Abstract

Abstract
Researchers rely on confidential and sensitive microdata provided by national statistical institutes and other organizations, making disclosure control a critical challenge. Manual output checking processes are time-consuming and require expert knowledge, limiting scalability. This paper presents an automated framework that integrates large language models (LLMs), prompt engineering, and workflow automation (n8n) for statistical disclosure control (SDC). The system introduces an AI-driven output validation, including code generation, data processing, and risk assessment, allowing researchers to pre-check their outputs via a Seamless Within-Activity Review (SWAR) approach. Key challenges such as computational costs, confidentiality concerns, and the need for human oversight are addressed, and reinforcement learning is proposed to enhance future risk evaluation. The framework marks a step toward scalable, privacy-compliant, AI-assisted disclosure control in official statistics.

Abstract
Deep learning and large language models (LLMs) have recently enabled studies in state-of-the-art technologies that enhance recommender systems. This research focuses on solving the next-item recommendation problem using these challenging technologies in Web applications, specifically focusing on a case study in the wine domain. This paper presents the characterization of the framework developed for the object of study: adaptive recommendation based on new modeling of the initial data to explore the user's dynamic taste profile. Following the design science research methodology, the following contributions are presented: (i) a novel dataset of wines called X-Wines; (ii) an updated recommender model called X-Model4Rec-eXtensible Model for Recommendation supported in attention and transformer mechanisms which constitute the core of the LLMs; and (iii) a collaborative Web platform to support adaptive wine recommendation to users in an online environment. The results indicate that the solutions proposed in this research can improve recommendations in online environments and promote further scientific work on specific topics.

Abstract

Abstract

Abstract