Summary of Using Zero-shot Prompting in the Automatic Creation and Expansion Of Topic Taxonomies For Tagging Retail Banking Transactions, by Daniel De S. Moraes et al.
Using Zero-shot Prompting in the Automatic Creation and Expansion of Topic Taxonomies for Tagging Retail Banking Transactions
by Daniel de S. Moraes, Pedro T. C. Santos, Polyana B. da Costa, Matheus A. S. Pinto, Ivan de J. P. Pinto, Álvaro M. G. da Veiga, Sergio Colcher, Antonio J. G. Busson, Rafael H. Rocha, Rennan Gaio, Rafael Miceli, Gabriela Tourinho, Marcos Rabaioli, Leandro Santos, Fellipe Marques, David Favaro
First submitted to arxiv on: 8 Jan 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper introduces an unsupervised approach to constructing and expanding topic taxonomies using instruction-based fine-tuned Large Language Models (LLMs). By combining topic modeling and keyword extraction techniques, the method creates initial topic taxonomies that are then post-processed by LLMs to create a hierarchical structure. The approach is novel in its application of zero-shot prompting to expand an existing taxonomy with new terms. To evaluate the effectiveness of this methodology, the authors used a retail bank dataset to assign tags to merchants based on the created taxonomies. The results showed high coherence rates (over 90%) and promising parent node prediction performance (f1-score above 70%). This research has potential applications in information retrieval, natural language processing, and knowledge management. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way for computers to organize and expand topics without needing human help. It uses special kinds of artificial intelligence called Large Language Models (LLMs) to build topic taxonomies and make them more detailed. The approach is unique because it lets the computer figure out where to add new information to an existing taxonomy without being trained on that specific task before. To test this method, researchers used a dataset about merchants from a bank and asked people to rate how well the resulting taxonomies worked. The results showed that most of the topics were coherent (over 90%) and that the computer was good at predicting parent nodes (f1-score above 70%). This research could be important for how we find and understand information in the future. |
Keywords
» Artificial intelligence » F1 score » Natural language processing » Prompting » Unsupervised » Zero shot
