|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/complex-analytics/
课程评论:没有评论
课程名称:高级分析 课程概述:自20世纪中叶以来,信息革命进入了新的发展阶段,云计算、新数据源、算法进步和物联网的崛起等重大趋势交汇,正在引发深刻的变化,将我们带入大数据时代。第一波信息革命围绕个人计算机和互联网建立,创造了大量新数据来源,例如网络博客、互联网搜索历史、大规模电子商务、零售交易、射频识别标签、全球定位系统、传感器网络、社交网络和移动计算等,这些都汇聚成我们现在所称的大数据。然而,没有计算能力来处理这些数据,这些数据将毫无用处。历史上,计算能力一直是稀缺资源,但近年来,全球范围的云计算使得各类组织都能以低成本、按需获得高端计算能力。第三个重要因素是新一组强大的算法方法,特别是机器学习和深度学习的突破,提供了处理这些日益复杂数据集的软件系统;这些算法能够从数据中学习,处理数百万个参数,协调庞大的数字平台,优化物流网络、自动执行金融交易、预测电网的维护需求。 随着新工具的出现,我们正用越来越复杂的算法框架透视庞大的非结构化数据集,探索我们以前无法看见的事物。许多人将其比作建造了一种新型的显微镜或望远镜。而与显微镜揭示生命的微观奥秘和望远镜揭示星空与星系不同,这种工具让我们看到了身边复杂的系统。数据拓宽了我们对周围事物的感知能力,展现了那些因复杂性而难以理解的社会、经济和技术系统的新面貌。其影响深远,正如望远镜改变了我们对宇宙位置的理解,复杂分析正在改变我们对周围世界的认知,影响决策和管理方式。 越来越多的政府、商业部门和机构开始意识到数据正成为最有价值的资产,其分析将成为竞争力的核心。如今,数据成为一种新的通用语言,掌握数据的能力能够决定体育比赛的胜利、电影的成功、选举的结果、智慧城市的建设,甚至可能改变战局。本课程将探讨复杂数据分析的世界:如何分析大数据并将这些数据流转化为洞察、决策和行动。复杂分析集中于如何从金融市场、交通网络或社交网络等复杂系统中提取数据,并将其处理成有意义的模式和可行的洞察。 课程分为四个部分:首先介绍大数据的兴起及其暗数据的扩展,探讨数据化进程及其带来的隐私问题;接着讨论算法的崛起,介绍机器学习系统及不同方法的运用,深入神经网络和深度学习,并评估算法的局限性;第三部分专注于智能系统,探讨机器学习与物联网的融合如何创造出具备适应性与响应能力的自主系统,并讨论相关安全问题;最后一部分探讨人类与技术的关系,分析新型数据驱动的网络组织及合成推理与分析推理的根本区别,以理解数字计算与人类推理之间的动态关系。 这门课程并非技术性课程,不会教授数据建模或构建机器学习系统的细节,而是提供对这一令人激动的重要新领域的全面概述,适用于几乎所有领域、研究人员、工程师、设计师、商业人士及公众。课程旨在全面概述复杂分析,包括这些重大技术发展的背景、相关方法的概念理解及技术底层的概览,并讨论这些技术发展带来的正面和负面影响。
The information revolution that began in the mid-twentieth century is entering a new stage of development; the confluence of major trends in cloud computing, new data sources, advances in algorithms and the rise of the internet of things are assuring in profound changes that take us into the ear of big data. The first wave of the information revolution based around the personal computer and the world wide web has created a torrent of new data sources from web blogs, internet search histories, large-scale e-commerce practices, retail transactions, RFID tags, GPS, sensor networks, social networks and mobile computing have all worked to create what we now call big data. But this mass of data would be no use without computing capacities to process it. Throughout human history, computing power was a scarce resource. However, with the recent advent of global scale cloud computing, high-end computing is now available to organizations of almost all size at low cost and on-demand. The third major element that has fallen into place is a powerful new set of algorithmic approaches. Breakthroughs in machine learning and deep learning in particular now provide the software systems to process these ever more complex data sets; algorithms that learn from data, that can deal with millions of parameters, that can coordinate vast digital platforms, optimizing logistics networks, automating financial trades, predicting maintenance on electrical grids. New Instrument With these new tools, we are now peering into massive unstructured data sets using ever more sophisticated algorithmic frameworks to see what we could never before see. Many compare this to building a new kind of microscope or telescope. But whereas with the microscope we revealed the microscopic mysteries of life and with the telescope the stars and galaxies, this tool lets us see the complex systems all around us. Data is opening up our ability to perceive things around us that were previously invisible; our evolved social, economic and technological systems that have become so complex we can no longer see them are being revealed to us in new ways. The implications of this are huge; just as the telescope changed our understanding of our place in the universe, complex analytics is changing our understanding of the world around us, the systems we form part of and this opens the door to a shift in the nature of how we make decisions and management is conducted. More and more governments, business sectors, and institutions begin to realize data is becoming the most valuable asset and its analysis is becoming core to competitiveness. Today data is becoming a new universal language, mastering it can win sports matches, can make movies a success, can win elections, can build smart cities, can make the right trade at the right time, it may even win wars. This course explores the world of complex data analytics: information systems that are able to analyze big data and transform these restless streams of data into insight, decisions, and action. Complex analytics focuses on how we extract the data from a complex system - such as a financial market, a transport network or a social network - and process that into meaningful patterns and actionable insights. Big Data After starting the course with an overview to the subject we will look at the emergence of big data and the expanding universe of dark data; we talk about the ongoing process of datafication, the quantification of more and more aspects of our lives and the many issues that it brings with respect to privacy. Advanced Algorithms In the second section, we will talk about the rise of algorithms as they are coming to effect ever more spheres of our world. We will introduce you to the workings of machine learning systems and the different approaches used, we go more in-depth on neural networks and deep learning before assessing the limitations of algorithms. Smart Systems The third section is dedicated to smart systems, as the convergence of machine learning with the internet of things is beginning to populate our world with systems that exhibit adaptive and responsive behavior, which are autonomous and can interact with humans in a natural way. Here we look at cyber-physical systems, smart platforms, and autonomous systems before discussing security issues. Data Driven Organizations The final section deals with the relationship between people and technology and the emergence of a new form of analytics and data-driven networked organization. We talk about the fundamental distinction between synthetic and analytical reasoning as a way of understanding the distinction between digital computation and human reasoning and as a means for interpreting the rapidly evolving relationship between the two. This is not a technical course where you will learn the details of data modeling or how to build machine learning systems. What it does provide is an overview of this very exciting and important new area that will be of relevance to almost all domains, researchers, engineers and designers, business and the general public alike. The course aims to be a comprehensive overview to complex analytics, it aims to be inclusive in scope. We try to provide an understanding of the context to these major technological developments; a conceptual understanding of the methods and approaches of big data modeling and analysis; an overview to the underlying technology and address the issues and consequences both positive and negative of such technological developments.