Web-aplikacija za demonstraciju genetskog programiranja

• Main Author: Nimac, Josip
• Other Authors: Botički, Ivica
• Format: Bachelor Thesis
• Language: Croatian
• Published: Sveučilište u Zagrebu. Fakultet elektrotehnike i računarstva. 2019
• Subjects: genetsko programiranje; evolucijsko računarstvo; evolucijski algoritam; web-aplikacija; Node.js; React.js; PostgreSQL; genetic programming; evolutionary computing; evolutionary algorithm; web-application; TEHNIČKE ZNANOSTI. Računarstvo; TECHNICAL SCIENCES. Computing; sami
• Online Access: https://zir.nsk.hr/islandora/object/fer:6218; https://urn.nsk.hr/urn:nbn:hr:168:771758; https://repozitorij.unizg.hr/islandora/object/fer:6218; https://repozitorij.unizg.hr/islandora/object/fer:6218/datastream/PDF

Tema ovog završnog rada bila je razviti web-aplikaciju koja bi korisnicima približila područja genetskog programiranja i evolucijskih algoritama na interaktivan način. Problem koji se rješava je problem treniranja virtualnog mrava na stazi s hranom. Korisnici sami sami mogu birati parametre koji ulaze u algoritam te mogu sami i dizajnirati stazu na kojoj se mravi treniraju. Po završetku treninga korisnicima se prikazuje animacija na kojoj se vidi ponašanje mrava. Poslužiteljski dio aplikacije je izrađen u Node.js okruženju s Postgres bazom podataka, dok je klijentski dio izrađen s JavaScript bibliotekom React.js. The subject of this final work was to develop a web-application that would bring users closer to subjects like genetic programming and evolutionary algorithms in an interactive way. Problem that can be solved is the problem of training a virtual ant to follow some food trail. Users can choose the parameters of the algorithm and they can even design treir own food trails. On training completion users are shown an animation that shows the ant's behaviour. Server side of the application was built with the Node.js environment with PostgreSQL as database, while client side was built with JavaScript library React.js.