论大数据分析的方法论意义

发布时间：2018-01-05 21:12

本文关键词：论大数据分析的方法论意义　出处：《武汉科技大学》2015年硕士论文　论文类型：学位论文

更多相关文章： 大数据 大数据分析 方法论 局限性

【摘要】：数据科学的发展经历了数据的产生、科学数据的形成和大数据的诞生三个阶段。三个阶段的不断发展改变了也加深了人们用数据看世界的观念。随着现代测量技术和智能设备的发展，数据的形式由传统单一的结构化数据演变为由结构化数据、半结构化数据和非结构化数据组成的复杂的数据形式。而互联网、移动互联网的快速发展给了人们以分享这些数据的平台，由此产生的海量数据被人们发现和利用，就产生了新型的大数据概念。大数据分析从技术手段上采用了最新的数据分析模型，通过数据之间特有的相关关系可以产生出许多有关联、有价值的结论。大数据分析在许多领域发挥了巨大的作用。在商业领域，，可收集的数据量非常巨大，不管是消费者的直接的消费数据，还是日常生活数据都可以被收集，经过对收集来的海量数据进行相关性分析之后，可以让营销者实现精准营销，增加成交量；在科学领域，比如海洋科学、生物科学、医疗健康科学以及天文学领域，由于历史积累的数据量本来就大，加上现在的先进技术产生的数据，导致数据量更是无比巨大，科学家通过大数据分析，可能会在这些海量数据基础上形成数据密集型的科学发现及第四范式。在其他能产生大量数据的领域，大数据分析技术都能产生较大的效用。从理论上说，大数据技术与传统的数据分析技术的根据基本一样。传统数据分析技术所具有的弊端，大数据同样可能有。尤其是在数据源和数据的完整性方面，大数据分析技术的局限性是非常明显的。因为海量的数据并不能彼此保证每个数据都是真实可靠的，而且所谓海量数据也只是相对的大量，并不是有关方面的全部数据。所以，通过大数据分析获得的结论可能具有更大的更多的实用价值，但并不必然为真。而且大数据分析对相关人群的隐私权可能造成某种程度的侵害。因此，在应用大数据技术时应科学合理，不可滥用。
[Abstract]:The development of data science has experienced the generation of data. The formation of scientific data and the birth of big data. The development of the three stages has changed and deepened the concept of people to see the world with data. With the development of modern measurement technology and intelligent equipment. The form of data changes from traditional single structured data to complex data form composed of structured data, semi-structured data and unstructured data. The rapid development of mobile Internet has given people a platform to share these data, and the resulting massive data has been discovered and utilized. Big data analysis adopts the latest data analysis model from the technical means, and through the special correlation relationship between the data, it can produce many relevant and valuable conclusions. Big data analysis has played a significant role in many areas. In business, the amount of data that can be collected is very large, whether it is consumer direct consumption data or daily life data can be collected. After the correlation analysis of the massive data collected, the marketers can achieve accurate marketing and increase the turnover. In the fields of science, such as marine science, biological science, medical and health sciences, and astronomy, the amount of data accumulated by history is already large, coupled with data generated by modern advanced technologies. Resulting in a huge amount of data, scientists through big data analysis, on the basis of these massive data may form a data-intensive scientific discovery and fourth paradigm, in other areas that can produce large amounts of data. Big data analysis technology can produce greater utility. In theory, the big data technology is basically the same as the traditional data analysis technology. The traditional data analysis technology has its drawbacks. Big data may also be. Especially in terms of data sources and data integrity, big data analysis technology limitations are very obvious, because the huge amount of data can not guarantee that every data is true and reliable. And the so-called massive data is only relatively large, not all the relevant data. Therefore, the conclusions obtained through big data analysis may have more practical value. But it is not necessarily true. And big data's analysis of the privacy of the people concerned may cause a certain degree of infringement. Therefore, the application of big data technology should be scientific and reasonable, can not be abused.
【学位授予单位】：武汉科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP311.13

【参考文献】