EUVIMEDEUVIMED entdecken
Crossref

Privacy Preservation and Personal Data Deletion: State-of-the-Art Review and a Proposed LLM-Based Multi-Agent Architecture

Hemali Dholam, Heeba Deshmukh, Sneha Ingale, Vidya Sagvekar

International Journal on Science and Technology · 06.11.2026 · 17 (4)

DOI: 10.71097/ijsat.v17.i4.11684 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal on Science and Technology, online erschienen am 06.11.2026. Verfasst von Hemali Dholam und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Privacy policies are the primary legal documents through which online platforms disclose their data practices, yet they remain difficult for users to understand and even harder to act upon. Parallel research has explored automated privacy policy analysis using natural language processing (NLP), large-scale web privacy measurement through browser instrumentation, and the emergence of agentic large language model (LLM) systems capable of tool use and decision making. However, these research directions largely exist in isolation. This review paper synthesizes advances across privacy policy Natural Language Processing (NLP), web privacy measurement frameworks, Retrieval - Augmented Generation (RAG) and knowledge-graph-enhanced retrieval for long and complex documents, agentic LLM architectures that combine reasoning with tool execution, and guardrails for safe autonomous agents. However, across these separate research threads, we identify a lack of any actionable direction for users to be able to control their own data. The paper highlights how the convergence of these technologies enables the design of an agentic privacy assistant that can interpret privacy policies, verify platform behavior against those policies, and autonomously execute user-authorized privacy actions such as data deletion; and outlines the architectural principles derived from current research to realize such a system.

Zitieren

APA

Dholam, H., Deshmukh, H., Ingale, S., & Sagvekar, V. (2026). Privacy Preservation and Personal Data Deletion: State-of-the-Art Review and a Proposed LLM-Based Multi-Agent Architecture. International Journal on Science and Technology, 17(4). https://doi.org/10.71097/ijsat.v17.i4.11684

Vancouver

Dholam H, Deshmukh H, Ingale S, Sagvekar V. Privacy Preservation and Personal Data Deletion: State-of-the-Art Review and a Proposed LLM-Based Multi-Agent Architecture. International Journal on Science and Technology. 2026;17(4). doi:10.71097/ijsat.v17.i4.11684

BibTeX
@article{Dholam2026,
  author = {Dholam, Hemali and Deshmukh, Heeba and Ingale, Sneha and Sagvekar, Vidya},
  title = {Privacy Preservation and Personal Data Deletion: State-of-the-Art Review and a Proposed LLM-Based Multi-Agent Architecture},
  journal = {International Journal on Science and Technology},
  year = {2026},
  volume = {17},
  number = {4},
  doi = {10.71097/ijsat.v17.i4.11684}
}
Kostenlos starten →