The development of new solutions incorporating artificial intelligence (AI) within the medical field is an area of great interest. However, access to comprehensive and diverse datasets is restricted due to the sensitive nature of the data. A potential solution to this is to generatesynthetic datasets based on real medical data. Synthetic data could protect the integrity of the subjects while preserving the inherent information necessary for training AI models and be generated in greater quantity than otherwise available. This thesis project aims to generate reliable clinical trial subject data using a generative adversarial network (GAN). The main data set used is a mock clinical trial dataset consisting of multiple subject visits, however an additional data set containing authentic medical data is also used for better insights into the model’s ability to learn underlying relationships. The thesis also investigates training strategies for simulating the temporal dimension and the missing values in the data. The GAN model used is an altered version of the Conditional Tabular GAN (CTGAN)made to be compatible with the preprocessed clinical trial mock data, and multiple model architectures and number of training epochs are examined. The results show great potential for GAN models on clinical trial datasets, especially for real-life data. One model, trained on the authentic dataset, generates near-perfect synthetic data with respect to column distributions and correlation between columns. The results also show that classification models trained on synthetic data and tested on real data have the potential to match the performance of classification models trained on real data. While the synthetic data replicates the missing values, no definitive conclusion can be drawn regarding the temporal characteristics due to the sparsity of the mock dataset and lack of real correlations in it. Although the results are promising, further experiments on authentic datasets with less sparsity are required.

The flight search engine is a technology used in the air travel industry. It allows the traveler to search and book for the best flight options, such as the combination of flights while keeping the best services, options, and price. The computation for a flight search query can be very intensive given its parameters and complexity. The project goal is to predict the flight search queries computation cost for a new flight search engine product when dealing with parameters change and optimizations. The problem of flight search cost prediction is a regression problem. We propose to solve the problem by delimiting the problem based on its business logic and meaning. Our problem has data defined as a graph, which is why we have chosen Graph Neural Network. We have investigated multiple pretraining strategies for the evaluation of node embedding concerning a realworld regression task, including using a line graph for the training. The embeddings are used for downstream regression tasks. Our work is based on some stateoftheart Machine Learning, Deep Learning, and Graph Neural Network methods. We conclude that for some business use cases, the predictions are suitable for production use. In addition, the prediction of tree ensemble boosting methods produces negatives predictions which further degrade the R2 score by 4% because of the business meaning. The Deep Neural Network outperformed the most performing Machine Learning methods by 8% to 12% of R2 score. The Deep Neural Network also outperformed Deep Neural Network with pretrained node embedding from the Graph Neural Network methods by 11% to 17% R2 score. The Deep Neural Network achieved 93%, 81%, and 63% R2 score for each task with increasing difficulty. The training time range from 1 hour for Machine Learning models, 2 to 10 hours for Deep Learning models, and 8 to 24 hours for Deep Learning model for tabular data trained end to end with Graph Neural Network layers. The inference time is around 15 minutes. Finally, we found that using Graph Neural Network for the node regression task does not outperform Deep Neural Network. / Flygsökmotor är en teknik som används inom flygresebranschen. Den gör det möjligt för resenären att söka och boka de bästa flygalternativen, t.ex. kombinationer av flygningar med bästa service, alternativ och pris. Beräkningen av en flygsökning kan vara mycket intensiv med tanke på dess parametrar och komplexitet. Projektets mål är att förutsäga beräkningskostnaden för flygsökfrågor för en ny produkt för flygsökmotor när parametrar ändras och optimeringar görs. Problemet med att förutsäga kostnaderna för flygsökning är ett regressionsproblem. Vi föreslår att man löser problemet genom att avgränsa det utifrån dess affärslogik och innebörd. Vårt problem har data som definieras som en graf, vilket är anledningen till att vi har valt Graph Neural Network. Vi har undersökt flera förträningsstrategier för utvärdering av nodinbäddning när det gäller en regressionsuppgift från den verkliga världen, bland annat genom att använda ett linjediagram för träningen. Inbäddningarna används för regressionsuppgifter i efterföljande led. Vårt arbete bygger på några toppmoderna metoder för maskininlärning, djupinlärning och grafiska neurala nätverk. Vi drar slutsatsen att förutsägelserna är lämpliga för produktionsanvändning i vissa Vi drar slutsatsen att förutsägelserna är lämpliga för produktionsanvändning i vissa fall. Dessutom ger förutsägelserna från trädens ensemble av boostingmetoder negativa förutsägelser som ytterligare försämrar R2poängen med 4% på grund av affärsmässiga betydelser. Deep Neural Network överträffade de mest effektiva metoderna för maskininlärning med 812% av R2poängen. Det djupa neurala nätverket överträffade också det djupa neurala nätverket med förtränad node embedding från metoderna för grafiska neurala nätverk med 11 till 17% av R2poängen. Deep Neural Network uppnådde 93, 81 och 63% R2poäng för varje uppgift med stigande svårighetsgrad. Träningstiden varierar från 1 timme för maskininlärningsmodeller, 2 till 10 timmar för djupinlärningsmodeller och 8 till 24 timmar för djupinlärningsmodeller för tabelldata som tränats från början till slut med grafiska neurala nätverkslager. Inferenstiden är cirka 15 minuter. Slutligen fann vi  att användningen av Graph Neural Network för uppgiften om regression av noder inte överträffar Deep Neural Network.

