Weakly Supervised Concept Map Generation through Task-Guided Graph Translation.

Abstract

Recent years have witnessed the rapid development of concept map generation techniques due to their advantages in providing well-structured summarization of knowledge from free texts. Traditional unsupervised methods do not generate task-oriented concept maps, whereas deep generative models require large amounts of training data. In this work, we present <i>GT-D2G</i> (Graph Translation-based Document To Graph), an automatic concept map generation framework that leverages generalized NLP pipelines to derive semantic-rich initial graphs, and translates them into more concise structures under the weak supervision of downstream task labels. The concept maps generated by <i>GT-D2G</i> can provide interpretable summarization of structured knowledge for the input texts, which are demonstrated through human evaluation and case studies on three real-world corpora. Further experiments on the downstream task of document classification show that <i>GT-D2G</i> beats other concept map generation methods. Moreover, we specifically validate the labeling efficiency of <i>GT-D2G</i> in the label-efficient learning setting and the flexibility of generated graph sizes in controlled hyper-parameter studies.