Data Analytics for Future Cities

“物联网”是一个短语，描述了支持互联网应用程序传播的一系列技术，系统和方法。最终，物联网将涉及物理对象无缝集成到信息网络中，以实现社会和经济效益。物联网的核心是数据-人类，技术和自然交互的数字记录。数据流是大规模的，多样的和快速变化的。除了收集、存储和共享这些数据这一重要但概念上简单的任务外，还迫切需要开发强大而高效的计算算法，以提取见解并做出有用的预测。这项提议将增加充分利用数据洪流所需的“聪明才智”。技术发展迅速。近年来，智能手机、平板电脑、笔记本电脑和台式电脑的外观、感觉和功能发生了巨大变化，出现了一系列新技术，如可穿戴设备、智能眼镜和植入式传感器。在它们被取代之前，其中许多将为技术革命添砖加瓦。然而，分析和利用物联网产生的大量数据的挑战是普遍的，支撑这些新技术的数学概念和由此产生的算法是基础性的，具有非常长期的价值。我的建议将开发新的分析概念，导致计算算法。研究结果将直接用于未来城市研究的应用--改善城市生活的社会、环境和经济方面。斯特拉斯克莱德大学是城市观测站的所在地，该观测站收集了来自格拉斯哥市的大量数据，包括空气质量传感器，交通信息，能源使用和人类行为许多方面的测量，例如在线活动，社交媒体互动，CCTV数据和零售客流量。事实上，格拉斯哥获得了2400万英镑的政府资金，成为英国这种数字驱动增强的试点城市。因此，该研究项目侧重于新的概念和算法，可以帮助我们理解非常大的，快速移动的数据流，描述组成物联网的组件之间的交互：人，设备和传感器。正如Google开始使用一种聪明的数学算法PageRank，能够为万维网带来秩序一样，我们的目标是开发能够总结这些大量数据的新算法。这项工作将与未来城市竞技场的利益相关者一起进行：社会科学研究人员;为广告，金融，娱乐，出版客户提供数据分析解决方案的中小企业;提供大规模IT解决方案的高科技行业外部合作伙伴;为社区服务的地方和国家政府雇员。通过一系列知识交流和外联活动，这些利益攸关方将有机会对成果进行批判性评价和反馈，并迅速部署新的想法。该研究将涉及的一些示例性应用包括：描述不同用户群的社交媒体特征，例如司机/骑自行车的人/行人，以预测针对每个群体的消息的最佳方式，根据他们的工作/休闲/购物社区成员资格组合对城市用户群体进行分层，以最大限度地利用城市中的能源/空间资源，预测即将到来的公共事件中的人群水平和人群行为，监视公众对正在进行的运动的感知，例如骑自行车上班计划，在真实的时间内监视公众对与计划的中断（例如政治游行）或不可预测的事件（例如交通事故）有关的响应并作出反应。

The "Internet of Things" is a phrase describing the set of technologies, systems and methodologies that underpin the spread of internet-enabled applications. Ultimately, the Internet of Things will involve physical objects seamlessly integrating into the information network for social and economic benefit.At the heart of the Internet of Things is data---digital records of human, technological and natural interactions. The data streams are large-scale, varied and rapidly changing. In addition to the important, but conceptually simple, tasks of gathering, storing and sharing this data, there is a pressing need to develop powerful and efficient computational algorithms that can extract insights and make useful predictions. This proposal will add to the "cleverness" that is needed to exploit fully the data deluge. Technology evolves rapidly. The look, feel and functionality of smartphones, tablets, notebooks and desktop PCs have changed dramatically in recent years, and there is a range of new technologies such as wearable devices, smart glasses and implantable sensors. Many of these will add to the technological revolution, before themselves being superseded. However, the challenge of analysing and exploiting the vast realms of data produced by the Internet of Things is universal, and the mathematical concepts and resulting algorithms that underpin these new technologies are fundamental and have very long-term value.My proposal will develop new analytical concepts that lead to computational algorithms. The results will be aimed directly at the application of Future Cities research---improving the social, environmental and economic aspects of city living. The University of Strathclyde is home to a City Observatory that collects a huge amount of data from the city of Glasgow, including air quality sensors, traffic information, energy usage and measurements of many aspects of human behaviour, such as on-line activity, social media interactions, CCTV data and retail footfall counts. Indeed, Glasgow received £24M of government funding to become the UK's pilot city for this type of digitally-driven enhancement. The research project therefore focusses on new concepts and algorithms that can help us to understand the very large, fast-moving streams of data describing the interactions between the components that make up the Internet of Things: people, devices and sensors. Just as Google began with a clever mathematical algorithm, PageRank, that was able to bring order to the world wide web, we aim to develop new algorithms that can summarize these vast quantities of data. The work will take place alongside stakeholders in the Future Cities arena: fellow-researchers in social science; SMEs who deliver data analytics solutions to clients in advertising, finance, entertainment, publishing; external partners in hi-tech industry who deliver larger-scale IT solutions; local and national government employees who serve the community. Through a range of knowledge exchange and outreach events, these stakeholders will have the opportunity to critically evaluate and feedback on the results and rapidly deploy the new ideas. Some illustrative applications that the research will address are: characterising the social media traits of different user bases, such as drivers/cyclists/pedestrians, to predict the best way to target messages at each group, stratifying the population of city users according to their portfolio of work/leisure/shopping community memberships in order to maximise the usage of energy/space resources in the city,predicting crowd levels and crowd behaviour at forthcoming public events,monitoring the public perception of an ongoing campaign, such as a cycle-to-work-scheme,monitoring and reacting in real time to the public response in relation to a planned disruption, such as a political march, or an unpredictable event, such as a traffic incident.