Uma série de equipamentos de eletrônica de potência é ligada à rede de corrente alternada através de transformadores, que exercem as funções de isolamento galvânico, elevação/abaixamento de tensão, etc. Estes transformadores podem sofrer saturação em condições transitórias (inrush) ou quando o conversor ligado a ele impõe valor médio não nulo de tensão. O problema é normalmente sanado nas malhas de controle do conversor, no entanto o ajuste destas malhas em ambiente de simulação computacional requer um modelo que represente adequadamente a característica de saturação do núcleo ferromagnético. Este trabalho apresenta um modelo de simulação computacional adequado para transformadores operando em saturação intensa, visando aplicações em projetos de equipamentos de eletrônica de potência com transformador monofásico alimentado por inversor. Além disso, é demonstrado um método experimental de obtenção da característica de saturação do núcleo, sem necessidade das suas características construtivas. Resultados de simulação obtidos são comparados com resultados experimentais para validação do modelo e método. / Several power electronics based equipments are connected to the alternating current network through transformers, which perform galvanic insulation, voltage increasing/lowering, etc. These transformers may experience saturation under transitory conditions (inrush), or when its converter imposes non-zero average voltage. Such problem is commonly treated in the converter\'s control loops, but the tuning of these loops in a computer simulation environment requires a transformer model that represents adequately the ferromagnetic core saturation effect. The current work proposes a computer simulation model that adequately describes the operation of the transformer at deep saturation, suitable for design of power electronics applications with single phase transformers connected to power inverter. Additionally, an experimental method for determining its core saturation characteristic is demonstrated. This method does not require previous knowledge of the core\'s constructive parameters. Simulation and experimental results are presented to confirm the validity of the model and the method.

Uma série de equipamentos de eletrônica de potência é ligada à rede de corrente alternada através de transformadores, que exercem as funções de isolamento galvânico, elevação/abaixamento de tensão, etc. Estes transformadores podem sofrer saturação em condições transitórias (inrush) ou quando o conversor ligado a ele impõe valor médio não nulo de tensão. O problema é normalmente sanado nas malhas de controle do conversor, no entanto o ajuste destas malhas em ambiente de simulação computacional requer um modelo que represente adequadamente a característica de saturação do núcleo ferromagnético. Este trabalho apresenta um modelo de simulação computacional adequado para transformadores operando em saturação intensa, visando aplicações em projetos de equipamentos de eletrônica de potência com transformador monofásico alimentado por inversor. Além disso, é demonstrado um método experimental de obtenção da característica de saturação do núcleo, sem necessidade das suas características construtivas. Resultados de simulação obtidos são comparados com resultados experimentais para validação do modelo e método. / Several power electronics based equipments are connected to the alternating current network through transformers, which perform galvanic insulation, voltage increasing/lowering, etc. These transformers may experience saturation under transitory conditions (inrush), or when its converter imposes non-zero average voltage. Such problem is commonly treated in the converter\'s control loops, but the tuning of these loops in a computer simulation environment requires a transformer model that represents adequately the ferromagnetic core saturation effect. The current work proposes a computer simulation model that adequately describes the operation of the transformer at deep saturation, suitable for design of power electronics applications with single phase transformers connected to power inverter. Additionally, an experimental method for determining its core saturation characteristic is demonstrated. This method does not require previous knowledge of the core\'s constructive parameters. Simulation and experimental results are presented to confirm the validity of the model and the method.

In this thesis, the possibility of detection of fault location in the cable distribution systems by using traveling waves due to fault and circuit breaker operations is investigated. Waveforms originated from both actions and fault steady state are separately analyzed. During such switching actions, high frequency variations which are absent in the steady state conditions, take place. In order to simulate high frequency changes properly, system elements are modeled accordingly. In other words, frequency dependent models are introduced, and they are used in Electro-Magnetic Transients Program (EMTP). Since the characteristics of waveforms are different for separately analyzed portions, different fault locating algorithms with their limitations are introduced.

