Back to search

Institutional Repository of Vilnius University

Didžiųjų kalbos modelių užklausų kūrimo sistema skaitmeninės rinkodaros kontekste /

Abstract

dc:description

This bachelor’s thesis examines the significance and application of large language model prompt creation in the context of digital marketing. The use of large language models is increasingly integrated into content creation and communication planning processes; however, their effectiveness directly depends on clearly defined and structured prompts. An unclear prompt structure, insufficient contextual information, and inconsistent prompt formulation lead to the manifestation of LLM limitations, increased time consumption, and the need for repeated revisions. Therefore, the need for structured prompt creation solutions remains relevant in digital marketing practice. The aim of the thesis is to investigate prompt creation solutions for large language models in the digital marketing context and, based on the obtained results, to develop a prototype of a structured prompt creation system. The research objectives included analyzing the application of large language models in digital marketing, examining the significance and structuring principles of prompt creation, evaluating the functionalities of existing prompt creation solutions, and designing, implementing, and testing the proposed system prototype. The research employed methods of scientific literature analysis, comparative analysis, and a semi-structured expert interview. The theoretical part analyzes aspects of large language model application in digital marketing and principles of structured prompt creation. The empirical part presents a comparative analysis of existing prompt creation systems and a semi-structured interview with digital marketing specialists, whose insights were used to substantiate the specification of the proposed system. Based on the research results, the system specification was formulated, the information system was designed, and a functional prototype of a structured prompt creation system was implemented. UML notation was applied during the design phase, while the implementation was carried out in the Visual Studio Code environment using the Next.js framework, the TypeScript programming language, integration with the OpenAI API, data storage on the Supabase platform, and deployment on the Vercel infrastructure. The developed system enables the collection of contextual information using a questionnaire-based approach, the formation of prompts based on structural principles for “ChatGPT” large language models, the management of created and saved prompt histories, and the administration of system-level prompt formulations. The results demonstrate that questionnaire-based structured prompt creation reduces the need for repeated corrections and creates conditions for obtaining more consistent responses from large language models in digital marketing tasks.

Degree

thesis:*
Grantor dc:publisher
Institutional Repository of Vilnius University
Year dc:date
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Vainauskaitė, Aušrinė,

Rights

dc:rights
Statement dc:rights
  • info:eu-repo/semantics/openAccess
Language dc:language
lit

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Vilnius University
Base URL
epublications.vu.lt/oai
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Vainauskaitė, Aušrinė,. Didžiųjų kalbos modelių užklausų kūrimo sistema skaitmeninės rinkodaros kontekste /. Institutional Repository of Vilnius University, 2026. https://repository.vu.lt/VU:ELABAETD277692576&prefLang=en_US