制图模型与算法课程具有理论、技术难、更新较快、与公安业务需求结合紧密、应用性强的特点,存在学生基础弱、实操能力差、内容与公安业务需求脱节、学习资源少、实验环境单一等真实问题。提出以公安业务需求为导向的“理论”“实验”...制图模型与算法课程具有理论、技术难、更新较快、与公安业务需求结合紧密、应用性强的特点,存在学生基础弱、实操能力差、内容与公安业务需求脱节、学习资源少、实验环境单一等真实问题。提出以公安业务需求为导向的“理论”“实验”双循环任务驱动式教学法,通过任务驱动、翻转课堂、第二课堂高阶知识拓展三个环节,实现培养的学生可以直接参与公安大数据工作,熟悉公安大数据建模平台,动手实验能力强,能够解决实际问题。教学途径通过重构教学内容“实战建模案例”、“理论 + 实验”双循环任务驱动教学、“滴灌式”课程思政融入、自主建设大数据建模平台、过程性终结考核评价体系,实现学生警务大数据建模能力的培养。The course of Visualization, Model and Algorithm has the characteristics of theory, technical difficulty, rapid update, close integration with the needs of public security business, and wide application. There are some real problems such as students’ weak foundation, poor practical operation ability, disconnection between content and the needs of public security business, few learning resources, and single experimental environment. We put forward the two-cycle task-driven teaching innovation guided by the actual needs of public security. Through three links of task-driven, flipped classroom and advanced knowledge expansion in the second classroom, the cultivated students can directly participate in the big data work of public security, be familiar with the big data modeling platform, have strong experimental ability and can solve practical problems. Teaching innovation through the reconstruction of teaching content “Practical Modeling Cases”, “Theory + Experiment” double cycle task-driven teaching, “Drip Irrigation” curriculum ideology and politics, independent construction of big data modeling platform, process of final assessment and evaluation system, can train students to have the ability of big data modeling.展开更多
针对用户个性化需求不断增多、系统复杂性呈井喷式增长的客观事实,通过瀑布型(Water fall Model)软件生命周期的案例实践,表明文档驱动的Water fall Model可实现复杂系统的功能,提高系统可靠性、可塑性和质量可信度。同时,瀑布模型也缺...针对用户个性化需求不断增多、系统复杂性呈井喷式增长的客观事实,通过瀑布型(Water fall Model)软件生命周期的案例实践,表明文档驱动的Water fall Model可实现复杂系统的功能,提高系统可靠性、可塑性和质量可信度。同时,瀑布模型也缺乏灵活性,为更好地克服瀑布模型的缺点,提出了CMM+瀑布模型。展开更多
针对聚类中的多视角和可解释的问题,提出多视角生成模型的可解释性聚类算法(interpretable clustering with multi-view generative model,ICMG).ICMG能够产生多个视角的聚类划分,并通过视角的语义信息对聚类结果进行定性和定量地解释....针对聚类中的多视角和可解释的问题,提出多视角生成模型的可解释性聚类算法(interpretable clustering with multi-view generative model,ICMG).ICMG能够产生多个视角的聚类划分,并通过视角的语义信息对聚类结果进行定性和定量地解释.首先,构建一种多视角生成模型(multi-view generative model,MGM),该模型使用贝叶斯程序学习(Bayesian program learning,BPL)和嵌入多视角因素的贝叶斯案例模型(multi-view Bayesian case model,MBCM)生成多个视角.其次,基于视角的匹配度进行聚类得到多种聚类方案.最后使用视角的原型和子空间所附带的语义信息定性和定量地解释聚类结果.实验结果表明:ICMG能够得到多种可解释的聚类结果,相比于传统多视角聚类算法具有较明显的优势.展开更多
文摘制图模型与算法课程具有理论、技术难、更新较快、与公安业务需求结合紧密、应用性强的特点,存在学生基础弱、实操能力差、内容与公安业务需求脱节、学习资源少、实验环境单一等真实问题。提出以公安业务需求为导向的“理论”“实验”双循环任务驱动式教学法,通过任务驱动、翻转课堂、第二课堂高阶知识拓展三个环节,实现培养的学生可以直接参与公安大数据工作,熟悉公安大数据建模平台,动手实验能力强,能够解决实际问题。教学途径通过重构教学内容“实战建模案例”、“理论 + 实验”双循环任务驱动教学、“滴灌式”课程思政融入、自主建设大数据建模平台、过程性终结考核评价体系,实现学生警务大数据建模能力的培养。The course of Visualization, Model and Algorithm has the characteristics of theory, technical difficulty, rapid update, close integration with the needs of public security business, and wide application. There are some real problems such as students’ weak foundation, poor practical operation ability, disconnection between content and the needs of public security business, few learning resources, and single experimental environment. We put forward the two-cycle task-driven teaching innovation guided by the actual needs of public security. Through three links of task-driven, flipped classroom and advanced knowledge expansion in the second classroom, the cultivated students can directly participate in the big data work of public security, be familiar with the big data modeling platform, have strong experimental ability and can solve practical problems. Teaching innovation through the reconstruction of teaching content “Practical Modeling Cases”, “Theory + Experiment” double cycle task-driven teaching, “Drip Irrigation” curriculum ideology and politics, independent construction of big data modeling platform, process of final assessment and evaluation system, can train students to have the ability of big data modeling.
文摘针对用户个性化需求不断增多、系统复杂性呈井喷式增长的客观事实,通过瀑布型(Water fall Model)软件生命周期的案例实践,表明文档驱动的Water fall Model可实现复杂系统的功能,提高系统可靠性、可塑性和质量可信度。同时,瀑布模型也缺乏灵活性,为更好地克服瀑布模型的缺点,提出了CMM+瀑布模型。
文摘针对聚类中的多视角和可解释的问题,提出多视角生成模型的可解释性聚类算法(interpretable clustering with multi-view generative model,ICMG).ICMG能够产生多个视角的聚类划分,并通过视角的语义信息对聚类结果进行定性和定量地解释.首先,构建一种多视角生成模型(multi-view generative model,MGM),该模型使用贝叶斯程序学习(Bayesian program learning,BPL)和嵌入多视角因素的贝叶斯案例模型(multi-view Bayesian case model,MBCM)生成多个视角.其次,基于视角的匹配度进行聚类得到多种聚类方案.最后使用视角的原型和子空间所附带的语义信息定性和定量地解释聚类结果.实验结果表明:ICMG能够得到多种可解释的聚类结果,相比于传统多视角聚类算法具有较明显的优势.
