The thesis investigates the automatic and semi-automatic evaluation of programming assignments in teaching of programming, exams and competitions. Manual evaluation of programming assignments is still dominant in programming teaching in Lithuania. In some contexts (i.e. programming competitions, basics of programming and algorithms course) there is widely applied automatic and semi-automatic evaluation of programming tasks based on black-box testing approach. However, automatic evaluation based on black-box testing is criticized for its inability to demonstrate the strengths and the weaknesses of the program being evaluated. Automatic evaluation is not capable of identifying the level of achievement in incomplete programs. The thesis addresses the problem of evaluation of practical programming assignments in Information Technology maturity exam (IT VBE). This problem involves requirements to evaluate reliably thousands of programming assignment solutions in a reasonable amount of time. The thesis presents analysis of systems for automatic evaluation of programming tasks and appropriate types of assignments. The analysis led to the improvement of the evaluation method. The most important result of the research is the proposed way to improve the semi-automatic evaluation and testing method by increasing its interactivity. A semi-automatic assessment system is expanded by adding the components responsible for the modification of submitted programs, monitoring changes in the... [to full text] / Disertacijoje nagrinėjamas automatinis ir pusiau automatinis programavimo užduočių vertinimas programavimo mokyme, egzaminuose ir varžybose. Šiuo metu informatikos mokyme Lietuvoje vis dar dominuoja rankinis programavimo užduočių vertinimas. Atskirose švietimo srityse (programavimo varžybos, algoritmavimo kursai) išplito automatinis ir pusiau automatinis programavimo užduočių vertinimas, paremtas juodosios dėžės testavimo principu. Tačiau automatinio vertinimo pagrindu gauti įverčiai kritikuojami kaip neatskleidžiantys visų pateiktos programos privalumų ir trūkumų. Ypatingai dideli vertinimo netikslumai gaunami vertinant nebaigtas programas. Disertacijoje sprendžiama Informacinių technologijų valstybinio brandos egzamino (IT VBE) praktinių programavimo užduočių vertinimo problema – siekiama gauti patikimus įvertinimus dideliam pateiktų programavimo užduočių sprendimų kiekiui per priimtiną laiką. Išanalizuotos automatinės programavimo užduočių vertinimo sistemos bei joms tinkami užduočių tipai atvedė prie idėjos tobulinti vertinimo metodą. Svarbiausias disertacijos rezultatas – pasiūlytas būdas tobulinti pusiau automatinį vertinimo ir testavimo metodą didinant jo interaktyvumą. Tam pusiau automatinė vertinimo sistema turi būti papildoma komponentais, leidžiančiais modifikuoti pateiktą programą, stebėti modifikuotos programos funkcionalumo pokyčius. Pasiūlytas patobulintas vertinimo metodas išbandytas praktiškai, sukurta IT VBE praktinių užduočių vertinimo sistema. Taikant šį... [toliau žr. visą tekstą]

Disertacijoje nagrinėjamas automatinis ir pusiau automatinis programavimo užduočių vertinimas programavimo mokyme, egzaminuose ir varžybose. Šiuo metu informatikos mokyme Lietuvoje vis dar dominuoja rankinis programavimo užduočių vertinimas. Atskirose švietimo srityse (programavimo varžybos, algoritmavimo kursai) išplito automatinis ir pusiau automatinis programavimo užduočių vertinimas, paremtas juodosios dėžės testavimo principu. Tačiau automatinio vertinimo pagrindu gauti įverčiai kritikuojami kaip neatskleidžiantys visų pateiktos programos privalumų ir trūkumų. Ypatingai dideli vertinimo netikslumai gaunami vertinant nebaigtas programas. Disertacijoje sprendžiama Informacinių technologijų valstybinio brandos egzamino (IT VBE) praktinių programavimo užduočių vertinimo problema – siekiama gauti patikimus įvertinimus dideliam pateiktų programavimo užduočių sprendimų kiekiui per priimtiną laiką. Išanalizuotos automatinės programavimo užduočių vertinimo sistemos bei joms tinkami užduočių tipai atvedė prie idėjos tobulinti vertinimo metodą. Svarbiausias disertacijos rezultatas – pasiūlytas būdas tobulinti pusiau automatinį vertinimo ir testavimo metodą didinant jo interaktyvumą. Tam pusiau automatinė vertinimo sistema turi būti papildoma komponentais, leidžiančiais modifikuoti pateiktą programą, stebėti modifikuotos programos funkcionalumo pokyčius. Pasiūlytas patobulintas vertinimo metodas išbandytas praktiškai, sukurta IT VBE praktinių užduočių vertinimo sistema. Taikant šį... [toliau žr. visą tekstą] / The thesis investigates the automatic and semi-automatic evaluation of programming assignments in teaching of programming, exams and competitions. Manual evaluation of programming assignments is still dominant in programming teaching in Lithuania. In some contexts (i.e. programming competitions, basics of programming and algorithms course) there is widely applied automatic and semi-automatic evaluation of programming tasks based on black-box testing approach. However, automatic evaluation based on black-box testing is criticized for its inability to demonstrate the strengths and the weaknesses of the program being evaluated. Automatic evaluation is not capable of identifying the level of achievement in incomplete programs. The thesis addresses the problem of evaluation of practical programming assignments in Information Technology maturity exam (IT VBE). This problem involves requirements to evaluate reliably thousands of programming assignment solutions in a reasonable amount of time. The thesis presents analysis of systems for automatic evaluation of programming tasks and appropriate types of assignments. The analysis led to the improvement of the evaluation method. The most important result of the research is the proposed way to improve the semi-automatic evaluation and testing method by increasing its interactivity. A semi-automatic assessment system is expanded by adding the components responsible for the modification of submitted programs, monitoring changes in the... [to full text]

