Banks are usually large and complex companies that face a number of challenges to support the rapid and effective sharing of information and content across their organizations. Extracting complex metadata from raw bank documents is therefore central to support intelligent data indexing, information circulation and to promote more complex predictive capabilities, e.g., compliance assessment problems. In this paper, we present a weakly-supervised neural methodology for creating semantic metadata from bank documents. It exploits a neural pre-training method optimized against legacy semantic resources able to minimize the training effort. We studied an application to business process design and management in banks and tested the method on documents from the Italian banking community. The measured impact of the proposed training approach to process-related metadata creation confirms its applicability.

Margiotta, D., Croce, D., Rotoloni, M., Cacciamani, B., Basili, R. (2022). Knowledge-Based Neural Pre-training for Intelligent Document Management. In AIxIA 2021 – Advances in Artificial Intelligence (pp.564-579). GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND : Springer Science and Business Media Deutschland GmbH [10.1007/978-3-031-08421-8_39].

Knowledge-Based Neural Pre-training for Intelligent Document Management

Margiotta D.;Croce D.;Basili R.
2022-01-01

Abstract

Banks are usually large and complex companies that face a number of challenges to support the rapid and effective sharing of information and content across their organizations. Extracting complex metadata from raw bank documents is therefore central to support intelligent data indexing, information circulation and to promote more complex predictive capabilities, e.g., compliance assessment problems. In this paper, we present a weakly-supervised neural methodology for creating semantic metadata from bank documents. It exploits a neural pre-training method optimized against legacy semantic resources able to minimize the training effort. We studied an application to business process design and management in banks and tested the method on documents from the Italian banking community. The measured impact of the proposed training approach to process-related metadata creation confirms its applicability.
20th International Conference of the Italian Association for Artificial Intelligence, AIxIA 2021
2021
20
Rilevanza internazionale
2022
Settore INF/01
Settore ING-INF/05
English
Bert-based NL inference
Domain knowledge modeling
Domain-specific neural learning
Zero-shot learning in NLP
Intervento a convegno
Margiotta, D., Croce, D., Rotoloni, M., Cacciamani, B., Basili, R. (2022). Knowledge-Based Neural Pre-training for Intelligent Document Management. In AIxIA 2021 – Advances in Artificial Intelligence (pp.564-579). GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND : Springer Science and Business Media Deutschland GmbH [10.1007/978-3-031-08421-8_39].
Margiotta, D; Croce, D; Rotoloni, M; Cacciamani, B; Basili, R
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2108/359273
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact