No abstract

No abstract

Organisations disclose their privacy practices by posting privacy policies on their websites. Even though internet users often care about their digital privacy, they usually do not read privacy policies, since understanding them requires a significant investment of time and effort. Natural language processing has been used to create experimental tools to interpret privacy policies, but there has been a lack of large privacy policy corpora to facilitate the creation of large-scale semi-supervised and unsupervised models to interpret and simplify privacy policies. Thus, we present the Pri-vaSeer Corpus of 1,005,380 English language website privacy policies collected from the web. The number of unique websites represented in PrivaSeer is about ten times larger than the next largest public collection of web privacy policies, and it surpasses the aggregate of unique websites represented in all other publicly available privacy policy corpora combined. We describe a corpus creation pipeline with stages that include a web crawler, language detection, document classification, duplicate and near-duplicate removal, and content extraction. We employ an unsupervised topic modelling approach to investigate the contents of policy documents in the corpus and discuss the distribution of topics in privacy policies at web scale. We further investigate the relationship between privacy policy domain PageRanks and text features of the privacy policies. Finally, we use the corpus to pretrain PrivBERT, a transformer-based privacy policy language model, and obtain state of the art results on the data practice classification and question answering tasks.

No abstract

Web privacy policies are used by organisations to disclose their privacy practices to users on the web. However, users often do not read privacy policies because they are too long, time consuming, or too complicated. Attempts to simplify privacy policies using natural language processing have achieved some success, but they face limitations of scalability and generalization. While this puts an onus on researchers and policy regulators to protect users against unfair privacy practices, they often lack a large-scale collection of policies to study the state of internet privacy. To remedy this bottleneck, we present PrivaSeer, the first privacy policy search engine. PrivaSeer has been indexed on 1,400,318 English language website privacy policies and can be used to search privacy policies based on text queries and several search facets. Results can be ranked by PageRank, query-based document relevance, and the probability that a document is a privacy policy. Results also can be filtered by readability, vagueness, industry, and mentions of tracking technology, self-regulatory bodies, or regulations and cross-border agreements in the policy text. PrivaSeer allows legal experts, researchers, and policy regulators to discover privacy trends and policy anomalies in privacy policies at scale. In this paper we present the search interface, ranking technique, and filtering techniques for PrivaSeer. We create two indexes of privacy policies: one including supplementary non-policy content present in privacy policy web pages and one without. We evaluate the functionality of PrivaSeer by comparing ranking techniques on these two indexes.