Objective
The project was born as a solution to a problem in managing HelpDesk inquiries for Human Resources in a telecommunications company. The resource consumption of the current service has been evaluated and compared with a solution developed using generative artificial intelligence (GenAI) in the form of a chatbot.
The chatbot is designed to provide accurate responses to employees by consulting personal policies through a retrieval-augmented generation (RAG) system, which searches for information in relevant documents.
For knowledge storage, both vector and relational databases were analyzed, with the relational database being chosen as the solution. This includes pre-processing of the text which is stored separately according to the index. Experiments have shown that this method provides better results than vector databases.
Experiments have shown that this method provides better results than vector databases.
BACHELOR’S THESIS BY:
CARMEN ABELLA MIRAVET
Academic Experience
- Dual Bachelor in Computer Engineer and Business Administration, University Carlos III de Madrid (september 2018 – may2024)
- Exchange year in University of Southern California: Marshall Business School, Los Angeles, USA (september 2022 – may2023)
Work Experience
- Data Science – Applied Intelligence Accenture (february 2024 – present)
- Research program in artificial intelligence with the Applied Artificial Intelligence UC3M group (september 2023 – may 2024)
Awards and Certifications
- Best creative idea in digital transformation in occupational safety and health – UC3M
- Academic Excellence Scholarship awarded by the Community of Madrid 2020-2023
Technical skills
- Programming languages Python, Java, C, SQL, R
- Development libraries: Pandas, Numpy, Tensorflow, Sci-kit Learn.