Benchmarking Large Language Models: A Comparative Study of DeepSeek and ChatGPT Across Diverse Domains
Bibliographic record
Abstract
The growing integration of Large Language Models (LLMs) like ChatGPT and DeepSeek across education and professional domains highlights the need to understand their domain-specific performance and response behavior. This study evaluates 400 prompts and 800 responses across four core fields—Business, Healthcare, Mathematics, and Neuroscience—using both simple and Retrieval-Augmented Generation (RAG) prompts. Responses were rated on Accuracy, Relevance, Complexity, and Runtime using a 5-point Likert scale. Correlation analysis using Pearson coefficients revealed key relationships between evaluation metrics, showing that Relevance and Accuracy are strongly aligned in high-performing domains, while Complexity often correlates with longer response times. Our findings indicate that ChatGPT excels in producing accurate and efficient outputs, while DeepSeek demonstrates strength in generating contextually rich and complex responses, particularly with RAG prompts. This research provides practical insights into how prompt design impacts model behavior and offers actionable recommendations for educators, developers, and learners to optimize prompt strategies when using generative Artificial Intelligence (AI) tools. The study underscores the role of prompt engineering in enhancing LLM performance and supports the development of tailored, domain-aware AI applications in education and beyond.
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 | 0.013 | 0.078 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".