Downloads · 30 days
0
HarmoniCA_anxiety
julia-pfarr/HarmoniCA_anxiety
HarmoniCA_anxiety is a text classification model from julia-pfarr. Use it when you need a label for a piece of text. It is set up for sentence-transformers. The card lists the license as cc-by-4.0.
Part of the HarmoniCA collection for harmonising psychiatric questionnaire items across studies. This model assigns items from anxiety questionnaires to one of two theoretically motivated symptom dimensions.
Access
Public
Updated Jun 8, 2026
Repo size
1.3 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.3 GB · 100%
At a glance
- Task
- Text Classification
- Library
- sentence-transformers
- License
- cc-by-4.0
- Access
- Public
- Created
- Jun 8, 2026
- Updated
- Jun 8, 2026
- SHA
- 9b2318ef
Try a prompt
Base models
Spaces
Model card
From the Hugging Face model README
HarmoniCA — Anxiety Model
Part of the HarmoniCA collection for harmonising psychiatric questionnaire items across studies. This model assigns items from anxiety questionnaires to one of two theoretically motivated symptom dimensions.
Dimensions
| ID | Label | Description |
|---|---|---|
| 1 | Somatic Anxiety | physiological manifestations of anxiety: racing heart, sweating, trembling, shortness of breath, autonomic symptoms |
| 2 | Cognitive Anxiety | worry, fear, apprehension, panic, distressing thoughts, anticipation of negative events |
Items rated as not belonging to the anxiety construct are assigned dimension -1 (Does not fit).
Model Architecture
- Type: ContrastiveModel (contrastive fine-tuning of a sentence transformer)
- Base encoder: sentence-transformers/gtr-t5-large
Training Data
- Training items: 159 expert-labeled questionnaire items (144 original + 10 augmented + 5 expert-reviewed)
- Questionnaires: ASensI, BAI, GAD-7, HADS, HAMA, MDS-UPDRS, NMSQuest, NPI, PAS, PSWQ, SCL-90, SMGAD, STAI, plus data augmentation
- Labels: Expert consensus from two rounds of surveys (psychiatrists/neurologists)
Evaluation
Evaluated on 66 held-out items from unseen questionnaires against expert labels.
| Cohen's κ | Accuracy |
|---|---|
| 0.877 | 93.9% |
Cross-validation (leave-one-questionnaire-out, 13 folds): mean accuracy 93.7% ± 10.7%.
Limitations
- Trained and validated on Parkinson's disease cohort questionnaires; performance may differ in other populations.
- Dimension structure reflects theoretical consensus among a specific expert panel and may not match all theoretical frameworks.
Citation
in preparation
- Task
- Text Classification
- Library
- sentence-transformers
- License
- cc-by-4.0
- Languages
- en
- Created
- Jun 8, 2026
- Updated
- Jun 8, 2026