Improving Keyphrase Extraction with Data Augmentation and Information Filtering

TL;DR

This paper introduces a new dataset and method for keyphrase extraction from video transcripts, utilizing data augmentation to improve performance across informal and diverse text domains.

Contribution

It presents a novel corpus and data augmentation technique specifically designed for keyphrase extraction from informal video transcripts, addressing a gap in existing research.

Findings

Data augmentation improves keyphrase extraction accuracy.

The proposed method outperforms baseline models on the new dataset.

Enrichment with background knowledge enhances model robustness.

Abstract

Keyphrase extraction is one of the essential tasks for document understanding in NLP. While the majority of the prior works are dedicated to the formal setting, e.g., books, news or web-blogs, informal texts such as video transcripts are less explored. To address this limitation, in this work we present a novel corpus and method for keyphrase extraction from the transcripts of the videos streamed on the Behance platform. More specifically, in this work, a novel data augmentation is proposed to enrich the model with the background knowledge about the keyphrase extraction task from other domains. Extensive experiments on the proposed dataset dataset show the effectiveness of the introduced method.