet. Sammanfattningsvis förbättrar denna avhandling datainteraktion med AI genom att utveckla en prototyp som möjliggör realtidsinteraktion med medicinska data. Den uppnår hög noggrannhet och effektiv resursanvändning samtidigt som den adresserar begränsningar i nuvarande AI-verktyg relaterade till realtidsbearbetning och integritetsproblem.

Multi-object tracking is a classic engineering problem wherein a system must keep track of the identities of a set of a priori unknown objects through a sequence, for example video. Perfect execution of this task would mean no spurious or missed detections or identities, neither swapped identities. To measure performance of tracking systems, the Higher Order Tracking Accuracy metric is often used, which takes into account both detection and association accuracy. Prior work in monocular vision-based multi-object tracking has integrated deep learning to various degrees, with deep learning based detectors and visual feature extractors being commonplace alongside motion models of varying complexities. These methods have historically combined the usage of position and appearance in their association stage using hand-crafted heuristics, featuring increasingly complex algorithms to achieve higher performance tracking. With an interest in simplifying tracking algorithms, we turn to the field of representation learning. Presenting a novel method using a Joint-Embedding Predictive Architecture, trained through a contrastive objective, we learn object feature embeddings initialized by detections from a pre-trained detector. The results are features that fuse both positional and visual features. Comparing the performance of our method on the complex DanceTrack and relatively simpler MOT17 datasets to that of the most performant heuristic-based alternative, Deep OC-SORT, we see a significant improvement of 66.1 HOTA compared to the 61.3 HOTA of Deep OC-SORT on DanceTrack. On MOT17, which features less complex motion and less training data, heuristics-based methods outperform the proposed and prior learned tracking methods. While the method lags behind the state of the art in complex scenes, which follows the tracking-by-attention paradigm, it presents a novel approach and brings with it a new avenue of possible research. / Spårning av multipla objekt är ett typiskt ingenjörsproblem där ett system måste hålla reda på identiteterna hos en uppsättning på förhand okända objekt genom en sekvens, till exempel video. Att perfekt utföra denna uppgift skulle innebära inga felaktiga eller missade detektioner eller identiteter, inte heller utbytta identiteter. För att mäta prestanda hos spårningssystem används ofta metriken HOTA, som tar hänsyn till både detektions- och associationsnoggrannhet. Tidigare arbete inom monokulär vision-baserad flerobjektsspårning har integrerat djupinlärning i olika grad, med detektorer baserade på djupinlärning och visuella funktionsutdragare som är vanliga tillsammans med rörelsemodeller av varierande komplexitet. Dessa metoder har historiskt kombinerat användningen av position och utseende i deras associationsfas med hjälp av handgjorda heuristiker, med alltmer komplexa algoritmer för att uppnå högre prestanda i spårningen. Med ett intresse för att förenkla spårningsalgoritmer, vänder vi oss till fältet för representationsinlärning. Vi presenterar en ny metod som använder en prediktiv arkitektur med gemensam inbäddning, tränad genom ett kontrastivt mål, där vi lär oss objekt representationer initierade av detektioner från en förtränad detektor. Resultatet är en funktion som sammansmälter både position och visuel information. När vi jämför vår metod på det komplexa DanceTrack och det relativt enklare MOT17-datasetet med det mest presterande heuristikbaserade alternativet, Deep OC-SORT, ser vi en betydande förbättring på 66,1 HOTA jämfört med 61,3 HOTA för Deep OC-SORT på DanceTrack. På MOT17, som har mindre komplex rörelse och mindre träningsdata, presterar heuristikbaserade metoder bättre än den föreslagna och tidigare lärande spårningsmetoderna. Även om metoden ligger efter den senaste utvecklingen i komplexa scener, som följer paradigm för spårning-genom-uppmärksamhet, presenterar den ett nytt tillvägagångssätt och för med sig möjligheter för ny forskning.

