From statistical methods to deep learning, automatic keyphrase prediction: A survey

Keyphrase prediction aims to generate phrases (keyphrases) that highly summarizes a given document. Recently, researchers have conducted in-depth studies on this task from various perspectives. In paper, we comprehensively summarize representative the perspectives of dominant models, datasets and evaluation metrics. Our work analyzes up 167 previous works, achieving greater coverage than surveys. Particularly, focus deep learning-based keyphrase prediction, which attracts increasing attention in recent years. Afterwards, conduct several groups experiments carefully compare models. To best our knowledge, is first attempt these models using identical commonly-used metric, facilitating analyses their disadvantages advantages. Finally, discuss possible research directions future.