排序方式: 共有164条查询结果,搜索用时 15 毫秒
1.
基于深度学习的中文专利自动分类方法研究 总被引:2,自引:0,他引:2
[目的/意义] 面向当前国内专利审查和专利情报分析工作中对于海量专利分类的客观需求,设计了7种基于深度学习的专利自动分类方法,对比各种方法的分类效果,从而助力专利分类效率和效果的提升。[方法/过程] 针对传统机器学习方法存在的缺陷,基于Word2Vec、CNN、RNN、Attention机制等深度学习技术,考虑专利文本语序特征、上下文特征以及分类关键特征,设计Word2Vec+TextCNN、Word2Vec+GRU、Word2Vec+BiGRU、Word2Vec+BiGRU+TextCNN等7种深度学习模型,以中国专利为例,选取IPC主分类号的"部"作为分类依据,对比这7种模型与3种传统分类模型在中文专利分类任务中的效果。[结果/结论] 实证研究效果显示,采用考虑语序特征、上下文特征及强化关键特征的深度学习方法进行中文专利分类具有更优的分类效果。 相似文献
2.
Lyu Sun-jian Yuan Xue-mei Zhang Hai-qi Shi Wei-da Hang Xiao-ying Liu Li Wu Ying-lei 《Journal of Zhejiang University. Science. B》2019,20(9):728-739
Journal of Zhejiang University-SCIENCE B - As one of the most important aquatic fish, Micropterus salmoides suffers lethal and epidemic disease caused by rhabdovirus at the juvenile stage. In this... 相似文献
3.
Huixiang Wang Fei Lyu Kapil Sugand Sheungting Wong Yanping Lin Qiugen Wang 《Anatomical sciences education》2019,12(6):655-663
Acetabular fractures are a real challenge for junior doctors as well as experienced orthopedic surgeons. Correct fracture classification is crucial for appreciating the fracture type, surgical planning, and predicting prognosis. Although three-dimensional (3D) tutorial is believed to improve the understanding of the complex anatomy structure, there have been few applications and randomized controlled trials to confirm it in orthopedics. This study aims to develop a 3D interactive software system for teaching acetabular fracture classification and evaluate its efficacy. Participants were randomly but evenly allocated into either the experimental group (who learned the acetabular fracture classification using a 3D software) or the control group (who used a traditional two-dimensional [2D] tutorial). Both groups were then tasked to classify 10 acetabular fractures and complete a five-point Likert scale on their satisfaction of each learning modality. To calculate significance (P < 0.05), independent t-test was used for normally distributed data whereas Mann-Whitney U test for non-normally distributed data. The experimental group significantly outperformed the control group (t (28) = 2.526, P = 0.017) with identifying correct acetabular fracture classification. Moreover, Likert scale score in the experimental group was also significantly higher than in the control group (Z = 2.477, P = 0.013). This 3D classification software has objectively and subjectively showed an advantage over the traditional 2D tutorial, resulting in an improved classification accuracy and higher Likert scale score. The 3D software has the potential to improve both clinical knowledge as well as identifying correct patient management in orthopedics. 相似文献
4.
5.
在界定内容策展概念的基础上,本文对国内外内容策展平台进行总结分析,并以国内策展平台——花瓣网为例,借助MetaStudio和DataScraper工具,对策展活动中策展主题、数据来源以及用户相关行为进行探讨。研究表明,策展数据的来源广泛,策展结果主题化鲜明且融入了个人知识的增值信息,内容策展的实例化已基本实现,但大众与小众主题的被关注度存在显著差异,同时平台中信息交流的表达不同于传统的社交平台,其主要以策展结果的方式展现。 相似文献
6.
在分析心理素质的内涵及作用的基础上,阐述了学员心理素质训练的内容,提出了加强学员心理素质训练的对策,进而为促进军事体育课程的教学效果,培养适应未来信息化战争的高素质军事人才打下基础。 相似文献
7.
史学原理告诉我们:完整的"历史记录",是由事件、事件过程及其事件背景构成的.虽然我们能清楚地看到,人们利用档案记录了无数的历史事件,但我们是否问过这样的问题:在利用档案记录历史的过程中,它是如何完成事件背景记录的? 相似文献
8.
9.
Fengjiao Lyu Manqi Xu Yunfeng Cheng Jinghang Xie Jianghong Rao Sindy K. Y. Tang 《Biomicrofluidics》2015,9(4)
This paper describes a method for the quantitative detection of cells expressing BlaC, a β-lactamase naturally expressed by Mycobacterium tuberculosis, intended for the diagnosis of tuberculosis. The method is based on the compartmentalization of bacteria in picoliter droplets at limiting dilutions such that each drop contains one or no cells. The co-encapsulation of a fluorogenic substrate probe for BlaC allows the quantification of bacteria by enumerating the number of fluorescent drops. Quantification of 10 colony forming units per milliliter is demonstrated. Furthermore, the encapsulation of single cell in drops maintains the specificity of the detection scheme even when the concentration of bacteria that do not express BlaC exceeds that expressing BlaC by one million-fold. 相似文献
10.