Prof. Dr.-Ing.

Bauckhage, Christian

Filters

1 1
2
Reset filters

Settings
Now showing 1 - 2 of 2
  • Publication
    Improving Zero-Shot Text Matching for Financial Auditing with Large Language Models
    ( 2023-08-22)
    Hillebrand, Lars Patrick
    ;
    Berger, Armin
    ;
    Deußer, Tobias orcid-logo
    ;
    Dilmaghani, Tim
    ;
    Khaled, Mohamed
    ;
    Kliem, B.
    ;
    Loitz, Rüdiger
    ;
    Pielka, Maren
    ;
    Leonhard, David
    ;
    Bauckhage, Christian
    ;
    Sifa, Rafet
    Auditing financial documents is a very tedious and time-consuming process. As of today, it can already be simplified by employing AI-based solutions to recommend relevant text passages from a report for each legal requirement of rigorous accounting standards. However, these methods need to be fine-tuned regularly, and they require abundant annotated data, which is often lacking in industrial environments. Hence, we present ZeroShotALI, a novel recommender system that leverages a state-of-the-art large language model (LLM) in conjunction with a domain-specifically optimized transformer-based text-matching solution. We find that a two-step approach of first retrieving a number of best matching document sections per legal requirement with a custom BERT-based model and second filtering these selections using an LLM yields significant performance improvements over existing approaches.
  • Publication
    KPI-BERT: A Joint Named Entity Recognition and Relation Extraction Model for Financial Reports
    ( 2022)
    Hillebrand, Lars Patrick
    ;
    Deußer, Tobias orcid-logo
    ;
    Dilmaghani, Tim
    ;
    Kliem, Bernd
    ;
    Loitz, Rüdiger
    ;
    Bauckhage, Christian
    ;
    Sifa, Rafet
    We present KPI-BERT, a system which employs novel methods of named entity recognition (NER) and relation extraction (RE) to extract and link key performance indicators (KPIs), e.g. "revenue"or "interest expenses", of companies from real-world German financial documents. Specifically, we introduce an end-to-end trainable architecture that is based on Bidirectional Encoder Representations from Transformers (BERT) combining a recurrent neural network (RNN) with conditional label masking to sequentially tag entities before it classifies their relations. Our model also introduces a learnable RNN-based pooling mechanism and incorporates domain expert knowledge by explicitly filtering impossible relations. We achieve a substantially higher prediction performance on a new practical dataset of German financial reports, outperforming several strong baselines including a competing state-of-the-art span-based entity tagging approach.