Towards automating Numerical Consistency Checks in Financial Reports
We introduce KPI-Check, a novel system that automatically identifies and cross-checks semantically equivalent key performance indicators (KPIs), e.g. „revenue“ or „total costs“, in real-world German financial reports. It combines a financial named entity and relation extraction module with a BERT-based filtering and text pair classification component to extract KPIs from unstructured sentences before linking them to synonymous occurrences in the balance sheet and profit \& loss statement. The tool achieves a high matching performance of 73.00\% micro F1 on a hold out test set and is currently being deployed for a globally operating major auditing firm to assist the auditing procedure of financial statements.
- Veröffentlicht in:
2022 IEEE International Conference on Big Data (Big Data) - Typ:
Inproceedings - Autoren:
- Jahr:
2022 - Source:
https://www.computer.org/csdl/proceedings-article/big-data/2022/10020308/1KfRFDJVlbq
Informationen zur Zitierung
: Towards automating Numerical Consistency Checks in Financial Reports, 2022 IEEE International Conference on Big Data (Big Data), 2022, https://www.computer.org/csdl/proceedings-article/big-data/2022/10020308/1KfRFDJVlbq, Hillebrand.etal.2022a,
@Inproceedings{Hillebrand.etal.2022a,
author={Hillebrand, Lars; Deußer, Tobias; Dilmaghani, Tim; Kliem, Bernd; Loitz, Rüdiger; Bauckhage, Christian; Sifa, Rafet},
title={Towards automating Numerical Consistency Checks in Financial Reports},
booktitle={2022 IEEE International Conference on Big Data (Big Data)},
url={https://www.computer.org/csdl/proceedings-article/big-data/2022/10020308/1KfRFDJVlbq},
year={2022},
abstract={We introduce KPI-Check, a novel system that automatically identifies and cross-checks semantically equivalent key performance indicators (KPIs), e.g. „revenue“ or „total costs“, in real-world German financial reports. It combines a financial named entity and relation extraction module with a BERT-based filtering and text pair classification component to extract KPIs from...}}
