{"id":{"repo_id":"calgary","oai_identifier":"oai:ucalgary.scholaris.ca:1880/123708"},"canonical_url":"https://search.dev.ndltd.org/etd/calgary/oai:ucalgary.scholaris.ca:1880/123708","repository":{"repo_id":"calgary","name":"University of Calgary","base_url":"https://ucalgary.scholaris.ca/server/oai/request"},"display":{"title":"LLM-based MaSE - A Software Development Framework for Developing Multi-Agent System","abstract":"The automation of software development prompted extensive research for decades. Recent advancements in AI, particularly large language models (LLMs), are driving transformative changes in the field. While LLMs have demonstrated impressive performance in code generation, challenges remain in generating complex and domain-specific code. To fully realize their potential, it is crucial to extend their application throughout the entire software development lifecycle. This necessity has driven us to explore and evaluate the role of these models in supporting all stages of the software engineering process. In this thesis, we propose an LLM-based framework tailored for the development of multiagent systems (MAS). By focusing on MAS and leveraging the MaSE methodology in conjunction with the JADE platform, our framework ensures the generation of accurate, traceable, and reliable software artefacts efficiently. The framework uses predefined prompts to guide the LLM through each step of the methodology, producing outputs that adhere to domain-specific requirements. Empirical evaluations on sample projects, through a case study and an experiment, demon-strate that this approach not only accelerates the development process but also results in artefacts that are more comprehensive and modular than those produced by human developers, showcasing the transformative potential of LLMs in software engineering.","abstract_html":"The automation of software development prompted extensive research for decades. Recent advancements in AI, particularly large language models (LLMs), are driving transformative changes in the field. While LLMs have demonstrated impressive performance in code generation, challenges remain in generating complex and domain-specific code. To fully realize their potential, it is crucial to extend their application throughout the entire software development lifecycle. This necessity has driven us to explore and evaluate the role of these models in supporting all stages of the software engineering process. In this thesis, we propose an LLM-based framework tailored for the development of multiagent systems (MAS). By focusing on MAS and leveraging the MaSE methodology in conjunction with the JADE platform, our framework ensures the generation of accurate, traceable, and reliable software artefacts efficiently. The framework uses predefined prompts to guide the LLM through each step of the methodology, producing outputs that adhere to domain-specific requirements. Empirical evaluations on sample projects, through a case study and an experiment, demon-strate that this approach not only accelerates the development process but also results in artefacts that are more comprehensive and modular than those produced by human developers, showcasing the transformative potential of LLMs in software engineering.","abstract_has_math":false,"creators":["Hajjar Zadeh, Sahar"],"institution":"University of Calgary","degree_name":"Master of Science (MSc)","degree_level":null,"degree_discipline":"Engineering – Electrical & Computer","degree_department":null,"school":null,"contributors":[],"advisors":["Far, Behrouz","Shakeri Hossein Abad, Zahra"],"committee_chairs":[],"committee_members":["Moussavi, Mahmood","Drew, Steve"],"year":2025,"date_issued":"2025-12-19","date_published":"2025-12-19","updated_at":"2026-07-24T01:30:22Z","subjects":["Large Language Models","Multi-Agent Systems","Software Development Methodology","MaSE","Statistical Analysis"],"languages":["en"],"rights":["Unless otherwise indicated, this material is protected by copyright and has been made available with authorization from the copyright owner. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission."],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://dx.doi.org/10.11575/PRISM/50921"],"render_values":[{"text":"https://dx.doi.org/10.11575/PRISM/50921","href":"https://dx.doi.org/10.11575/PRISM/50921","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/1880/123708","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Far, Behrouz","Shakeri Hossein Abad, Zahra"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Moussavi, Mahmood","Drew, Steve"]},{"key":"dc:creator","label":"Author","values":["Hajjar Zadeh, Sahar"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026-02"]},{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-01-09T22:42:21Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-12-19"]},{"key":"dc:type","label":"Dc Type","values":["master thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Engineering – Electrical & Computer"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science (MSc)"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Calgary"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Large Language Models","Multi-Agent Systems","Software Development Methodology","MaSE","Statistical Analysis"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Unless otherwise indicated, this material is protected by copyright and has been made available with authorization from the copyright owner. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://dx.doi.org/10.11575/PRISM/50921"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1880/123708"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The automation of software development prompted extensive research for decades. Recent advancements in AI, particularly large language models (LLMs), are driving transformative changes in the field. While LLMs have demonstrated impressive performance in code generation, challenges remain in generating complex and domain-specific code. To fully realize their potential, it is crucial to extend their application throughout the entire software development lifecycle. This necessity has driven us to explore and evaluate the role of these models in supporting all stages of the software engineering process. In this thesis, we propose an LLM-based framework tailored for the development of multiagent systems (MAS). By focusing on MAS and leveraging the MaSE methodology in conjunction with the JADE platform, our framework ensures the generation of accurate, traceable, and reliable software artefacts efficiently. The framework uses predefined prompts to guide the LLM through each step of the methodology, producing outputs that adhere to domain-specific requirements. Empirical evaluations on sample projects, through a case study and an experiment, demon-strate that this approach not only accelerates the development process but also results in artefacts that are more comprehensive and modular than those produced by human developers, showcasing the transformative potential of LLMs in software engineering."]},{"key":"dc:title","label":"Title","values":["LLM-based MaSE - A Software Development Framework for Developing Multi-Agent System"]}]}],"canonical_facts":{"dc:contributor.advisor":["Far, Behrouz","Shakeri Hossein Abad, Zahra"],"dc:contributor.committeemember":["Moussavi, Mahmood","Drew, Steve"],"dc:creator":["Hajjar Zadeh, Sahar"],"dc:date":["2026-02"],"dc:date.accessioned":["2026-01-09T22:42:21Z"],"dc:date.issued":["2025-12-19"],"dc:description.abstract":["The automation of software development prompted extensive research for decades. Recent advancements in AI, particularly large language models (LLMs), are driving transformative changes in the field. While LLMs have demonstrated impressive performance in code generation, challenges remain in generating complex and domain-specific code. To fully realize their potential, it is crucial to extend their application throughout the entire software development lifecycle. This necessity has driven us to explore and evaluate the role of these models in supporting all stages of the software engineering process. In this thesis, we propose an LLM-based framework tailored for the development of multiagent systems (MAS). By focusing on MAS and leveraging the MaSE methodology in conjunction with the JADE platform, our framework ensures the generation of accurate, traceable, and reliable software artefacts efficiently. The framework uses predefined prompts to guide the LLM through each step of the methodology, producing outputs that adhere to domain-specific requirements. Empirical evaluations on sample projects, through a case study and an experiment, demon-strate that this approach not only accelerates the development process but also results in artefacts that are more comprehensive and modular than those produced by human developers, showcasing the transformative potential of LLMs in software engineering."],"dc:identifier.doi":["https://dx.doi.org/10.11575/PRISM/50921"],"dc:identifier.uri":["https://hdl.handle.net/1880/123708"],"dc:language.iso":["en"],"dc:rights":["Unless otherwise indicated, this material is protected by copyright and has been made available with authorization from the copyright owner. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission."],"dc:subject":["Large Language Models","Multi-Agent Systems","Software Development Methodology","MaSE","Statistical Analysis"],"dc:title":["LLM-based MaSE - A Software Development Framework for Developing Multi-Agent System"],"dc:type":["master thesis"],"thesis:degree_discipline":["Engineering – Electrical & Computer"],"thesis:degree_name":["Master of Science (MSc)"],"thesis:institution_name":["University of Calgary"]},"updated_at":"2026-07-24T01:30:22Z"}