This thesis tries to answer if a semi-automatic tool can speed up the process of segmenting tumors to find the area of a slice in the tumor or the volume of the entire tumor. A few different 2D semi-automatic tools were considered. The final choice was to implement live-wire. The implemented live-wire was evaluated and improved upon with hands-on testing from developers. Two methods were found for extending live-wire to 3D bodies. The first method was to interpolate the seed points and create new contours using the new seed points. The second method was to let the user segment contours in two orthogonal projections. The intersections between those contours and planes in the third orthogonal projection were then used to create automatic contours in this third projection. Both tools were implemented and evaluated. The evaluation compared the two tools to manual segmentation on two cases posing different difficulties. Time-on-task and accuracy were measured during the evaluation. The evaluation revealed that the semi-automatic tools could indeed save the user time while maintaining acceptable (80%) accuracy. The significance of all results were analyzed using two-tailed t-tests.

The main goal of this thesis is to create a generator of scientific web portals which makes the creation of portals much easier. The administrators of the portals will be able to show different information to specific user groups. The generator is based on modular structure that uses Nette framework (PHP), Doctrine ORM and jQuery framework (Javascript, Ajax). The portal presents the results depending on settings of each portal and its basic data is saved in local database. The local database is a stand-alone project (ReReSearch). The portal works in different modes which are classified into three categories: automatic, semi-automatic and manual mode. The priority mode is the semi-automatic. The generator works also as a feedback for the local database, which provides relevant information from specified fields of science.

Self-driving vehicles has become a hot topic in today's industry during the past years and companies all around the globe are attempting to solve the complex task of developing vehicles that can safely navigate roads and traffic without the assistance of a driver.  As deep learning and computer vision becomes more streamlined and with the possibility of using fisheye cameras as a cheap alternative to external sensors some companies have begun researching the possibility for assisted driving on vehicles such as electrical scooters to prevent injuries and accidents by detecting dangerous situations as well as promoting a sustainable infrastructure. However training such a model requires gathering large amounts of data which needs to be labeled by a human annotator. This process is expensive, time consuming, and requires extensive quality checking which can be difficult for companies to afford. This thesis presents an application that allows for semi-automatically labeling a dataset with the help of a human annotator and an object detector. The application trains an object detector together with an active learning framework on a small part of labeled data sampled from the woodscape dataset of fisheye distorted images and uses the knowledge of the trained model as well as using a human annotator as assistance to label more data. This thesis examines the labeled data produced by using the application described in this thesis and compares them with the quality of the annotations in the woodscape dataset. Results show that the model can't make any quality annotations compared to the woodscape dataset and resulted in the human annotator having to label all of the data, and the model achieved an accuracy of 0.00099 mAP.

Title: Structured Data Extraction from Unstructured Text Author: Bc. Peter Kóša Department: Department of Software Engineering Supervisor: Mgr. Martin Nečaský, Ph.D., Department of Software Engineering Abstract: In the last 20 years, there has been an ever-growing amount of information present on the Internet and in published texts. However, this information is often in a non-structured format and this causes various problems such as the inability to efficiently search in diverse collections of texts (medical reports, ads, etc.). To overcome these problems, we need efficient tools capable of automatic processing, extracting the important information and storing of these results in some form for later reuse. The purpose of this thesis is to compare existing solutions as well as to compare them with our solution, which was created in the scope of software project SemJob. The SemJob project is introduced and the reader can therefore obtain knowledge about its inner structure and workings. Keywords: structured data extraction, extraction rules, (semi)automatic wrapper induction