Depuis un demi-siècle, le développement de la traction électrique ferroviaire en courant monophasé en France s'est appuyé sur les progrès réalisés aussi bien au niveau des installations fixes de traction qu'au niveau du matériel roulant. Toutefois, au cours des deux dernières décennies, l'augmentation du trafic et l'introduction de locomotives avec des chaines de traction innovantes ont été à l'origine de phénomènes électriques qui se sont avérés néfastes pour l'exploitation du système. Les premiers phénomènes observés ont été à l'origine de dégâts matériels à bord de locomotives. Il s'agissait de surtensions résultant d'une interaction défavorable entre l'impédance interne de l'infrastructure et les harmoniques générés par les engins moteurs équipés de redresseurs à thyristors. Plus récemment, suite à l'introduction massive d'engins équipés de redresseurs à absorption sinusoïdale de courant, un phénomène de modulation très basse fréquence de la tension caténaire est apparu et a provoqué la mise hors tension des locomotives voire la disjonction de la sous station alimentant le secteur concerné. Ceci constitue aujourd'hui un obstacle majeur à l'utilisation généralisée de la nouvelle technologie à bord des engins. Ces perturbations affectent l'exploitation du système en entrainant généralement des retards voire des annulations de circulation. Elles peuvent aussi dégrader la qualité d'énergie du réseau d'électricité amont à un niveau tel que la sous-station d'alimentation doit être déconnectée. La direction de l'ingénierie de la SNCF a donc pris des dispositions pour comprendre puis éviter l'apparition des phénomènes observés. Une collaboration interne entre le centre d'ingénierie du matériel et la division des installations fixes de traction électrique ainsi qu'un partenariat avec le LAPLACE ont été mis en place. Le présent document est le fruit de cette collaboration. L'objectif de cette thèse est donc d'étudier et de modéliser les interactions entre les engins et les installations fixes de traction sur le réseau français 25kV/50Hz. Ce manuscrit comporte deux parties principales qui s'organisent ainsi : La première partie est consacrée à l'étude du phénomène de modulation très basse fréquence de la tension caténaire. Les modèles des deux principaux composants du système sont d'abord présentés. Les études ainsi menés permettent de comprendre l'origine du phénomène, puis ensuite de développer une méthode de caractérisation des engins permettant de retrouver les limites de stabilité dans les secteurs problématiques du réseau ferré. Ceci nous a conduit à proposer une représentation générale des locomotives modernes sous forme d'une matrice admittance qu'il est possible d'obtenir par une mesure directe sur des engins réels. La deuxième partie concerne l'étude des interactions harmoniques à l'origine de surtensions sur la caténaire. L'analyse systématique du phénomène est basée sur des outils de simulation de circuits électroniques de puissance utilisant une bibliothèque de modèles élémentaires. La première étape consiste à développer un modèle « moyenne fréquence » du réseau d'alimentation afin de mettre en évidence les fréquences de résonance de l'ensemble ligne/sous-station. La deuxième étape consiste à modéliser les locomotives afin de prendre en compte leur réponse harmonique. Au final, il devient possible de savoir si un engin donné va générer des déformations de la tension en vérifiant si l'une des composantes harmoniques du courant absorbé coïncide avec une des résonances caractéristique du circuit d'alimentation. Pour compléter cette deuxième partie, une modélisation plus fine, intégrant l'effet de peau et l'effet de proximité est abordée. Elle s'appuie sur la caractérisation expérimentale en moyenne fréquence d'un transformateur 50Hz. Ceci nous permet de vérifier l'influence de ces phénomènes sur le comportement fréquentiel du réseau d'alimentation. / For a half a century, the increasing development of AC electrical traction railway networks in France relied on the progress made in the infrastructure power supply an in the rolling stock. However, over the past two decades, increased traffic and the introduction of modern locomotives were the cause of electrical phenomena that have proven harmful to the operation of the railway network. The first events that occurred induced serious faults on board locomotives. It was overvoltages resulting from unfavourable interaction between the internal impedance of the infrastructure and the harmonics generated by the electrical vehicles using thyristor controlled rectifiers. More recently, with the massive introduction of active front-end locomotives, problems of low frequency oscillations and instability were observed causing power off locomotives or disjunction of the sector sub-station. The objective of this thesis is to study and model the interactions between locomotives and fixed installations for electric traction on the French rail network 25kV/50Hz. This script has two main parts, which organized as follows: The first part is devoted to the study of the phenomenon of very low frequency modulation of the catenary voltage. The models of the two main components of the system, namely the single-phase power and the active front-end locomotives are first presented. Studies conducted this way, help to understand the origin of the phenomenon and then to develop a method to characterize the vehicle to find the stability limits in problems sectors of the rail network. This led us to propose a general representation of modern locomotives as an admittance matrix that can be obtained by direct measurement on real machines. The second part is the study of harmonic interactions causing overvoltages on the catenary. Systematic analysis of the phenomenon is based on simulation tools of power electronics circuits using a collection of specific elementary models. The first step consists in developing a “medium frequency” model of the power network in order to highlight the resonance and anti-resonance frequencies of the line/sub-station set. The second step is to model locomotives to take into account their harmonic response. In the end, it becomes possible to know whether a particular machine will generate deformations of the catenary voltage, by checking if any of the harmonic components of the consumed current coincides with one of the characteristic resonances of the traction electric circuit. To complete the second part, a more detailed model is discussed incorporating physical phenomena that occur with an increasing frequency (skin effect in the insulted conductors or proximity effect between conductors).It is based on the experimental characterization on medium frequency of a 50Hz transformer. This allows us to check the influence of these phenomena on the frequency behaviour of the supply network.

