| KEYWORD |
LLM-enhanced Retrieval for Data Analytics
keywords ARTIFICIAL INTELLIGENCE, CHAT-GPT, DATA ANALYSIS, DATA ANALYTICS, DATA SCIENCE, LLM, MACHINE LEARNING, MACHINE LEARNING, ARTIFICIAL NEURAL NETWORKS
Reference persons DANIELE APILETTI
Research Groups DAUIN - GR-04 - DATABASE AND DATA MINING GROUP - DBDM
Thesis type ANALYTICAL AND EXPERIMENTAL, DEVELOPMENT, SPERIMENTAL APPLIED
Description Create an integrated system using heterogenous data, from sales to surveys, to train end-to-end algorithms, leveraging both traditional deep learning models and large language models (LLMs) like LLAMA/ChatGPT, where deep learning models validate LLM-generated results based on structural knowledge.
Project Overview:
- Combine heterogenous data, from sales to surveys, for training a comprehensive statistical data analytics system
- Employ traditional deep learning models and LLMs to produce analytic reports
- Use deep learning models to validate and enhance the accuracy of LLM-generated outputs through structural knowledge integration
The final goal is to generate high-quality analytic reports from diverse sources, investigating the possibility of enhancing LLMs reliability through deep learning model validation. Enable more robust decision-making based on accurate, integrated data.
Required skills Python, coding
Deadline 19/02/2025
PROPONI LA TUA CANDIDATURA