Sustav automatskog upravljanja poslovanjem u stvarnom vremenu baziran na umjetnoj inteligenciji uz ljudski nadzor

master's thesis

• Title (english): Automatic real time business management system based on artificial intelligence with human in the loop
• Author: Jurica Kovačević
• Mentor: Tomislav Jagušt (mentor)
• Committee member: Tomislav Jagušt (predsjednik povjerenstva)
• Committee member: Ivica Botički (član povjerenstva)
• Committee member: Marko Horvat (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-06, Croatia

Abstract

Diplomski rad opisuje sustave za automatizirano upravljanje rizikom pomoću umjetne inteligencije u kontekstu kladionice. Svaku izmjenu parametra korisnika zbog GDPRa mora odobriti čovjek, stoga se svaka izmjena prikazuje kao prijedlog internom stručnjaku. Pri arhitekturi sustava uzeto je u obzir okruženje promjenjivih zahtjeva sa više razvojnih timova. Prilagodljivost novim zahtjevima, minimizacija kompleksnosti i smanjenje međuovisnosti timova postignuti su kombinacijom mikroservisne arhitekture i arhitekture zasnovane na događajima. Opisane su arhitekture te odabrane tehnologije za implementaciju platforme za notifikacije, platforme za upravljanje prijedlozima, te sustava za automatsku segmentaciju korisnika i automatsku izmjenu koeficijenta ograničenja. Također, implementiran je prototip sustava za upravljanje prijedlozima.

Abstract (english)

This master thesis describes systems for AI-based automatization of a betting companies risk management. Every change of a customers parameters must be approved by a human due to GDPR, so every change is presented to an internal expert as a suggestion. The architecture described in this thesis is optimized for changing requirements and a development environment with multiple teams. Adaptability, minimization of complexity, and decoupling of teams is achieved by leveraging microservice and event driven architecture approaches. The architecture and technology choices of a notification platform, suggestion management platform, system for automatic segmentation of users and betting limit coefficient change is described, and a prototype of the suggestions system is implemented.

• Keywords: upravljanje rizikom; automatizacija; klađenje; mikroservisna arhitektura; komunikacija porukama; NoSQL; serijalizacijski formati; DynamoDB; Apache Kafka; Protobuf; JSON; gRPC; SSE; WebSockets
• Keywords (english): risk management; automatization; betting; microservice architecture; event driven architecture; NoSQL; serialization formats; DynamoDB; Apache Kafka; Protobuf; JSON; gRPC; SSE; WebSockets
• Language: croatian
• URN:NBN: urn:nbn:hr:168:930954
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-10-11 09:21:59