Large-Scale Multi-Label Text Classification on EU Legislation

Ilias Chalkidis, Manos Fergadiotis, Prodromos Malakasiotis, Ion Androutsopoulos

2019-06-05ACL 2019 7Text Classification Multi Label Text Classification text-classification General Classification Classification Zero-Shot Learning Multi-Label Text Classification

Paper PDF Code(official)

Abstract

We consider Large-Scale Multi-Label Text Classification (LMTC) in the legal domain. We release a new dataset of 57k legislative documents from EURLEX, annotated with ~4.3k EUROVOC labels, which is suitable for LMTC, few- and zero-shot learning. Experimenting with several neural classifiers, we show that BIGRUs with label-wise attention perform better than other current state of the art methods. Domain-specific WORD2VEC and context-sensitive ELMO embeddings further improve performance. We also find that considering only particular zones of the documents is sufficient. This allows us to bypass BERT's maximum text length limit and fine-tune BERT, obtaining the best results in all but zero-shot learning cases.

Results

Task	Dataset	Metric	Value	Model
Multi-Label Text Classification	EUR-Lex	Micro F1	73.2	bert-base
Multi-Label Text Classification	EUR-Lex	P@5	68.7	bert-base
Multi-Label Text Classification	EUR-Lex	RP@5	79.6	bert-base
Multi-Label Text Classification	EUR-Lex	nDCG@5	82.3	bert-base
Text Classification	EUR-Lex	Micro F1	73.2	bert-base
Text Classification	EUR-Lex	P@5	68.7	bert-base
Text Classification	EUR-Lex	RP@5	79.6	bert-base
Text Classification	EUR-Lex	nDCG@5	82.3	bert-base
Classification	EUR-Lex	Micro F1	73.2	bert-base
Classification	EUR-Lex	P@5	68.7	bert-base
Classification	EUR-Lex	RP@5	79.6	bert-base
Classification	EUR-Lex	nDCG@5	82.3	bert-base

Related Papers

Making Language Model a Hierarchical Classifier and Generator2025-07-17 Adversarial attacks to image classification systems using evolutionary algorithms2025-07-17 GLAD: Generalizable Tuning for Vision-Language Models2025-07-17 Efficient Calisthenics Skills Classification through Foreground Instance Selection and Depth Estimation2025-07-16 Safeguarding Federated Learning-based Road Condition Classification2025-07-16 DEARLi: Decoupled Enhancement of Recognition and Localization for Semi-supervised Panoptic Segmentation2025-07-14 AI-Enhanced Pediatric Pneumonia Detection: A CNN-Based Approach Using Data Augmentation and Generative Adversarial Networks (GANs)2025-07-13 GNN-CNN: An Efficient Hybrid Model of Convolutional and Graph Neural Networks for Text Representation2025-07-10