eprintid: 10195658 rev_number: 6 eprint_status: archive userid: 699 dir: disk0/10/19/56/58 datestamp: 2024-08-09 13:03:29 lastmod: 2024-08-09 13:03:29 status_changed: 2024-08-09 13:03:29 type: article metadata_visibility: show sword_depositor: 699 creators_name: Kurylek, Bartosz creators_name: Camara, Arthur creators_name: Nandi, Akash creators_name: Markopoulos, Evangelos title: A Novel Agent-Based Framework for Conversational Data Analysis and Personal AI Systems ispublished: pub divisions: UCL divisions: B04 divisions: F49 keywords: Conversational AI, Agent-Based Systems, Large Language Models (LLMs) note: © The Authors 2024. The authors of papers published in the AHFE Open Access Proceedings will retain full copyrights as specified by the provisions of the Creative Commons: (http://creativecommons.org/licenses/by/4.0/). abstract: This paper introduces a novel agent-based framework that leverages conversational data to enhance Large Language Models (LLMs) with personalized knowledge, enabling the creation of Artificial Personal Intelligence (API) systems. The proposed framework addresses the challenge of collecting and analysing unstructured conversational data by utilizing LLM agents and embeddings to efficiently process, organize, and extract insights from conversations. The system architecture integrates knowledge data aggregation and agent-based conversational data extraction. The knowledge data aggregation method employs LLMs and embeddings to create a dynamic, multi-level hierarchy for organizing information based on conceptual similarity and topical relevance. The agent-based component utilizes an LLM Agent to handle user queries, extracting relevant information and generating specialized theme datasets for comprehensive analysis. The framework's effectiveness is demonstrated through empirical analysis of real-world conversational data and a user survey. However, limitations such as the need for further testing of scalability and performance under large-scale, real-world conditions and potential biases introduced by LLMs are acknowledged. Future research should focus on extensive real-world testing and the integration of additional conversational qualities to further enhance the framework's capabilities, ultimately enabling more personalized and context-aware AI assistance. date: 2024 date_type: published official_url: https://doi.org/10.54941/ahfe1004649 oa_status: green full_text_type: pub language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 2303991 doi: 10.54941/ahfe1004649 lyricists_name: Markopoulos, Evangelos lyricists_id: EMARK80 actors_name: Markopoulos, Evangelos actors_name: Kalinowski, Damian actors_id: EMARK80 actors_id: DKALI47 actors_role: owner actors_role: impersonator full_text_status: public publication: Artificial Intelligence and Social Computing volume: 122 pagerange: 124-136 citation: Kurylek, Bartosz; Camara, Arthur; Nandi, Akash; Markopoulos, Evangelos; (2024) A Novel Agent-Based Framework for Conversational Data Analysis and Personal AI Systems. Artificial Intelligence and Social Computing , 122 pp. 124-136. 10.54941/ahfe1004649 <https://doi.org/10.54941/ahfe1004649>. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10195658/1/Markopoulos_A%20Novel%20Agent-Based%20Framework%20for%20Conversational%20Data%20Analysis%20and%20Personal%20AI%20Systems_VoR.pdf