深度学习用于使用NLP进行文本总结,用于自动化新闻摘要
K M Rani Krishna1, K Somasundaram2, P Arulmozhivarman3
1Department of Mathematics, Amrita School of Physical Science, Amrita Vishwa Vidyapeetham, Coimbatore, India.
Scientific reports
|October 17, 2025
概括
这项研究比较了对文本总结的深度学习模型,发现基于Rouge分数的PEGASUS提供了最佳性能. 这项研究解决了自然语言处理中的挑战,以实现有效的文本凝聚.
科学领域:
- 自然语言处理 (NLP) 是一种自然语言处理.
- 人工智能 (AI) 是一种人工智能.
- 机器学习 机器学习
背景情况:
- 文本总结对于缩小信息,同时保持意义至关重要.
- 深度学习模型面临诸如语义理解和处理长文件等挑战.
- 现有的深度学习方法提供了诸如节省时间和个性化等好处.
研究的目的:
- 提出和评估用于文本总结的深度学习模型.
- 为了比较T5-基,T5-大,BART CNN和PEGASUS模型的性能.
- 根据评估指标,确定产生最高质量的摘要的模型.
主要方法:
- 数据预处理和探索性数据分析 (EDA).
- 实施和培训T5-基础,T5-大,BART CNN和PEGASUS模型.
- 评估使用红色和蓝色分数来评估总结质量.
主要成果:
- 该研究量化了每个深度学习模型的性能.
- 红色和蓝色分数在训练后计算,以衡量有效性.
- 对比分析的重点是确定 Rouge 评分最高的模型.
结论:
- 深度学习模型在推进文本总结技术方面表现有前途.
- 模型的性能各不相同,而PEGASUS被认为是一个强的表现者.
- 进一步的研究正在进行中,以解决基于深度学习的总结方面的挑战.
相关概念视频
Improving Translational Accuracy
3.5K
3.5K
lncRNA - Long Non-coding RNAs
3.5K
3.5K
Trimmed Mean
3.3K
While measuring the mean of a data set, care needs to be taken when associating the mean to its central tendency. The same goes for the arithmetic mean, the geometric mean, or the harmonic mean. This is because the presence of a single outlier data value can significantly affect the mean. That is, the mean is sensitive to fluctuations in the data set.
Although certain measures of central tendency are not sensitive to outliers, there are alternative versions of the mean that get around the...
Although certain measures of central tendency are not sensitive to outliers, there are alternative versions of the mean that get around the...
3.3K
Autonomic Nervous System: Overview
7.3K
The human nervous system is divided into two main parts: the central nervous system (CNS) and the peripheral nervous system (PNS). The CNS is composed of the brain and spinal cord, while the PNS contains nerve cells, clusters of nerve cells, and the sensory receptors that are outside the CNS. The PNS has two types of nerve cells: sensory (afferent) and motor (efferent). Sensory cells send signals to the CNS from receptors, and motor cells carry signals from the CNS to organs, muscles, and...
7.3K
Nonsense-mediated mRNA Decay
3.3K
3.3K
Nucleic Acids
8.8K
8.8K
