Avtomatizirano odgovarjanje na e-pošto s pomočjo sistema rag in vektorske baze

Poredoš, Rok

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Avtomatizirano odgovarjanje na e-pošto s pomočjo sistema rag in vektorske baze
Authors:	ID Poredoš, Rok (Author) ID Rajkovič, Uroš (Mentor) More about this mentor...
Files:	VS_Poredos_Rok_2025.pdf (3,02 MB) MD5: B64372BF3E3BBAAE358CB3F2288B1BDE
Language:	Slovenian
Work type:	Bachelor thesis/paper
Typology:	2.11 - Undergraduate Thesis
Organization:	FOV - Faculty of Organizational Sciences in Kranj
Abstract:	V digitalnem okolju se podjetja soočajo z velikim obsegom ponavljajočih se e-poštnih vprašanj. To diplomsko delo je naslovilo ta izziv z razvojem in implementacijo funkcionalnega prototipa avtomatiziranega sistema za generiranje odgovorov. Glavni cilj je bil zmanjšati čas, potreben za pripravo odgovorov, in povečati učinkovitost dela. Projekt je uporabil pristop Retrieval-Augmented Generation (RAG) v kombinaciji z vektorsko bazo podatkov, pri čemer so bili uporabljeni n8n, Docker in PostgreSQL z razširitvijo pgvector. Za generiranje odgovorov so bili uporabljeni veliki jezikovni modeli (LLM), dostopni prek storitve Gemini. V okviru eksperimentalnega dela je bil razvit delovni tok, ki avtomatizira obravnavo e-pošte in uspešno dokazuje, da sistem hitro najde relevantne informacije in ustvari ustrezen osnutek odgovora. S testiranjem je bilo potrjeno, da sistem zmanjša čas priprave odgovorov za povprečno 85 %. Dodana vrednost te raziskave je v predstavitvi praktične rešitve, ki predstavlja pomembno izhodišče za nadaljnji razvoj avtomatiziranih sistemov v poslovnem okolju.
Keywords:	avtomatizirano odgovarjanje na e-pošto, delovni tokovi, RAG, LLM
Place of publishing:	Maribor
Year of publishing:	2025
PID:	20.500.12556/DKUM-95265
COBISS.SI-ID:	254142723
Publication date in DKUM:	21.10.2025
Views:	123
Downloads:	47
Metadata:
Categories:	FOV
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:	CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International

Link:	http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:	The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:	10.09.2025

Secondary language

Language:	English
Title:	Automated email response generation using a rag system and a vector database
Abstract:	In the digital environment, companies face a large volume of repetitive email inquiries. This thesis addressed this challenge by developing and implementing a functional prototype of an automated system for generating responses. The main goal was to reduce the time required to prepare responses and increase work efficiency. The project used a Retrieval-Augmented Generation (RAG) approach in combination with a vector database, utilizing tools such as n8n, Docker, and PostgreSQL with the pgvector extension. Large language models (LLMs), accessible via the Gemini service, were used to generate responses. Within the experimental part, a workflow was developed that automates email processing and successfully demonstrates that the system quickly finds relevant information and creates a suitable draft response. Testing confirmed that the system reduces the response preparation time by an average of 85 %. The added value of this research lies in presenting a practical solution that provides an important starting point for the further development of automated systems in a business environment.
Keywords:	automated email response, workflows, RAG, LLM

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back