Title: Structured Data Extraction from Unstructured Text Author: Bc. Peter Kóša Department: Department of Software Engineering Supervisor: Mgr. Martin Nečaský, Ph.D., Department of Software Engineering Abstract: In the last 20 years, there has been an ever-growing amount of information present on the Internet and in published texts. However, this information is often in a non-structured format and this causes various problems such as the inability to efficiently search in diverse collections of texts (medical reports, ads, etc.). To overcome these problems, we need efficient tools capable of automatic processing, extracting the important information and storing of these results in some form for later reuse. The purpose of this thesis is to compare existing solutions as well as to compare them with our solution, which was created in the scope of software project SemJob. The SemJob project is introduced and the reader can therefore obtain knowledge about its inner structure and workings. Keywords: structured data extraction, extraction rules, ontologies, (semi)automatic wrapper induction

Les systèmes distribués collaboratifs, en particulier les systèmes pair-à-pair, forment l’infrastructure sous-jacente de nombreuses applications Internet, certaines parmi les plus populaires (ex : partage de fichiers, streaming multimédia). Ils se situent également à la base d’un ensemble de technologies émergentes telles que la blockchain et l’Internet des Objets. Le succès de ces systèmes repose sur la contribution volontaire, de la part des nœuds participants, aux ressources partagées (ex : bande passante réseau, puissance de calcul, stockage de données). Or ces nœuds sont des entités autonomes qui peuvent considérer comme plus avantageux de se comporter de manière égoïste, c’est-à-dire de refuser de collaborer. De tels comportements peuvent fortement impacter les performances et la stabilité opérationnelles du système cible. Prendre en compte et prévenir les comportements égoïstes des nœuds est donc essentiel pour garantir l’efficacité et la fiabilité des systèmes coopératifs. Cependant, cela exige du développeur, en dépit de la grande quantité de techniques et d’approches proposées dans la littérature, des connaissances multisectorielles approfondies. L'objectif de cette thèse est de concevoir et étudier de nouveaux outils théoriques et pratiques pour aider les concepteurs de systèmes distribués collaboratifs à faire face à des nœuds égoïstes. La première contribution, basée sur une analyse exhaustive de la littérature sur les comportements égoïstes dans les systèmes distribués, propose un modèle de classification pour identifier et analyser les comportements égoïstes les plus importants sur lesquels il est important de se concentrer lors de la conception d'un système coopératif. Dans la deuxième contribution, nous proposons RACOON, un framework pour la conception et la configuration de systèmes coopératifs résilients aux comportements égoïstes. Outre un ensemble de mécanismes d'incitation à la coopération, RACOON fournit une méthodologie semi-automatique d’intégration et de calibration de ces mécanismes de manière à garantir le niveau de performance souhaité. RACOON s’appuie sur une analyse du système cible fondée sur la théorie des jeux et sur des simulations pour prédire l’existence de nœuds égoïstes dans le système. RACOON a été étendu en un deuxième framework, RACOON++. Plus précis, plus flexible, RACOON++ offre également une plus grande facilité d'utilisation. Une dernière contribution, SEINE, propose un framework pour la modélisation et l'analyse des différents types de comportements égoïstes dans un système coopératif. Basé sur un langage dédié, développé pour décrire les scénarios de comportement égoïstes, SEINE fournit un support semi-automatique pour la mise en œuvre et l'étude de ces scénarios dans un simulateur choisi sur la base de l’état de l’art (PeerSim). / Cooperative distributed systems, particularly peer-to-peer systems, are the basis of several mainstream Internet applications (e.g., file-sharing, media streaming) and the key enablers of new and emerging technologies, including blockchain and the Internet of Things. Essential to the success of cooperative systems is that nodes are willing to cooperate with each other by sharing part of their resources, e.g., network bandwidth, CPU capability, storage space. However, as nodes are autonomous entities, they may be tempted to behave in a selfish manner by not contributing their fair share, potentially causing system performance degradation and instability. Addressing selfish nodes is, therefore, key to building efficient and reliable cooperative systems. Yet, it is a challenging task, as current techniques for analysing selfishness and designing effective countermeasures remain manual and time-consuming, requiring multi-domain expertise. In this thesis, we aim to provide practical and conceptual tools to help system designers in dealing with selfish nodes. First, based on a comprehensive survey of existing work on selfishness, we develop a classification framework to identify and understand the most important selfish behaviours to focus on when designing a cooperative system. Second, we propose RACOON, a unifying framework for the selfishness-aware design and configuration of cooperative systems. RACOON provides a semi-automatic methodology to integrate a given system with practical and finely tuned mechanisms to meet specified resilience and performance objectives, using game theory and simulations to predict the behaviour of the system when subjected to selfish nodes. An extension of the framework (RACOON++) is also proposed to improve the accuracy, flexibility, and usability of RACOON. Finally, we propose SEINE, a framework for fast modelling and evaluation of various types of selfish behaviour in a given cooperative system. SEINE relies on a domain-specific language for describing the selfishness scenario to evaluate and provides semi-automatic support for its implementation and study in a state-of-the-art simulator.

