Information Extraction in Domain and Generic Documents: Findings from Heuristic-based and Data-driven Approaches

TL;DR

This study compares heuristic-based and data-driven information extraction methods across document genres and lengths, revealing that no single approach dominates and that document features significantly influence extraction accuracy.

Contribution

It investigates the impact of document genre and length on IE tasks, providing insights into the effectiveness of different methods in diverse document settings.

Findings

Data-driven methods outperform symbolic approaches in NER, especially in short texts.

Heuristic and syntax-based models outperform pure data-driven approaches in SRL.

Different semantic roles have varying extraction accuracies depending on the method.

Abstract

Information extraction (IE) plays very important role in natural language processing (NLP) and is fundamental to many NLP applications that used to extract structured information from unstructured text data. Heuristic-based searching and data-driven learning are two main stream implementation approaches. However, no much attention has been paid to document genre and length influence on IE tasks. To fill the gap, in this study, we investigated the accuracy and generalization abilities of heuristic-based searching and data-driven to perform two IE tasks: named entity recognition (NER) and semantic role labeling (SRL) on domain-specific and generic documents with different length. We posited two hypotheses: first, short documents may yield better accuracy results compared to long documents; second, generic documents may exhibit superior extraction outcomes relative to domain-dependent documents due to training document genre limitations. Our findings reveals that no single method demonstrated overwhelming performance in both tasks. For named entity extraction, data-driven approaches outperformed symbolic methods in terms of accuracy, particularly in short texts. In the case of semantic roles extraction, we observed that heuristic-based searching method and data-driven based model with syntax representation surpassed the performance of pure data-driven approach which only consider semantic information. Additionally, we discovered that different semantic roles exhibited varying accuracy levels with the same method. This study offers valuable insights for downstream text mining tasks, such as NER and SRL, when addressing various document features and genres.