C# provides static analysis libraries for template-based code analysis and code fixing. These libraries have been used by the open-source community to generate numerous NuGet packages for different use-cases. However, due to the unstructured vastness of these packages, it is difficult to find the ones required for a project and creating new analyzers and fixers take time and effort to create. Therefore, this thesis proposes a neural network, which firstly imitates existing fixers and secondly extrapolates to fixes of unseen diagnostics. To do so, the state-of-the-art of static analysis NuGet packages is examined and further used to generate a dataset with diagnostics and corresponding code fixes for 24,622 data points. Since many C# fixers apply formatting changes, all formatting is preserved in the dataset. Furthermore, since the fixers also apply identifier changes, the tokenization of the dataset is varied between splitting identifiers by camelcase and preserving them. The neural network uses a sequence-to-sequence learning approach with the Transformer model and takes file context, diagnostic message and location as input and predicts a diff as output. It is capable of imitating 46.3% of the fixes, normalized by diagnostic type, and for data points with unseen diagnostics, it is able to extrapolate to 11.9% of normalized data points. For both experiments, splitting identifiers by camelcase produces the best results. Lastly, it is found that a higher proportion of formatting tokens in input has minimal positive impact on prediction success rates, whereas the proportion of formatting in output has no impact on success rates. / C# tillhandahåller statiska analysbibliotek för mallbaserad kodanalys och kodfixering. Dessa bibliotek har använts av open source-gemenskapen för att generera många NuGet-paket för olika användningsfall. Men på grund av mängden av dessa paket är det svårt att hitta de som krävs för ett projekt och att skapa nya analysatorer och fixare tar tid och ansträngning att skapa. Därför föreslår denna avhandling ett neuralt nätverk, som för det första imiterar befintliga korrigeringar och för det andra extrapolerar till korrigeringar av osynlig diagnostik. För att göra det har det senaste inom statisk analys NuGetpaketen undersökts och vidare använts för att generera en datauppsättning med diagnostik och motsvarande kodfixar för 24 622 datapunkter. Eftersom många C# fixers tillämpar formateringsändringar, bevaras all formatering i datasetet. Dessutom, eftersom fixarna också tillämpar identifieringsändringar, varieras tokeniseringen av datamängden mellan att dela upp identifierare efter camelcase och att bevara dem. Det neurala nätverket använder en sekvenstill- sekvens-inlärningsmetod med Transformer-modellen och tar filkontext, diagnostiskt meddelande och plats som indata och förutsäger en skillnad som utdata. Den kan imitera 46,3% av korrigeringarna, normaliserade efter diagnostisk typ, och för datapunkter med osynlig diagnostik kan den extrapolera till 11,9% av normaliserade datapunkter. För båda experimenten ger uppdelning av identifierare efter camelcase de bästa resultaten. Slutligen har det visat sig att en högre andel formateringstokens i indata har minimal positiv inverkan på åndelen korrekta förutsägelser, medan andelen formatering i utdata inte har någon inverkan på åndelen korrekta förutsägelser.

