Challenges and Possibilities for the Household Medicine Lease （HML） System Viewed in Light of CRM： HML-Based Everyday Self-Medication in Ubiquitous Network Society

Household medicine lease （HML） industry originated way back in the Edo period （17C-19C）, when it was promoted by the local fiefdom government to revitalize the economy. Accumulations of wealth, acquired thereafter from everywhere outside the region, have culminated in the formation of the present-day industrial cluster in Toyama, the largest in the whole area facing the Sea of Japan. Today an adaptation of the quasi-CRM （customer relationship management） business model of the HML system has proved to be a success in Mongolia. This fact seems to offer the authors some clues for dealing with those problems that healthcare and medical services in Japan and elsewhere are riddled with. In this paper, focusing on the common critical success factors （CSFs） behind the success of the authors＇ prototype HML system and its recent successful application in Mongolia, the authors will analyze these factors from the perspective of CRM. The authors will then clarify the following： （1） the usefulness of the business model for ensuring primary healthcare for people in developing countries; （2） the usefulness in our ubiquitous network society of applying ICT to the HML system; （3） possible contributions that the use of the system can make toward improving the quality of our everyday healthcare and medical services in our prominently aging society; and the authors will also suggest （4） the importance of elevating ＂individual self-medication＂ to ＂community-based self-medication＂.

Shaping the future of application for information and communication technology in national economic analysis

Characterized by networking,digitization,and intellectualization,the new generation of Information and Communication Technologies(ICTs)has promoted integration of various national economic industries,with a more significant leading role in the socio-economic development.In order to describe the influence of ICT industries on economic growth,this paper has comprehensively analyzed the contribution mechanism of ICTs to current economy.Crucial indicators,such as Total Telecommunications Business(TTB),Data Traffic(DT),and Computing Power Development Index(CPDI),have been innovatively selected to sufficiently describe the data circulation scale and the processing capacity in the era of digital economy.Based on relevant statistics of 31 provinces(municipalities or autonomous regions)in China from 2011 to 2020,quantitative results have been put forward to interpret the impact of ICT industries on the national economy development by using correlation coefficient analysis and data regression analysis methods.The measurement results show that TTB,DT,and CPDI are positively correlated with the national economic development,indicating that each 1%increasement of TTB,DT,and CPDI increases Gross Domestic Product(GDP)by 0.08%,0.09%,and 0.6%,respectively.

基于BR和GBDT的电力信息通信客服系统多标签文本分类

现有电力信息通信(ICT)客服系统主要依靠客服坐席员经验,根据电力ICT系统用户报修信息进行故障类型分类判别,存在在线处理及时性较差、准确性不足的问题。针对上述问题,提出了一种基于集成学习的电力ICT客服系统文本数据的多标签文本分类方法,实现对电力ICT系统的复杂故障类型进行自动化、高准确率分类识别。首先,针对电力ICT系统故障类型识别准确率偏低且低效的问题,提出了基于二元相关性(BR)和梯度提升决策树(GBDT)集成学习的多标签分类方法,将BR和GBDT有机结合实现自动化、高准确率的故障多标签分类。其次,针对电力ICT客服文本数据的多标签分类训练集难以获取的问题,提出一种面向电力ICT客服文本数据的多标签训练集自动化构建方法,实现了高效的电力ICT客服文本多标签分类。实验表明,BR-GBDT方法可以高效处理电力ICT系统复杂故障类型的多标签分类任务,分类性能也优于BR+逻辑回归(LR)和多标签k最近邻(ML-kNN)等典型的集成学习多标签分类方法。

电力地理信息系统及其发展

综述了地理信息系统技术在电力系统中的三种应用形式,详细论述了电力地理信息系统在发电、输变电、配电和客户服务系统等领域的应用情况,给出了电力地理信息系统的体系结构和平台选择策略,展望了电力地理信息系统今后的应用发展方向。

基于层次化类别嵌入的电力ICT系统故障分类

为解决电力信息通信客服系统在故障研判时存在故障分类准确率低甚至误分的问题,提出基于层次化类别嵌入的文本分类方法,进行电力信息通信系统故障识别。首先,基于电力信息通信系统故障的用户保修工单文本数据构建电力信息通信系统层次化电力故障标签;其次,提出了基于层次化深层金字塔卷积神经网络和基于层次化中断循环神经网络2种层次化文本分类方法,采用层次化类别嵌入方法逐层进行故障类型分类。实验结果表明,基于层次化深层金字塔卷积神经网络的方法效果最优,可以提供高效、准确的故障识别服务。