一个文字-语音多式联络中文命名实体识别模型,用于作物疾病和害虫
Ruilin Liu1, Xuchao Guo2, HongMei Zhu2
1School of Information Management, Nanjing Agricultural University, Nanjing, Jiangsu, China.
Scientific reports
|February 13, 2025
概括
本研究引入了一种新的多式联运命名实体识别 (NER-CDP) 模型,使用音频线索来改善作物疾病和害虫识别. 通过整合文本和声音,CDP-MCNER模型提高了准确性,克服了纯文本方法的局限性.
科学领域:
- 农业信息学 农业信息学
- 自然语言处理自然语言处理.
- 机器学习 机器学习
背景情况:
- 对作物疾病和害虫的命名实体识别 (NER-CDP) 对于农业数据提取至关重要.
- 现有的NER-CDP方法与单词细分错误作斗争,限制了性能.
- 在NER-CDP中,多式联络方法尚未得到充分探索.
研究的目的:
- 为农作物疾病和害虫提出一个多式联名实体识别模型 (CDP-MCNER).
- 为了解决NER-CDP中词段化错误造成的性能下降.
- 为了利用音频模式,在这个领域改进中文单词细分.
主要方法:
- 开发了一种多式联通命名实体识别模型 (CDP-MCNER),利用跨式联通注意力.
- 引入了音频模式,使用中文单词细分辅助的暂停.
- 采用数据增强技术 (文字编码器干扰,声频增强).
- 使用的蒙面连接主义时间分类 (CTC) 语义对齐的损失.
主要成果:
- 在定制数据集上实现了最佳精度 (91.32%),回忆 (93.05%) 和F1得分 (92.18%).
- 在公开数据集上获得了81.05% (CNERTA) 和79.23% (Ai-SHELL) 的最佳F1分数.
- 与纯文本,词汇增强和其他多式模式相比,表现出优异的性能.
结论:
- CDP-MCNER模型有效地整合了NER-CDP的文字和声学信息.
- 提出的方法显著提高了对单词细分错误的准确性和稳定性.
- 该模型在作物疾病和害虫识别方面显示出强大的有效性和概括能力.
相关概念视频
02:16Light Acquisition
8.4K
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
8.4K
02:35Key Elements for Plant Nutrition
18.6K
Like all living organisms, plants require organic and inorganic nutrients to survive, reproduce, grow and maintain homeostasis. To identify nutrients that are essential for plant functioning, researchers have leveraged a technique called hydroponics. In hydroponic culture systems, plants are grown—without soil—in water-based solutions containing nutrients. At least 17 nutrients have been identified as essential elements required by plants. Plants acquire these elements from the...
18.6K