<p dir="ltr">This research involves numerous network intrusion techniques through novel applications of graph representation learning and image representation learning. The methods are tested on multiple publicly available network flow datasets.</p>

Background. New ways of modeling abstract concepts have been enabled due to the recent boom in generative machine learning brought on by transformer architecture. By modeling abstract concepts within high-dimensional vectors their semantic meaning can be inferred and compared, which allows for methods such as embedding-based retrieval and the groundwork for a retrieval-augmented generation. Large language models can augment their parametric generative capabilities by introducing non-parametric information through retrieval processes. Objectives. Previous studies have explored different uses of embedding-based retrieval and retrieval-augmented generation, and this study examines the impact of these methods when used as an aid to support technicians. Methods. By developing and deploying a proof-of-concept system using embedding-based retrieval and retrieval-augmented generation to the Södra ITs service desk, the thesis could monitor system performance. Introducing a system to the service desk that generates instructional solutions to the support tickets and presenting them to the technician. The thesis investigates both systems' perceived performance based on the participating IT technician's input along with the retention of generated solutions and the quality of the solutions. Results. With 75.4% of the systems generated solutions being classified as reasonable solutions to ticket problems the system was deployed to the service desk. After an evaluation period where the technicians had been working with the system, it was shown that the solutions had a retention rate of 38.4%. These results were validated by a survey conducted at the service desk where the inputs were gathered from the technicians, showing a great deal degree of user engagement but a varying opinion on the system's helpfulness. Conclusions. Despite the varying degrees of opinion on the usefulness of the system among the technicians the numbers from the production test show that a significant amount of tickets were solved with the help of the system. Still, there's a huge dependency on seamless integration with the technicians and ticket quality from the requester. / Bakgrund. Nya sätt att modellera abstrakta begrepp har möjliggjorts tack vare den senaste tidens tillväxt inom generativ maskininlärning tack vare transformatorarkitekturen. Genom att modellera abstrakta begrepp i högdimensionella vektorer kan deras semantiska innebörd tolkas och jämföras, vilket möjliggör metoder som inbäddningsbaserad hämtning och grunden för en hämtningsförstärkt generation. Stora språkmodeller kan utvidga sina parametriska generativa förmågor genom att införa icke-parametrisk information genom hämtningsprocesser. Syfte. Tidigare studier har behandlat olika användningsområden för inbäddningsbaserad hämtning och hämtningsförstärkt generering, och i det här examensarbetet undersöks vilken inverkan dessa metoder har när de används som ett hjälpmedel för supporttekniker. Metod. Genom att utveckla och driftsätta ett prototypsystem som använder inbäddningsbaserad hämtning och hämtningsförstärkt generering till Södra ITs servicedesk, kunde examensarbetet övervaka systemets prestanda. Detta genom att införa ett system i servicedesken som genererar instruktionslösningar till supportärendena och presentera dem för teknikern. Examensarbetet undersöker både systemens upplevda prestanda baserat på den deltagande IT-teknikerns synpunkter tillsammans med kvarhållandet av genererade lösningar och kvaliteten på lösningarna. Resultat. Då 75,4% av de systemgenererade lösningarna klassificerades som rimliga för problemen i ärendena driftsattes systemet i servicedesken. Efter en utvärderingsperiod där teknikerna hade arbetat med systemet visade det sig att lösningarna hade en kvarhållningsgrad på 38,4%. Dessa resultat validerades av en undersökning som utförts vid servicedesken där synpunkter samlades in från teknikerna, vilket visade på en hög grad av användarengagemang men en varierande syn på systemets användbarhet. Slutsatser. Trots de varierande synpunkterna på systemets användbarhet bland teknikerna visar siffrorna från produktionstestningen att en betydande mängd ärenden löstes med hjälp av systemet. Dock är man fortfarande mycket beroende av en smidig integration med teknikerna och en god kvalitet på ärendena från beställaren.

