Intent Classification and Slot Filling for Privacy Policies
This work provides a new benchmark and baseline models for natural language processing tasks (intent classification and slot filling) specifically tailored for privacy policies, which is important for users to understand their data privacy.
This paper introduces PolicyIE, a new English corpus of 5,250 intent and 11,788 slot annotations from 31 privacy policies, to address the challenge of understanding privacy policies. The authors evaluate two neural baselines, finding that a sequence-to-sequence method significantly outperforms a joint sequence tagging approach for slot filling.
Understanding privacy policies is crucial for users as it empowers them to learn about the information that matters to them. Sentences written in a privacy policy document explain privacy practices, and the constituent text spans convey further specific information about that practice. We refer to predicting the privacy practice explained in a sentence as intent classification and identifying the text spans sharing specific information as slot filling. In this work, we propose PolicyIE, an English corpus consisting of 5,250 intent and 11,788 slot annotations spanning 31 privacy policies of websites and mobile applications. PolicyIE corpus is a challenging real-world benchmark with limited labeled examples reflecting the cost of collecting large-scale annotations from domain experts. We present two alternative neural approaches as baselines, (1) intent classification and slot filling as a joint sequence tagging and (2) modeling them as a sequence-to-sequence (Seq2Seq) learning task. The experiment results show that both approaches perform comparably in intent classification, while the Seq2Seq method outperforms the sequence tagging approach in slot filling by a large margin. We perform a detailed error analysis to reveal the challenges of the proposed corpus.