Transforming Sensitive Documents into Quantitative Data: An AI-Based Preprocessing Toolchain for Structured and Privacy-Conscious Analysis

TL;DR

This paper introduces a privacy-preserving, modular AI-based preprocessing toolchain that standardizes, anonymizes, and transforms unstructured sensitive text data into structured embeddings, facilitating large-scale analysis in privacy-sensitive fields.

Contribution

The authors develop an open-weight, local hardware-compatible toolchain that standardizes, anonymizes, and converts sensitive unstructured text into embeddings, enabling scalable, privacy-conscious research.

Findings

Effective removal of personally identifiable information

High semantic content retention after anonymization

Successful application to Swedish court decision corpus

Abstract

Unstructured text from legal, medical, and administrative sources offers a rich but underutilized resource for research in public health and the social sciences. However, large-scale analysis is hampered by two key challenges: the presence of sensitive, personally identifiable information, and significant heterogeneity in structure and language. We present a modular toolchain that prepares such text data for embedding-based analysis, relying entirely on open-weight models that run on local hardware, requiring only a workstation-level GPU and supporting privacy-sensitive research. The toolchain employs large language model (LLM) prompting to standardize, summarize, and, when needed, translate texts to English for greater comparability. Anonymization is achieved via LLM-based redaction, supplemented with named entity recognition and rule-based methods to minimize the risk of disclosure. We demonstrate the toolchain on a corpus of 10,842 Swedish court decisions under the Care of Abusers Act (LVM), comprising over 56,000 pages. Each document is processed into an anonymized, standardized summary and transformed into a document-level embedding. Validation, including manual review, automated scanning, and predictive evaluation shows the toolchain effectively removes identifying information while retaining semantic content. As an illustrative application, we train a predictive model using embedding vectors derived from a small set of manually labeled summaries, demonstrating the toolchain's capacity for semi-automated content analysis at scale. By enabling structured, privacy-conscious analysis of sensitive documents, our toolchain opens new possibilities for large-scale research in domains where textual data was previously inaccessible due to privacy and heterogeneity constraints.