L'analogie, c'est-à-dire une correspondance entre deux entités, est considérée une capacité de raisonnement importante. L'analogie proportionnelle, écrite $a:b::c:d$ et qui se lit ``$a$ est à $b$ ce que $c$ est à $d$'', en est un cas particulier où la correspondance tient de par la relation entre les éléments de deux paires d'objets. Le mémoire évalue certaines méthodes issues de l'usage de représentations distributionnelles vectorielles dans la résolution d'analogies proportionnelles verbales et les mène à leur prolongement naturel, la phrase. Nous ciblons la compétence de modèles de langue et des représentations qui peuvent en être extraites à la résolution d'analogies proportionnelles formées sur la base de relations syntaxiques, sémantiques, ou de connaissance encyclopédique. Peu d'ensembles de données existent pour les analogies de phrase et sinon comprennent pour la plupart des analogies au niveau de la forme, composées de phrases construites à partir de gabarits, ou bien variant peu dans les relations sémantiques qui tiennent entre les phrases. Nous construisons donc un ensemble de données contenant des phrases en paires relationnelles qui nous permet de construire des analogies en appariant deux paires. Nous essayons différentes variations de méthodes qui comportent un objectif de recouvrement par un modèle vectoriel. D'autres méthodes de résolution d'analogies proportionnelles sont explorées par voie de génération de texte. Nous expérimentons par le peaufinement du modèle de langue Flan-T5, pré-entraîné sur des paires instruction-réponse, sur nos analogies par une tâche séquence à séquence, ainsi que par l'incitation avec peu d'exemples en utilisant des versions de ce modèle en variant la capacité jusque dans la gamme des milliards de paramètres. En somme, la performance observée est faible pour toutes les tâches. Nous concluons, de l'utilisation de plongements de phrase, quelques mises en garde similaires à celles que l'on trouve avec la résolution d'analogies verbales par plongements lexicaux. Nos expérimentations génératives démontrent l'importance de données à la fois de bonne qualité et de bonne quantité, ainsi que le potentiel de l'apprentissage en contexte. Nous ajoutons à cela un aperçu qualitatif de la disparité entre l'habileté de modèles probabilistes entraînés pour prédire, à partir d'une instruction, la séquence correcte, et celle d'un modèle peaufiné par la méthode d'apprentissage par renforcement avec commentaires humains, à savoir ChatGPT. / Analogy, the correspondence between two things, has been hailed as an important reasoning capability. Proportional analogy, denoted $a:b::c:d$, read ``$a$ is to $b$ as $c$ is to $d$'' is a special case of this where a correspondence is made in the relation that holds between the elements of two pairs. This thesis evaluates methods originating in the recent use of distributional vector representations for solving four-part word analogies, bringing them to their natural extension, sentences. Few datasets of proportional sentence analogies exist, typically comprising purely formal analogies or sentences constructed by templates, and where semantic relations are typically limited in the variety we would hope to capture. Thus, for the purposes of our experiments, we curate a dataset of pairs of sentences for which a given relation holds and from which analogies can be constructed by matching pairs within a relation together. We target the analogy-solving ability of language models and representations derived therefrom, specifically as regards proportional sentence analogies formed on the basis of syntax, semantics, or encyclopedic knowledge. Different variations on previous methods are explored, all based on retrieval of the solution in a vector space model. Other methods of solving proportional sentence analogies by generation are attempted. We experiment with finetuning the instruction-trained Flan-T5 language model on sentence analogies as a sequence-to-sequence task, as well as prompting model checkpoints up into the billion-parameter range with few-shot examples. Overall performance at the task is poor in both settings. We find that similar caveats which apply to analogical reasoning with word vectors apply to sentence embeddings as well. Our generative experiments show the importance of data of suitable quality and quantity, as well the potential of in-context learning. Some qualitative insights are shown as to the disparity in task ability of instruction-trained probabilistic language models and one finetuned by reinforcement learning with human feedback, namely ChatGPT.

