ReactIE: Enhancing Chemical Reaction Extraction with Weak Supervision
This addresses the cost-prohibitive data annotation issue for chemists in tasks like drug design, though it is incremental as it builds on existing weak supervision techniques.
The paper tackles the problem of extracting structured chemical reaction information from scientific literature by proposing ReactIE, a method that uses weak supervision for pre-training, achieving substantial improvements and outperforming all existing baselines.
Structured chemical reaction information plays a vital role for chemists engaged in laboratory work and advanced endeavors such as computer-aided drug design. Despite the importance of extracting structured reactions from scientific literature, data annotation for this purpose is cost-prohibitive due to the significant labor required from domain experts. Consequently, the scarcity of sufficient training data poses an obstacle to the progress of related models in this domain. In this paper, we propose ReactIE, which combines two weakly supervised approaches for pre-training. Our method utilizes frequent patterns within the text as linguistic cues to identify specific characteristics of chemical reactions. Additionally, we adopt synthetic data from patent records as distant supervision to incorporate domain knowledge into the model. Experiments demonstrate that ReactIE achieves substantial improvements and outperforms all existing baselines.