农业科技服务云平台构建与农业时空推荐算法研究

发布时间：2018-04-26 07:26

  本文选题：农业科技 + 农民培训　； 参考：《吉林农业大学》2017年硕士论文

【摘要】：科技是第一生产力,农民科技素质对培养新型职业农民,发展现代农业具有重要意义。当前的农业科技服务主要包括农民培训、农技推广和成果推介,随着信息技术的快速发展,以上三种服务模式在各种互联网平台上也得到了充分利用。但存在以下问题:第一,线上农业科技资源分散,缺少整合与共享;第二,线上资源的制作没有考虑到农民时间的碎片化特点,导致资源利用率不高,发挥作用不够。第三,现有的农业科技资源管理平台没有考虑到农业生产的时空性特点,没有实现农业科技资源的精准化和个性化推荐。针对以上问题,本文开展了以下研究。第一,针对现有线上农业科技服务模式的不足,分析与设计了一种基于大数据的互联网+农业科技服务云平台。该平台面向三种用户:农民、农业专家和系统管理员,实现三种应用模式:电脑端B/S模式、移动端Web模式和微信公众号模式。整个平台由前台和后台两部分组成。前台主要功能包括:公开课、科教片、专家坐堂、农村书屋等;后台主要功能包含:数据统计与分析、资源共享与重用、信息搜索与推荐、视频压缩与优化等。第二,针对农民时间的碎片化特点,提出一种基于微课的农业科技资源制作方法和翻转式培训方法。该方法有2个基本要点:(1)制作以问题为导向的微课在互联网平台上发布;(2)农民在线上学习农业专家参与录制的微课程,农业专家在线下组织农民进行讨论并对农民的提问进行答疑。第三,针对农业生产的时空性特点,提出一种农业资源时空推荐模型。该模型以农民所在地区和访问平台时的节气为依据,为农民提供精准化、个性化资源推荐服务。该模型符合农业生产实际,是对已有推荐方法的一个突破。第四,在以上三项研究的基础上,开发出基于大数据的农业科技服务云平台-吉农在线。平台采用spark大数据架构和MongoDB数据库,实现了数据的快速性访问和资源的智能化管理。
[Abstract]:Science and technology is the first productive force, and the scientific and technological quality of farmers is of great significance to cultivate new professional farmers and develop modern agriculture. The current agricultural science and technology service mainly includes farmer training, agricultural technology extension and achievement promotion. With the rapid development of information technology, the above three service models have been fully utilized on various Internet platforms. But there are the following problems: first, the online agricultural science and technology resources scattered, lack of integration and sharing; second, the production of online resources did not take into account the fragmentation of farmers' time characteristics, resulting in a low utilization of resources, not enough to play a role. Third, the existing agricultural science and technology resources management platform does not take into account the space-time characteristics of agricultural production, and does not realize the precision and personalized recommendation of agricultural science and technology resources. In view of the above problems, this paper carried out the following research. Firstly, a cloud platform of Internet agricultural science and technology service based on big data is analyzed and designed in view of the deficiency of the existing online agricultural science and technology service model. The platform is designed for three kinds of users: farmers, agricultural experts and system administrators. It implements three application modes: computer side B / S mode, mobile side Web mode and WeChat public number mode. The whole platform consists of foreground and background. The main functions of the front desk include: open class, science and education film, expert sitting room, rural book house, etc. The backstage main functions include: data statistics and analysis, resource sharing and reuse, information search and recommendation, video compression and optimization, etc. Secondly, according to the fragmentation of farmer's time, a method of making agricultural science and technology resources based on microcourse and training method of flipping is put forward. This method has two basic points: 1) making problem-oriented microcourses and publishing them on Internet platforms.) Farmers learn microcourses recorded by agricultural experts online. Agricultural experts left the line to organize farmers to discuss and answer questions from farmers. Thirdly, according to the spatiotemporal characteristics of agricultural production, a spatio-temporal recommendation model of agricultural resources is proposed. The model is based on the solar terms of the area where the farmers are located and when visiting the platform, and provides the accurate and individualized resource recommendation service for the farmers. The model accords with the reality of agricultural production and is a breakthrough to the existing recommended methods. Fourth, on the basis of the above three studies, a cloud platform of agricultural science and technology based on big data-Jinong online was developed. The platform adopts spark big data structure and MongoDB database to realize fast data access and intelligent resource management.
【学位授予单位】：吉林农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.3;TP393.09

【参考文献】

中国期刊全文数据库 前10条

1 梅辉;韦琴;张军林;;Camtasia Studio 8.6微课制作软件的应用——以分子生物学微课制作为例[J];黑龙江畜牧兽医;2017年04期

2 陈程程;张宵;宫健丽;于合龙;;农业微课制作方法研究[J];中国农机化学报;2017年02期

3 张应辉;司彩霞;;基于用户偏好和项目特征的协同过滤推荐算法[J];计算机技术与发展;2017年01期

4 薛红霞;;微课的制作与应用:基于29节微课的评析与反思[J];中小学管理;2016年12期

5 周晓玲;刘燕梅;;大学英语翻转课堂教学模式的实验研究[J];广西师范大学学报(哲学社会科学版);2016年05期

6 陈虹君;吴雪琴;;基于Hadoop平台的Spark快数据推荐算法分析与应用[J];现代电子技术;2016年10期

7 黎玲萍;毛克彪;付秀丽;马莹;王芳;刘R，

本文编号：1805059