metadata

license: other
language:
  - en
pipeline_tag: text-classification
tags:
  - legal
  - privacy
  - intent
  - privacy policies

privacy_intent for privacy policy intent classification

This model is fine-tuned version of mukund/privbert model on PolicyIE dataset .

Reference Paper: Intent Classification and Slot Filling for Privacy Policies.
The back translation method (data augmentation) resulted in a 1% improvement in performance when applied to imbalanced samples
F1 Score: 88 (%4 performance increase compared to original work)

5 Intents (Labels):

(1) Data Collection/Usage: What, why and how user information is collected;

(2) Data Sharing/Disclosure: What, why and how user information is shared with or collected by third parties;

(3) Data Storage/Retention: How long and where user information will be stored;

(4) Data Security/Protection: Protection measures for user information;

(5) Other: Other privacy practices that do not fall into the above four categories.

from transformers import pipeline

pipe = pipeline("text-classification", "remzicam/privacy_intent")
text="At any time during your use of the Services, you may decide to share some information or content publicly or privately."
pipe(text)

Output

[{'label': 'data-sharing-disclosure', 'score': 0.8373807072639465}]