基于数据挖掘的图书馆长尾服务研究

发布时间：2018-03-01 07:11

本文关键词： 数据挖掘长尾理论图书馆服务离群数据挖掘　出处：《北京交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：互联网的高速发展影响着当今社会的各行各业,图书馆服务业首当其冲。电子信息出版商、互联网内容提供商选择积极抢占图书信息市场,与搜索引擎、信息咨询公司等信息服务提供商达成创新模式合作。这导致图书馆的用户粘性逐渐减弱,以往的忠实读者纷纷转而投向互联网大数据资源的怀抱。图书馆面临着严峻的发展形势。如何打破传统思维模式的桎梏,将图书馆的传统优势与先进的理论和技术相结合,提高图书馆服务效益,是当下图书馆服务业最为艰巨的任务。传统的图书馆服务业遵循"二八定律",认为20%的馆藏资源能满足80%的读者需求。本文通过文献分析、理论论证和实证研究,指出在互联网和大数据的新经济形势下,传统的"二八定律"已经不能满足用户多元化、个性化的需求。曾在商业经济领域被成功验证的"长尾理论"也适用于图书馆服务领域。本文在准确把握图书馆服务过程的两个主体——图书馆资源和用户的基础上,从长尾理论的三大推动力量和图书馆服务发展历程的三个阶段的角度,阐释了长尾现象在图书馆服务领域产生、发展和繁荣的历程;从图书馆服务的产业属性角度,论证了长尾理论在图书馆服务领域的适用性;从当今图书馆所处的竞争形势和信息共享的角度,论证了长尾理论应用在图书馆服务领域的必要性。本文创新性地提出了"满足需求"和"创造需求"两种图书馆长尾服务的开发策略。在后者的指导下,对长尾理论、数据挖掘、文本挖掘等相关理论、概念和方法进行研究,选择适用于探索图书馆长尾服务模式的理论和方法,构建了"图书馆长尾服务数据挖掘体系",提出了"基于两次聚类的文本离群数据挖掘算法"。最后,运用学位论文的真实数据集进行了实证研究,得到了有价值的结果。
[Abstract]:The rapid development of the Internet has affected all kinds of industries in today's society, and the library service industry bears the brunt. Electronic information publishers and Internet content providers choose to actively seize the book and information market, and search engines, Information service providers, such as information consulting firms, have agreed on innovative modes of collaboration. This has led to a gradual decline in the user stickiness of libraries. The loyal readers of the past have turned to the embrace of big data resources on the Internet. The library is facing a severe development situation. How to break the shackles of the traditional thinking mode and combine the traditional advantages of the library with the advanced theory and technology, Improving the efficiency of library service is the most arduous task of the library service industry at present. The traditional library service industry follows the "law of two or eight" and holds that 20% of the collection resources can meet the needs of 80% readers. Theoretical argumentation and empirical research point out that under the new economic situation of the Internet and big data, the traditional "two-eighth law" can no longer satisfy the diversity of users. The "long tail theory", which has been successfully verified in the field of commercial economy, is also applicable to the field of library service. This paper, on the basis of accurately grasping the two main bodies of library service process-library resources and users, From the perspective of the three driving forces of the long tail theory and the three stages of the development of library service, this paper expounds the process of the emergence, development and prosperity of the long tail phenomenon in the field of library service, and from the angle of the industrial attribute of the library service, This paper demonstrates the applicability of the long tail theory in the field of library service, and from the point of view of the competitive situation and information sharing in the library today, This paper demonstrates the necessity of the application of long-tail theory in library service field. This paper innovatively puts forward two kinds of development strategies of "satisfying demand" and "creating demand". Under the guidance of the latter, the author discusses the theory of long tail. Data mining, text mining and other related theories, concepts and methods are studied. This paper constructs a "data mining system of long tail service of library", and puts forward a "text outlier data mining algorithm based on two-cluster clustering". Finally, an empirical study is carried out using the real data set of the dissertation, and valuable results are obtained.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：G252

【相似文献】