(Résumé en français) "Exercices de style" est-elle une oeuvre simplement comique et acrobatique ? Sa réception positive mais plutôt superficielle auprès du public semble avoir dissimulé ses aspects plus profonds et plus problématiques. A nos yeux, en revanche, les 99 "Exercices" d'écriture sont tout à fait aptes à éclairer les problèmes essentiels de la création littéraire et de la transmission de l'énoncé qu'est la communication verbale. La structure multi-dimensionnelle du livre, où les "Exercices" s'enchaînent, se complètent et se répondent, nous rend sensibles non seulement aux réseaux intratextuels qu'entretiennent les "Exercices", mais également aux liens intertextuels qui lient cetains d'entre eux et les discours littéraires et non littéraires préexistants. D'autres "Exercices" témoignent du regard autoréflexif de l'écrivain, ceux qui peuvent être considérés comme autoparodie. Par ailleurs, la virtuosité des variations stylistiques exige parfois du lecteur une attention particulière - face à quelques variations hermétiques et presque inintelligibles, on devra recourir à d'autres composants du livre qui serviront de "traductions". La terme "traduction" devra être compris non seulement dans le sens de la transmission de messages entre différentes langues, mais aussi dans le sens de la transposition d'un signifiant dans d'autres signifiants, ou bien de la "réécriture" d'un énoncé, tout en restant dans la même langue. Si le principe des "Exercices de style" est de renouveler à l'infini des exercices d'écrire, ou plutôt de réécrire LE texte original - "qui est d'ailleurs inexistant" -, nous pouvons poser, semble-t-il, que la Littérature est basée sur le même procédé de tâtonnements,auquel le lecteur est enctraîné à participer. (Abstract in English) Is "Exercises in style" just a comic and acrobatic book ? The fact that the readers welcomed it so favourably - but rather superficially - seems to have overshadowed its more serious and problematic aspects. In our opinion indeed, Queneau's ninety-nine writing "Exercises" can clearly shed light on the essential problems of literary creation and utterance transmission, i.e. verbal communication. The book presents an intricate structure:the "Exercises" are linked together, echo each other and complement one another. Through this multidimensional structure, we can see the intratextual networks between the "Exercises" as well as their intertextual relations with pre-existent literary and non-literary discourses. Other "Exercises" show the author's autoreflective, autoparodic attitude. Furthermore, the virtuosity of the stylistic variations sometimes requires particular attention from the reader. To understand some abstruse, sometimes almost unintelligible "Exercises", the reader has to resort to other parts of the book, which will serve as "translation" for these enigmatic passages. The word "translation" here means not only transmission of messages from a language to another, but also transposition of a signifier to other signifiers - in other words, "rewriting" of an utterance in the same language. If the principle of "Exercices in style" is to practice writing endlessly, or rather rewriting of THE original text - "which actually does not exist" - , we can reasonably deduce that Literature is based on the same trial and error process the reader will inevitably take part in.

At the moment, there is a low level of students’ knowledge in natural sciences. The problem is particularly acute in the field of tasks solving. This paper offers actual ways to solve a task by developing a universal technology for solving tasks. This technology will become the basis for training courses aimed at filling students' knowledge gaps and improving their ability to solve tasks. The paper also considers the automation of educational activities of the Network Engineering School. As a result of the work, the current level of schoolchildren’s training was determined and the causes of the problem were identified. The existing methods of solving tasks were analyzed and identified their shortcomings. A universal technology for solving problems using the method of tabular analysis has been developed and its mathematical justification has been given. Compiled methodological manual and published in scientific journals. The business processes of the Network Engineering School were also automated: a distance learning system was introduced into the learning process. The obtained results have already been applied in practice: the developed methodology has become the base of mathematics course at the Network Engineering School, in which classes are already being held. Also, the effectiveness of the methodology was evaluated during a study conducted on the basis of the school Olympiad. In addition, an assessment of the effectiveness of the project on the implementation of a distance learning system was made and it was revealed that the project is profitable. / В настоящий момент наблюдается низкий уровень подготовки школьников по естественно-научным направлениям. Особенно остро ощущается проблема в области решения задач. Данная работа предлагает актуальные пути решения проблемы путем разработки универсальной технологии решения задач. Данная технология станет основой для учебных курсов, направленных восполнить пробелы в знаниях учеников и подтянуть их способности решать задачи. Также в работе рассматривается автоматизация учебной деятельности Сетевой инженерно-технической школы. В результате работы был определен текущий уровень подготовки школьников и определены причины возникновения проблемы. Проанализированы существующие методики решения задач и выявлены их недостатки. Разработана универсальная технология решения задач методом табличного анализа и дано ее математическое обоснование. Составлено методическое пособие и опубликовано в научных изданиях. Также были автоматизированы бизнес-процессы Сетевой инженерно-технической школы: в процесс обучения была внедрена система дистанционного обучения. Полученные результаты уже применены на практике: разработанная методика стала основной курса по математике в Сетевой инженерно-технической школе, по которому уже проводятся занятия. Также эффективность методики была оценена во время исследования, проведенного на базе школьной Олимпиады. А также была произведена оценка эффективности проекта по внедрению системы дистанционного обучения и выявлено, что проект является прибыльным.