LTP: A New Active Learning Strategy for CRF-Based Named Entity Recognition

TL;DR

This paper introduces LTP, a simple and effective active learning strategy for CRF-based named entity recognition that reduces annotation effort without modifying the model.

Contribution

The paper proposes LTP, a novel uncertainty-based active learning method that avoids sequence length bias and model invasion, improving annotation efficiency in NER tasks.

Findings

LTP slightly outperforms traditional strategies in accuracy and F1-score.

LTP reduces annotation tokens needed for training.

LTP does not require model modification or favor long sequences.

Abstract

In recent years, deep learning has achieved great success in many natural language processing tasks including named entity recognition. The shortcoming is that a large amount of manually-annotated data is usually required. Previous studies have demonstrated that active learning could elaborately reduce the cost of data annotation, but there is still plenty of room for improvement. In real applications we found existing uncertainty-based active learning strategies have two shortcomings. Firstly, these strategies prefer to choose long sequence explicitly or implicitly, which increase the annotation burden of annotators. Secondly, some strategies need to invade the model and modify to generate some additional information for sample selection, which will increase the workload of the developer and increase the training/prediction time of the model. In this paper, we first examine traditional active learning strategies in a specific case of BiLstm-CRF that has widely used in named entity recognition on several typical datasets. Then we propose an uncertainty-based active learning strategy called Lowest Token Probability (LTP) which combines the input and output of CRF to select informative instance. LTP is simple and powerful strategy that does not favor long sequences and does not need to invade the model. We test LTP on multiple datasets, and the experiments show that LTP performs slightly better than traditional strategies with obviously less annotation tokens on both sentence-level accuracy and entity-level F1-score. Related code have been release on https://github.com/HIT-ICES/AL-NER