TR-EduVSum: New Dataset and Framework for Turkish Educational Video Summarization
Researchers introduce TR-EduVSum, a dataset of 82 Turkish educational videos with 3281 human summaries, and AutoMUP, a consensus-based summarization framework. This work advances automatic summarization for educational content in Turkish.
Researchers have developed TR-EduVSum, a new dataset focused on Turkish educational videos. The dataset includes 82 videos on "Data Structures and Algorithms" along with 3281 independent human summaries. This resource aims to improve automated summarization for educational content in Turkish.
The study introduces the AutoMUP (Automatic Meaning Unit Pyramid) method, inspired by pyramid-based evaluation approaches. AutoMUP generates gold-standard summaries automatically by analyzing multiple human summaries, ensuring reproducibility. This framework could enhance the accessibility and efficiency of educational materials in Turkish.
The implications of this research extend beyond language-specific applications. By providing a robust dataset and evaluation framework, the study sets a precedent for other languages and domains. Future work may explore the adaptability of AutoMUP to different educational contexts and languages, potentially revolutionizing how educational content is summarized and distributed.