The availability and use of knowledge graphs have become commonplace as a compact storage of information and for lookup of facts. However, the discrete representation makes the knowledge graph unavailable for tasks that need a continuous representation, such as predicting relationships between entities, where the most probable relationship needs to be found. The need for a continuous representation has spurred the development of knowledge graph embeddings. The idea is to position the entities of the graph relative to each other in a continuous low-dimensional vector space, so that their relationships are preserved, and ideally leading to clusters of entities with similar characteristics. Several methods to produce knowledge graph embeddings have been created, from simple models that minimize the distance between related entities to complex neural models. Almost all of these embedding methods attempt to create an accurate static representation of each entity and relation. However, as with words in natural language, both entities and relations in a knowledge graph hold different meanings in different local contexts.  With the recent development of Transformer models, and their success in creating contextual representations of natural language, work has been done to apply them to graphs. Initial results show great promise, but there are significant differences in archi- tecture design across papers. There is no clear direction on how Transformer models can be best applied to create contextual knowledge graph embeddings. Two of the main differences in previous work is how the attention mask is applied in the model and what input graph structures the model is trained on.  This report explores how different attention masking methods and graph inputs affect a Transformer model (in this report, BERT) on a link prediction task for triples. Models are trained with five different attention masking methods, which to varying degrees restrict attention, and on three different input graph structures (triples, paths, and interconnected triples).  The results indicate that a Transformer model trained with a masked language model objective has the strongest performance on the link prediction task when there are no restrictions on how attention is directed, and when it is trained on graph structures that are sequential. This is similar to how models like BERT learn sentence structure after being exposed to a large number of training samples. For more complex graph structures it is beneficial to encode information of the graph structure through how the attention mask is applied. There also seems to be some indications that the input graph structure affects the models’ capabilities to learn underlying characteristics in the knowledge graph that is trained upon.

This dissertation develops systems engineering and machine learning models for aviation maintenance support. With the constant increase in demand for air travel, aviation organizations compete to maintain airworthy aircraft to ensure the safety of passengers. Given the importance of aircraft safety, the aviation sector constantly needs technologies to enhance the maintenance experience, ensure system safety, and limit aircraft downtime. Based on the current literature, the aviation maintenance sector still relies on outdated technologies to maintain aircraft maintenance documentation, including paper-based technical orders. Aviation maintenance documentation contains a mixture of structured and unstructured technical text, mainly inputted by operators, making them prone to error, misunderstanding communication, and inconsistency. This dissertation intends to develop decision support models based on systems engineering and artificial intelligence models that can automate the maintenance documentation system, extract useful information from maintenance work orders, and predict the aircraft's top degrader signals based on textual data. The first chapter of this dissertation introduces the significant setbacks of the aviation industry and provides a working ground for the following chapters. The dissertation's second chapter develops a system engineering framework using model-based systems engineering (MBSE) methodology to model the aviation maintenance process using the systems engineering language (SysML). The outcome of this framework is the design of an automated maintenance system model that can be used to automate maintenance documentation, making it less prone to error. The third chapter of the dissertation uses textual data in maintenance work orders to develop a hybrid approach that uses natural language processing (NLP) and transformer models to predict the readiness of a legacy aircraft. The model was tested using a real-life case study of the EA-6B military aircraft. The fourth chapter of this dissertation develops an ensemble transformer model based on three different transformer models. The ensemble model leverages the benefits of three different transformer architectures and is used to classify events based on an aviation log-based dataset. This dissertation's final and fifth chapter summarizes key findings, proposes future work directions, and provides the dissertation's limitations.

