Calpric: Inclusive and Fine-grain Labeling of Privacy Policies with Crowdsourcing and Active Learning

TL;DR

Calpric introduces a cost-effective method combining automated segmentation, crowdsourcing, and active learning to generate detailed, balanced privacy policy datasets, enabling more accurate deep learning models.

Contribution

It presents a novel integrated approach that reduces labeling costs and improves data quality for privacy policy analysis using crowdsourcing and active learning.

Findings

Generated 16K labeled privacy policy segments across 9 categories

Achieved labeling costs of roughly $0.92-$1.71 per segment

Produced models with improved accuracy and fine-grain labeling

Abstract

A significant challenge to training accurate deep learning models on privacy policies is the cost and difficulty of obtaining a large and comprehensive set of training data. To address these challenges, we present Calpric , which combines automatic text selection and segmentation, active learning and the use of crowdsourced annotators to generate a large, balanced training set for privacy policies at low cost. Automated text selection and segmentation simplifies the labeling task, enabling untrained annotators from crowdsourcing platforms, like Amazon's Mechanical Turk, to be competitive with trained annotators, such as law students, and also reduces inter-annotator agreement, which decreases labeling cost. Having reliable labels for training enables the use of active learning, which uses fewer training samples to efficiently cover the input space, further reducing cost and improving class and data category balance in the data set. The combination of these techniques allows Calpric to produce models that are accurate over a wider range of data categories, and provide more detailed, fine-grain labels than previous work. Our crowdsourcing process enables Calpric to attain reliable labeled data at a cost of roughly $0.92-$1.71 per labeled text segment. Calpric 's training process also generates a labeled data set of 16K privacy policy text segments across 9 Data categories with balanced positive and negative samples.