Combining Language and Graph Models for Semi-structured Information Extraction on the Web
This addresses the challenge of noisy and data-dependent information extraction on the web, offering a more efficient and generalizable solution for mining web knowledge.
The paper tackled the problem of extracting targeted relations from semi-structured web pages using only a short description, without domain-specific training data, and achieved a 34.8% improvement in F1 scores in zero-shot settings.
Relation extraction is an efficient way of mining the extraordinary wealth of human knowledge on the Web. Existing methods rely on domain-specific training data or produce noisy outputs. We focus here on extracting targeted relations from semi-structured web pages given only a short description of the relation. We present GraphScholarBERT, an open-domain information extraction method based on a joint graph and language model structure. GraphScholarBERT can generalize to previously unseen domains without additional data or training and produces only clean extraction results matched to the search keyword. Experiments show that GraphScholarBERT can improve extraction F1 scores by as much as 34.8\% compared to previous work in a zero-shot domain and zero-shot website setting.