Diagnosis code assignment is a field that looks at automatically assigning diagnosis codes to free-text clinical notes. Assigning a diagnosis code to clinical notes manually needs expertise and time. Being able to do this automatically makes getting structured data from free-text clinical notes in Electronic Health Records easier. Furthermore, it can also be used as decision support for clinicians where they can input their notes and get back diagnosis codes as a second opinion. This project investigates the effects of using the hierarchies the diagnosis codes are structured in when training the diagnosis code assignment models compared to models trained with a standard loss function, binary cross-entropy. This has been done by using the hierarchy of two systems of diagnosis codes, ICD-9 and SNOMED CT, where one hierarchy is more detailed than the other. The results showed that hierarchical training increased the recall of the models regardless of what hierarchy was used. The more detailed hierarchy, SNOMED CT, increased the recall more than what the use of the less detailed ICD-9 hierarchy did. However, when using the more detailed SNOMED CT hierarchy the precision of the models decreased while the differences in precision when using the ICD-9 hierarchy was not statistically significant. The increase in recall did not make up for the decrease in precision when training with the SNOMED CT hierarchy when looking at the F1-score that is the harmonic mean of the two metrics. The conclusions from these results are that using a more detailed hierarchy increased the recall of the model more than when using a less detailed hierarchy. However, the overall performance measured in F1-score decreased when using a more detailed hierarchy since the other metric, precision, decreased by more than what recall increased. The use of a less detailed hierarchy maintained its precision giving an increase in overall performance. / Diagnoskodstilldeling är ett fält som undersöker hur man automatiskt kan tilldela diagnoskoder till fri-text läkaranteckningar. En manuell tildeling kräver expertis och mycket tid. Förmågan att göra detta automatiskt förenklar utvinning av strukturerad data från fri-text läkaranteckningar i elektroniska patientjournaler. Det kan även användas som ett hjälpverktyg för läkare där de kan skriva in sina läkaranteckningar och få tillbaka diagnoskoder som en andra åsikt. Detta arbete undersöker effekterna av att ta användning av hierarkierna diagnoskoderna är strukturerade i när man tränar modeller för diagnoskodstilldelning jämfört med att träna modellerna med en vanlig loss-funktion. Det här kommer att göras genom att använda hierarkierna av två diagnoskod-system, SNOMED CT och ICD-9, där en av hierarkierna är mer detaljerad. Resultaten visade att hierarkisk träning ökade recall för modellerna med båda hierarkierna. Den mer detaljerade hierarkien, SNOMED CT, gav en högre ökning än vad träningen med ICD-9 gjorde. Trots detta minskade precision av modellen när man den tränades med SNOMED CT hierarkin medan skillnaderna i precision när man tränade hierarkiskt med ICD-9 jämfört med vanligt inte var statistiskt signifikanta. Ökningen i recall kompenserade inte för minskningen i precision när modellen tränades med SNOMED CT hierarkien som man kan see på F1-score vilket är det harmoniska medelvärdet av de recall och precision. Slutsatserna man kan dra från de här resultaten är att en mer detaljerad hierarki kommer att öka recall mer än en mindre detaljerad hierarki ökar recall. Trots detta kommer den totala prestandan, som mäts av F1-score, försämras med en mer detaljerad hierarki eftersom att recall minskar mer än vad precision ökar. En mindre detaljerad hierarki i träning kommer bibehålla precision så att dens totala prestandan förbättras.

<p dir="ltr">Semantic pairs, which consist of related entities or concepts, serve as the foundation for comprehending the meaning of language in both written and spoken forms. These pairs enable to grasp the nuances of relationships between words, phrases, or ideas, forming the basis for more advanced language tasks like entity recognition, sentiment analysis, machine translation, and question answering. They allow to infer causality, identify hierarchies, and connect ideas within a text, ultimately enhancing the depth and accuracy of automated language processing.</p><p dir="ltr">Nevertheless, the task of extracting semantic pairs from sentences poses a significant challenge, necessitating the relevance of syntactic dependency patterns (SDPs). Thankfully, semantic relationships exhibit adherence to distinct SDPs when connecting pairs of entities. Recognizing this fact underscores the critical importance of extracting these SDPs, particularly for specific semantic relationships like hyponym-hypernym, meronym-holonym, and cause-effect associations. The automated extraction of such SDPs carries substantial advantages for various downstream applications, including entity extraction, ontology development, and question answering. Unfortunately, this pivotal facet of pattern extraction has remained relatively overlooked by researchers in the domains of natural language processing (NLP) and information retrieval.</p><p dir="ltr">To address this gap, I introduce an attention-based supervised deep learning model, ASPER. ASPER is designed to extract SDPs that denote semantic relationships between entities within a given sentential context. I rigorously evaluate the performance of ASPER across three distinct semantic relations: hyponym-hypernym, cause-effect, and meronym-holonym, utilizing six datasets. My experimental findings demonstrate ASPER's ability to automatically identify an array of SDPs that mirror the presence of these semantic relationships within sentences, outperforming existing pattern extraction methods by a substantial margin.