Interactions are formed and shaped differently from culture to culture. This thesis focuses on this phenomenon and in the first part deals with (from a theoretical view) the question, how the interplay between language-use and culture can be described. A cross-disciplinary approach within the scope of cross-cultural communication research is developed as is a semiotic concept, based on mainly linguistic, interactional and anthropological theories and methods. In order to describe different culture-specific interactional styles the semiotic field called embedding culture is outlined as an important resource for participants organizing talk-in-interactions. In the second part of this book the relevance of this approach is applied and tested. With the aid of various video recordings of ‘lunch-talks’ among colleagues at a restaurant in Sweden and Germany, this specific type of action and their cultural patterns are analyzed in order to describe cultural-specific styles in face-to-face-interaction. The methodological problem of recorded interaction is pointed out and the role of the camera highlighted. As the analyzed data is mainly based on interactions between men also gender-aspects are discussed. Furthermore, the scenario, time aspects and the customs and rituals of interactions at table are taken into account as relevant features of the embedding culture, all going into a culture-specific style of interaction. The results of the empirical study are, finally, correlated with other, not interaction-based analyses in the field of cross-cultural communication, and the specific national-cultural dimensions are critically discussed. / Interaktioner utformas och gestaltas olika från kultur till kultur. Avhandlingen fokuserar på denna aspekt och behandlar först ur ett teoretiskt perspektiv hur detta fenomen kan beskrivas. Med en tvärvetenskaplig ansats inom ramen av den s.k. interkulturella kommunikationsforskningen (cross-cultural communication) kombineras kulturanalytiska med lingvistiska och samtalsanalytiska teorier och metoder. Ett semiotiskt koncept utarbetas med hjälp av vilka olika kulturspecifika interaktionsstilar kan beskrivas. Det semiotiska fältet som koncipierats kallas för Einbettungskultur (’inbäddningskulturen’). I den andra delen används och prövas detta koncept. Med hjälp av ett flertal videoinspelade lunchsamtal i Sverige och Tyskland analyseras konkreta exempel på kommunikativ praxis och deras kulturella mönster i jämförbara interaktionssituationer. Det beskrivs hur den kulturspecifika interaktionsstilen vid lunchen bland svenska arbetskollegor kan skilja sig från det tyskspråkiga sättet att gestalta samma interaktionstyp. Inspelningssituationen problematiseras och kamerans roll i interaktionen diskuteras. Eftersom analyserna huvudsakligen baseras på samtal mellan män tas genusperspektivet upp. Vid sidan av själva samtalen analyseras även scenariot, tidsaspekten och ritualiseringen av interaktionen vid bordet som tre aspekter av inbäddningskulturen som tillsammans bidrar till en kulturspecifik interaktionsstil. Resultaten av den kulturella analysen som tar sin utgångspunkt i en konkret och jämförbar interaktionssituation i olika kulturer relateras avslutningsvis till andra, icke-interaktionsbaserade analyser inom den interkulturella kommunikationsforskningen och nationalkulturella beskrivningar problematiseras.

This Master Thesis deals with the drawing algorithms of graphs known from the mathematical theory. These algorithms deals with an appropriate distribution of the graph vertices in order to obtain the most clear and readable graphs for human readers. The main objective of this work was also to implement the drawing algorithm in the application that would allow to edit the graph. This work deals also with graphs representation in computers.

Multiobjective Stochastic Linear Programming is a relevant topic. As a matter of fact, many real life problems ranging from portfolio selection to water resource management may be cast into this framework. There are severe limitations in objectivity in this field due to the simultaneous presence of randomness and conflicting goals. In such a turbulent environment, the mainstay of rational choice does not hold and it is virtually impossible to provide a truly scientific foundation for an optimal decision. In this thesis, we resort to the bounded rationality and chance-constrained principles to define satisficing solutions for Multiobjective Stochastic Linear Programming problems. These solutions are then characterized for the cases of normal, exponential, chi-squared and gamma distributions. Ways for singling out such solutions are discussed and numerical examples provided for the sake of illustration. Extension to the case of fuzzy random coefficients is also carried out. / Decision Sciences

Nowadays, the near-horizon regions of black holes have enjoyed great attention thanks to their role in the popular AdS/CFT correspondence and their specific geometry suitable for formulations of uniqueness theorems in higher dimensions. A strictly general-relativistic point of view reveals also many interesting phenomena taking place near black-hole horizons. Our aim was to investigate how horizon multiplicity affects near-horizon geometry, geodesical distance, radial motion of photons and massive, charged particles, and also the possibility of collision processes leading to unbound collision energies near the horizon. We chose the Reissner-Nordström-de Sitter metric, which, on the one hand, is simple thanks to being static and spherically symmetric but which, on the other hand, is rich enough to enable the existence of up to a doubly degenerate ultra-extreme horizon. After discussing the physical feasibility of the near-horizon limit, we applied it to single, double, and triple horizons, their near-horizon geometries, and local collision processes. We found continuous coordinate systems covering all types of horizons and analytic solutions for motion of radial photons and special or critical, massive, charged particles in their vicinity. We addressed particle collisions in the immediate vicinity of horizons...

