Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
PersonaBOT: Bringing Customer Personas to Life with LLMs and RAG: Generating and Utilizing Customer Personas with LLMs and Retrieval-Augmented Generation
Jönköping University, School of Engineering, JTH, Department of Computing, Jönköping AI Lab (JAIL).
2025 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

The introduction of Large Language Models (LLMs) has significantly transformed Natural LanguageProcessing (NLP) applications by enabling more advanced analysis of customer personas. At VolvoConstruction Equipment (VCE), customer personas are traditionally developed through qualitativemethods, which are time-consuming and lack scalability. The main objective of this research is to generate synthetic customer personas and integrate them into a Retrieval-Augmented Generation (RAG)chatbot to support decision-making in business processes. The study utilizes the Design Science Research Methodology (DSRM) in three iterative cycles. The first iteration focuses on developing apersona-based RAG chatbot integrated with verified personas. In the second iteration, synthetic personas were generated using Few-Shot and Chain-of-Thought (CoT) prompting techniques and evaluated based on completeness, relevance, and consistency using McNemar’s test. In the final iteration,the chatbot knowledge base is augmented with synthetic personas and additional segment informationto assess improvements in response accuracy and practical utility. Key findings indicate that Few-Shotprompting outperformed CoT in generating more complete personas, while CoT demonstrated greaterefficiency in terms of response time and token usage. After augmenting the knowledge base, the average accuracy rating of the chatbot increased from 5.88 to 6.42 on a 10-point scale, and 81.82% ofparticipants found the updated system useful in business contexts.

Place, publisher, year, edition, pages
2025. , p. 49
Keywords [en]
LLM, Prompt Engineering, Customer Persona, RAG
National Category
Computer and Information Sciences Artificial Intelligence
Identifiers
URN: urn:nbn:se:hj:diva-68438OAI: oai:DiVA.org:hj-68438DiVA, id: diva2:1968625
External cooperation
Volvo Construction Equipment
Subject / course
JTH, Computer Engineering
Available from: 2025-06-13 Created: 2025-06-13 Last updated: 2025-10-13Bibliographically approved

Open Access in DiVA

fulltext(1277 kB)460 downloads
File information
File name FULLTEXT02.pdfFile size 1277 kBChecksum SHA-512
399e1462b815d30a8d8c73bb3786470b6e2d3454eea98371555331b76aa7a06810c748264a0847a9797ef771806fb4f1ec4744a8acb9a099b7e656ce79249d34
Type fulltextMimetype application/pdf

By organisation
Jönköping AI Lab (JAIL)
Computer and Information SciencesArtificial Intelligence

Search outside of DiVA

GoogleGoogle Scholar
Total: 460 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 810 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf