Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/149544
Title: OSINT Infohound – Síntesis de datos de fuentes abiertas por medio de modelos de lenguaje de gran tamaño (LLM)
Author: Casado Herrero, Marcos
Tutor: Guijarro, Jordi  
Others: Garcia-Font, Victor  
Abstract: This thesis analyses the feasibility of integrating large language models (LLMs) into open-source intelligence (OSINT) collection platforms, with the aim of improving the efficiency of open-source intelligence analysis. It is argued that LLMs can add versatility, flexibility, and value to these platforms, enabling the search, analysis, and synthesis of large amounts of data. To illustrate this idea, a practical case of integrating an LLM into the InfoHound platform is presented. InfoHound is a tool from the research and innovation institute i2cat of Catalunya that allows organisations to perform reverse analysis on information indexed about them. Integrating an LLM into this platform would open up a wide range of possibilities, such as synthesising the CVs of individuals associated with an organisation, or classifying individuals based on their political thinking derived from social media information. The practical case study applied to InfoHound consists of collecting user profiles from open-sources and storing their data in a disorganised way, with different formats or sources, so that later an LLM model container can be asked to analyse the information and generate a professional summary for each person collected.
Keywords: OSINT
AI
LLM
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 9-Jan-2024
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
mcasadohTFM0124memoria.pdfMemoria del TFM1,75 MBAdobe PDFThumbnail
View/Open
Share:
Export:
View statistics

This item is licensed under aCreative Commons License Creative Commons