Les principaux objets étudiés dans cette thèse sont les équations décrivant le morphisme de groupe sur une variété abélienne, plongée dans un espace projectif, et leurs applications en cryptographie. Notons g sa dimension et k son corps de définition. Ce mémoire est composé de deux parties. La première porte sur l'étude des courbes d'Edwards, un modèle pour les courbes elliptiques possédant un sous-groupe de points k-rationnels cyclique d'ordre 4, connues en cryptographie pour l'efficacité de leur loi d'addition et la possibilité qu'elle soit définie pour toute paire de points k-rationnels (loi d'addition k-complète). Nous en donnons une interprétation géométrique et en déduisons des formules explicites pour le calcul du couplage de Tate réduit sur courbes d'Edwards tordues, dont l'efficacité rivalise avec les modèles elliptiques couramment utilisés. Cette partie se conclut par la génération, spécifique au calcul de couplages, de courbes d'Edwards dont les tailles correspondent aux standards cryptographiques actuellement en vigueur. Dans la seconde partie nous nous intéressons à la notion de complétude introduite ci-dessus. Cette propriété est cryptographiquement importante car elle permet d'éviter des attaques physiques, comme les attaques par canaux cachés, sur des cryptosystèmes basés sur les courbes elliptiques ou hyperelliptiques. Un précédent travail de Lange et Ruppert, basé sur la cohomologie des fibrés en droite, permet une approche théorique des lois d'addition. Nous présentons trois résultats importants : tout d'abord nous généralisons un résultat de Bosma et Lenstra en démontrant que le morphisme de groupe ne peut être décrit par strictement moins de g+1 lois d'addition sur la clôture algébrique de k. Ensuite nous démontrons que si le groupe de Galois absolu de k est infini, alors toute variété abélienne peut être plongée dans un espace projectif de manière à ce qu'il existe une loi d'addition k-complète. De plus, l'utilisation des variétés abéliennes nous limitant à celles de dimension un ou deux, nous démontrons qu'une telle loi existe pour leur plongement projectif usuel. Finalement, nous développons un algorithme, basé sur la théorie des fonctions thêta, calculant celle-ci dans P^15 sur la jacobienne d'une courbe de genre deux donnée par sa forme de Rosenhain. Il est désormais intégré au package AVIsogenies de Magma. / The main objects we study in this PhD thesis are the equations describing the group morphism on an abelian variety, embedded in a projective space, and their applications in cryptograhy. We denote by g its dimension and k its field of definition. This thesis is built in two parts. The first one is concerned by the study of Edwards curves, a model for elliptic curves having a cyclic subgroup of k-rational points of order 4, known in cryptography for the efficiency of their addition law and the fact that it can be defined for any couple of k-rational points (k-complete addition law). We give the corresponding geometric interpretation and deduce explicit formulae to calculate the reduced Tate pairing on twisted Edwards curves, whose efficiency compete with currently used elliptic models. The part ends with the generation, specific to pairing computation, of Edwards curves with today's cryptographic standard sizes. In the second part, we are interested in the notion of completeness introduced above. This property is cryptographically significant, indeed it permits to avoid physical attacks as side channel attacks, on elliptic -- or hyperelliptic -- curves cryptosystems. A preceeding work of Lange and Ruppert, based on cohomology of line bundles, brings a theoretic approach of addition laws. We present three important results: first of all we generalize a result of Bosma and Lenstra by proving that the group morphism can not be described by less than g+1 addition laws on the algebraic closure of k. Next, we prove that if the absolute Galois group of k is infinite, then any abelian variety can be projectively embedded together with a k-complete addition law. Moreover, a cryptographic use of abelian varieties restricting us to the dimension one and two cases, we prove that such a law exists for their classical projective embedding. Finally, we develop an algorithm, based on the theory of theta functions, computing this addition law in P^15 on the Jacobian of a genus two curve given in Rosenhain form. It is now included in AVIsogenies, a Magma package.