A área médica tem se beneficiado das ferramentas construídas pela Computação e, ao mesmo tempo, tem impulsionado o desenvolvimento de novas técnicas em diversas especialidades da Computação. Dentre estas técnicas a segmentação tem como objetivo separar em uma imagem objetos de interesse, podendo chamar a atenção do profissional de saúde para áreas de relevância ao diagnóstico. Além disso, os resultados da segmentação podem ser utilizados para a reconstrução de modelos tridimensionais, que podem ter características extraídas que auxiliem o médico em tomadas de decisão. No entanto, a segmentação de imagens médicas ainda é um desafio, por ser extremamente dependente da aplicação e das estruturas de interesse presentes na imagem. Esta dissertação apresenta uma técnica de segmentação semiautomática do endocárdio do ventrículo esquerdo em conjuntos de imagens cardíacas de Ressonância Magnética Nuclear. A principal contribuição é a segmentação considerando todas as imagens provenientes de um exame, por meio da propagação dos resultados obtidos em imagens anteriormente processadas. Os resultados da segmentação são avaliados usando-se métricas objetivas como overlap, entre outras, comparando com imagens fornecidas por especialistas na área de Cardiologia / The medical field has been benefited from the tools built by Computing and has promote the development of new techniques in diverse Computer specialties. Among these techniques, the segmentation aims to divide an image into interest objects, leading the attention of the specialist to areas that are relevant in diagnosys. In addition, segmentation results can be used for the reconstruction of three-dimensional models, which may have extracted features that assist the physician in decision making. However, the segmentation of medical images is still a challenge because it is extremely dependent on the application and structures of interest present in the image. This dissertation presents a semiautomatic segmentation technique of the left ventricular endocardium in sets of cardiac images of Nuclear Magnetic Resonance. The main contribution is the segmentation considering all the images coming from an examination, through the propagation of the results obtained in previously processed images. Segmentation results are evaluated using objective metrics such as overlap, among others, compared to images provided by specialists in the Cardiology field

A área médica tem se beneficiado das ferramentas construídas pela Computação e, ao mesmo tempo, tem impulsionado o desenvolvimento de novas técnicas em diversas especialidades da Computação. Dentre estas técnicas a segmentação tem como objetivo separar em uma imagem objetos de interesse, podendo chamar a atenção do profissional de saúde para áreas de relevância ao diagnóstico. Além disso, os resultados da segmentação podem ser utilizados para a reconstrução de modelos tridimensionais, que podem ter características extraídas que auxiliem o médico em tomadas de decisão. No entanto, a segmentação de imagens médicas ainda é um desafio, por ser extremamente dependente da aplicação e das estruturas de interesse presentes na imagem. Esta dissertação apresenta uma técnica de segmentação semiautomática do endocárdio do ventrículo esquerdo em conjuntos de imagens cardíacas de Ressonância Magnética Nuclear. A principal contribuição é a segmentação considerando todas as imagens provenientes de um exame, por meio da propagação dos resultados obtidos em imagens anteriormente processadas. Os resultados da segmentação são avaliados usando-se métricas objetivas como overlap, entre outras, comparando com imagens fornecidas por especialistas na área de Cardiologia / The medical field has been benefited from the tools built by Computing and has promote the development of new techniques in diverse Computer specialties. Among these techniques, the segmentation aims to divide an image into interest objects, leading the attention of the specialist to areas that are relevant in diagnosys. In addition, segmentation results can be used for the reconstruction of three-dimensional models, which may have extracted features that assist the physician in decision making. However, the segmentation of medical images is still a challenge because it is extremely dependent on the application and structures of interest present in the image. This dissertation presents a semiautomatic segmentation technique of the left ventricular endocardium in sets of cardiac images of Nuclear Magnetic Resonance. The main contribution is the segmentation considering all the images coming from an examination, through the propagation of the results obtained in previously processed images. Segmentation results are evaluated using objective metrics such as overlap, among others, compared to images provided by specialists in the Cardiology field