Spatial intelligence and contextual relevance in AI-driven health information retrieval
Bibliographic record
Abstract
The evolution of large language models (LLMs) has already significantly influenced online health information retrieval. As these models gain more widespread use, it is important to understand their ability to contextualize responses based on spatial and geographic information. This study investigates whether LLMs can vary responses based on geographic and spatial context. Using a structured set of prompts submitted to ChatGPT, responses were analyzed to discern patterns based on prompt question and geographic identifiers included in queries. The analysis used word frequency analysis and bidirectional encoder representations from transformers (BERT) embeddings to evaluate the variation in responses concerning geographic specificity. The results provide some evidence that LLMs can generate geographically tailored responses when the query specifies such a need, thereby supporting localized information retrieval. Moreover, prompt responses exhibit an association between spatial distance and word frequency/sentence embedding differences between texts. This result suggests a nuanced representation of spatial information, which could impact user experience by providing more relevant health information based on the user's location. This study lays the groundwork for further exploration into the spatial intelligence of LLMs and their impact on the accessibility of health information online. • Large language models, like ChatGPT, can be used to search for health information. • Responses show an association between spatial distance and differences between texts. • ChatGPT may discriminate as to when geographic context matters in responses. • Further work is needed to determine if these results are genralizable
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".