This study proposes a sentence-level framework to identify, analyze, and trace the evolution of motivations for mentioning algorithms in academic papers, using natural language processing as a case study. The researchers classify these motivations using pretrained models and data augmentation, revealing that deep learning models outperform traditional machine learning approaches.
- Deep learning models trained with augmented data achieve higher performance than traditional machine learning models in motivation classification.
- In NLP papers, over half of algorithm-related sentences express direct use, while improvement is the least frequent motivation.
- Grammar-based algorithms are more often mentioned for description, whereas machine learning algorithms are more frequently cited for use.
- Use motivations have gradually replaced description motivations over time, and the number of motivation types associated with individual algorithms has declined significantly.
The findings provide a basis for future research on algorithm relationship identification and algorithm impact evaluation by revealing how authors mention algorithm entities in academic writing.