Development of an RAG-Based LLM Chatbot for Enhancing Technical Support Service
Bibliographic record
Abstract
The global shortage of manpower for technical support is a critical issue in the digital transformation era. Recently, Large Language Models (LLMs) have made significant strides in natural language processing, leading to the development of AI chatbots to address this problem. However, LLMs have notable limitations in handling domain-specific information, often generating incorrect responses when queries go beyond the coverage of the training data or require the most up-to-date information. A promising solution is the Retrieval-Augmented Generation (RAG) approach, which incorporates domain-specific data retrieval into the generative process. Our team has developed a domain-specific and RAG-based LLM chatbot to enhance the software house technical support of an IT consultant in Canada. The chatbot was implemented and evaluated in real-world production environments. Preliminary results show that the system has achieved high scores of 38%, 188%, and 40% in the ROUGE-I, ROUGE-2, and ROUGE-L measures, respectively, compared to using only a general LLM model. End-user feedback also reflected that the enhanced system produced more accurate and efficient replies, thereby enhancing overall customer satisfaction.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.014 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".