Search

This study introduces a novel performance-based weighting scheme for ensemble learning using the Shapley value. The weighting uses the reciprocal of binary cross-entropy as a base learner's performance metric and estimates its Shapley value to measure the overall contribution of a learner to an equally weighted ensemble of various sizes. Two variants of this strategy were empirically compared with a single monolith model and other static weighting strategies using two large banking-related datasets. A variant that discards learners with a negative Shapley value was ranked as first or at least second when constructing homogeneous ensembles, whereas for heterogeneous ensembles this strategy resulted in a better or at least similar detection performance to other weighting strategies tested. The main limitation being the computational complexity of Shapley calculations, the explored weighting strategy could be considered as a generalization of performance-based weighting.

Blockchain smart contracts can support the decentralisation of business processes, but due to smart contracts’ specifics, their development is a complicated process. Introducing model-driven development principles in smart contract development can facilitate requirement specification, design, and implementation activities. This paper presents a model-driven development method MDAsmartCD (Model-Driven Architecture-based Smart Contract Development) to alleviate smart contract development by supporting the complete MDA life cycle, covering the definition of Computation-Independent Model, Platform-Independent Model, and two instances of Platform-Specific Models. In MDAsmartCD, model transformations (model-to-model and model-to-text) are used to produce smart contract code in the Hyperledger Fabric platform Go and the Ethereum platform Solidity programming languages. The method application was demonstrated by implementing the smart contract for the hackathon solution and executing the generated Solidity and Go smart contracts in the workflow of issuing certificates for hackathon participants. During the execution of the workflow, both deployed smart contracts behaved identically and recorded analogous results in respective blockchain data storages. This demonstrated that the MDAsmartCD method enables the generation of compilable and executable smart contract code, ready for deployment on a blockchain platform.

Blockchain technology and smart contract development currently lacks clarity in its implementation. The complicated architecture of blockchain is an obstacle that developers face during design and implementation of blockchain-based systems. In this paper we propose a method based on Model Driven Architecture, which could be used for defining and specifying blockchain structure and behavior. Such approach could be used as one of the ways for describing blockchain-based systems in a more general language in order to facilitate blockchain development process.

Dėl didelio internetinių informacinių sistemų kūrimo tempo, dažnu atveju šios informacinės sistemos yra prastai dokumentuojamos, taip apsunkinant jų palaikymą ir tobulinimą. Tokia pat situacija susidaro, kai veikiančioje informacinėje sistemoje įvykdomi pakeitimai ar informacinės sistemos funkcionalumas plečiamas, o sistemos projektas nėra papildomas. Dėl to informacinės sistemos dokumentacija greitai atitrūksta nuo esamo funkcionalumo ir tampa ne taip gerai panaudojama informacinės sistemos plėtojimo ir palaikymo metu. Šiame darbe apžvelgiamos galimybės sistemos UML modelius išgauti atgalinės inžinerijos būdu, siekiant palengvinti sistemos dokumentavimą. Išanalizavus rinkoje esančius sprendimus ir mokslinius tyrimus atliktus šioje srityje, nebuvo rastas sprendimas, kuris leistų neturint sistemos programinio kodo sugeneruoti panaudojimo atvejų modelį. Todėl nuspręsta kurti naują metodą ir įrankį, kuris, nepriklausomai nuo internetinei sistemai naudotos kūrimo technologijos, leistų išgauti jos UML panaudojimo atvejų modelį. Šiame darbe pateikiamas siūlomas metodas, leidžiantis iš užregistruotų sistemos naudotojo veiksmų ir informacijos, išgautos iš grafinės sąsajos, sudaryti panaudojimo atvejų modelį. Taip pat pateikiamas metodo veikimui pademonstruoti sukurto įrankio projektas ir jo realizacijos aprašymas. Eksperimento metu, siekiant patikrinti metodo ir įrankio veikimo teisingumą, sugeneruoti keturi panaudojimo atvejų modeliai. Du panaudojimo atvejų modeliai įvertinti, juose gautas diagramas ir jų elementus palyginus su analitikų suprojektuotais modeliais. Kiti du modeliai pateikti ekspertų vertinimui, siekiant išsiaiškinti sugeneruotų panaudojimo atvejų modelių kokybę. Įvertinus eksperimentų rezultatus, nustatyta, kad darbe sukurtas metodas yra tinkamas panaudojimo atvejų modelio generavimui ir jo kokybė, ekspertų yra vertinama aukštais balais, As websites exists in highly competitive environment, most of website developers tend to sacrifice quality of website documentation to spend more resources on website functionality to get competitive edge. As websites are poorly documented it is harder to maintain or introduce new functionality. The same kind of situation happens, when websites are updated without updating the corresponding documentation. This way websites’ documentation loses its value and is not as useful while maintaining or updating. In this thesis, capabilities of reverse engineering websites to extract UML models are analyzed, in order to simplify website documenting. After the analysis of tools and research in the area, no tool or method was found, capable of generating Use Case model without source code analysis. It was concluded that new method and tool should be created, that would be able to generate UML Use Case model not depending on platform or programing language. In this thesis the method, that provides ability to generate use case model from use interface data and recorded usage on websites, was developed, accompanied by the method prototype implementation. To test the correctness of the created method and prototype tool, four Use Case models were generated. Two of them were evaluated by comparing them to Use Case models created by analysts. The other two were presented to UML modeling experts, to evaluate generated models’ quality. Experiment results indicate that method and tools presented in this thesis are capable of creating Use Case models, and experts evaluated the quality of generated models with the high scores.

Gamification in education helps to enhance courses to increase user engagement, productivity and motivation. The current Information System Design course at Kaunas University of Technology covers a broad range of topics on UML and students tend to lose their motivation and engagement during the learning process. To combat this problem an idea of gamifying the course was proposed. In this paper we have presented an implementation of gamified UML and system design teaching course in Moodle environment. The Moodle course was developed, which utilizes additional plugins for implementing the required gamification elements. In 2017 autumn semester students were invited to participate in gamified course, their activity in the course was recorded and analyzed. Surveys' results indicate that students enjoyed gamified course, and plan to continue using it in the future.