Cognitive Distortion Detector (Stage 1)

This model is a Turkish NLP classifier specialized in identifying cognitive distortions in text. It is designed as the initial component of a hierarchical classification system.

Model Description

The model acts as a binary classifier (Distorted vs. Not Distorted). Its primary goal is to filter text for potential cognitive biases before more specific analysis.

  • Task: Binary Classification
  • Language: Turkish
  • Focus: Mental health-related language processing

Hierarchical Workflow

This model serves as the Gatekeeper:

  1. Stage 1 (This Model): Detects if any cognitive distortion is present.
  2. Stage 2 (Subsequent Classifier): Categorizes the detected distortion into specific types (e.g., Catastrophizing, Overgeneralization, Personalization).

Usage

You can easily integrate this model using the Hugging Face pipeline:

from transformers import pipeline

pipe = pipeline("text-classification", model="mfurkanerkan15/cognitive-distortion-detector-tr")
result = pipe("Her şey her zaman kötü gitmek zorunda.")
print(result)
Downloads last month
24
Safetensors
Model size
0.1B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support