Moving Beyond ChatGPT: Local Large Language Models (LLMs) and the Secure Analysis of Confidential Unstructured Text Data in Social Work Research
Bibliographic record
Abstract
Purpose: Large language models (LLMs) have demonstrated remarkable abilities in natural language tasks. However, their use in social work research is limited by confidentiality and security concerns when processing sensitive data. This study addresses these challenges by evaluating the performance of local LLMs (LocalLLMs) in classifying and extracting substance-related problems from unstructured child welfare investigation summaries. LocalLLMs allow researchers to analyze data on their own computers without transmitting information to external servers for processing. Methods: Four state-of-the-art LocalLLMs—Mistral-7b, Mixtral-8 × 7b, LLama3-8b, and Llama3-70b—were tested using zero-shot prompting on 2,956 manually coded summaries. Results: The LocalLLMs achieved exceptional results comparable to human experts in classification and extraction, demonstrating their potential to unlock valuable insights from confidential, unstructured child welfare data. Conclusions: This study highlights the feasibility of using LocalLLMs to efficiently analyze large amounts of textual data while addressing the confidentiality issues associated with proprietary LLMs.
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 | 0.014 | 0.046 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.007 |
| Open science | 0.001 | 0.004 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".