</p><p dir="ltr">Second, I want to use the SDPs to extract semantic pairs from sentences. I choose to extract cause-effect entities from medical literature. This task is instrumental in compiling various causality relationships, such as those between diseases and symptoms, medications and side effects, and genes and diseases. Existing solutions excel in sentences where cause and effect phrases are straightforward, such as named entities, single-word nouns, or short noun phrases. However, in the complex landscape of medical literature, cause and effect expressions often extend over several words, stumping existing methods, resulting in incomplete extractions that provide low-quality, non-informative, and at times, conflicting information. To overcome this challenge, I introduce an innovative unsupervised method for extracting cause and effect phrases, PatternCausality tailored explicitly for medical literature. PatternCausality employs a set of cause-effect dependency patterns as templates to identify the key terms within cause and effect phrases. It then utilizes a novel phrase extraction technique to produce comprehensive and meaningful cause and effect expressions from sentences. Experiments conducted on a dataset constructed from PubMed articles reveal that PatternCausality significantly outperforms existing methods, achieving a remarkable order of magnitude improvement in the F-score metric over the best-performing alternatives. I also develop various PatternCausality variants that utilize diverse phrase extraction methods, all of which surpass existing approaches. PatternCausality and its variants exhibit notable performance improvements in extracting cause and effect entities in a domain-neutral benchmark dataset, wherein cause and effect entities are confined to single-word nouns or noun phrases of one to two words.</p><p dir="ltr">Nevertheless, PatternCausality operates within an unsupervised framework and relies heavily on SDPs, motivating me to explore the development of a supervised approach. Although SDPs play a pivotal role in semantic relation extraction, pattern-based methodologies remain unsupervised, and the multitude of potential patterns within a language can be overwhelming. Furthermore, patterns do not consistently capture the broader context of a sentence, leading to the extraction of false-positive semantic pairs. As an illustration, consider the hyponym-hypernym pattern <i>the w of u</i> which can correctly extract semantic pairs for a sentence like <i>the village of Aasu</i> but fails to do so for the phrase <i>the moment of impact</i>. The root cause of this limitation lies in the pattern's inability to capture the nuanced meaning of words and phrases in a sentence and their contextual significance. These observations have spurred my exploration of a third model, DepBERT which constitutes a dependency-aware supervised transformer model. DepBERT's primary contribution lies in introducing the underlying dependency structure of sentences to a language model with the aim of enhancing token classification performance. To achieve this, I must first reframe the task of semantic pair extraction as a token classification problem. The DepBERT model can harness both the tree-like structure of dependency patterns and the masked language architecture of transformers, marking a significant milestone, as most large language models (LLMs) predominantly focus on semantics and word co-occurrence while neglecting the crucial role of dependency architecture.</p><p dir="ltr">In summary, my overarching contributions in this thesis are threefold. First, I validate the significance of the dependency architecture within various components of sentences and publish SDPs that incorporate these dependency relationships. Subsequently, I employ these SDPs in a practical medical domain to extract vital cause-effect pairs from sentences. Finally, my third contribution distinguishes this thesis by integrating dependency relations into a deep learning model, enhancing the understanding of language and the extraction of valuable semantic associations.</p>

The amount of data available and consumed by people globally is growing. To reduce mental fatigue and increase the general ability to gain insight into complex texts or documents, we have developed an application to aid in this task. The application allows users to upload documents and ask domain-specific questions about them using our web application. A summarized version of each document is presented to the user, which could further facilitate their understanding of the document and guide them towards what types of questions could be relevant to ask. Our application allows users flexibility with the types of documents that can be processed, it is publicly available, stores no user data, and uses state-of-the-art models for its summaries and answers. The result is an application that yields near human-level intuition for answering questions in certain isolated cases, such as Wikipedia and news articles, as well as some scientific texts. The application shows a decrease in reliability and its prediction as to the complexity of the subject, the number of words in the document, and grammatical inconsistency in the questions increases. These are all aspects that can be improved further if used in production. / Mängden data som är tillgänglig och konsumeras av människor växer globalt. För att minska den mentala trötthet och öka den allmänna förmågan att få insikt i komplexa, massiva texter eller dokument, har vi utvecklat en applikation för att bistå i de uppgifterna. Applikationen tillåter användare att ladda upp dokument och fråga kontextspecifika frågor via vår webbapplikation. En sammanfattad version av varje dokument presenteras till användaren, vilket kan ytterligare förenkla förståelsen av ett dokument och vägleda dem mot vad som kan vara relevanta frågor att ställa. Vår applikation ger användare möjligheten att behandla olika typer av dokument, är tillgänglig för alla, sparar ingen personlig data, och använder de senaste modellerna inom språkbehandling för dess sammanfattningar och svar. Resultatet är en applikation som når en nära mänsklig intuition för vissa domäner och frågor, som exempelvis Wikipedia- och nyhetsartiklar, samt viss vetensaplig text. Noterade undantag för tillämpningen härrör från ämnets komplexitet, grammatiska korrekthet för frågorna och dokumentets längd. Dessa är områden som kan förbättras ytterligare om den används i produktionen.