基于结构和内容的股权众筹网站群体分析与挖掘

发布时间：2018-01-06 20:28

本文关键词：基于结构和内容的股权众筹网站群体分析与挖掘　出处：《南京大学》2017年硕士论文　论文类型：学位论文

【摘要】：经过20多年的发展,万维网已经成为了世界上最大的公共数据源,对这些海量且丰富的网络资源进行分析和挖掘具有极大的价值。通常情况下,可以从网络的结构、内容和使用三个角度着手。随着越来越多的机构将业务搬上互联网,各类网站就成了分析与挖掘的热门对象,包括各级政府官网、学术机构网站和众多的商业网站。股权众筹作为企业融资的新手段,是指在网站上发布某个项目,企业筹集一定数额的资金并出让相应的股权,投资者付出超过最低限额的资金获得相应的股权,网站在此过程中承担资金托管人的作用,并收取一定比例的佣金。股权众筹一定程度上缓解了中小企业融资困难的问题。经过了近几年的快速发展,目前国内股权众筹行业增长放缓,热度减退,行业发展回归理性,作为新的事物股权众筹网站有很多值得研究分析的地方,国内网站主要有创投类、实体店铺类和影视类三大类别,笔者认为创投类和实体店铺类更贴近一般民众的创业方式,所以选择这两类网站作为本次研究的对象。国外对股权众筹的研究有很多,涉及金融、法律、社会效益、用户行为等各个方面,而国内的相关研究主要集中在金融、法律领域,情报学视角下的研究很少。对国内股权众筹的发展现状也只能通过一些众筹行业报告得知,缺乏全面深入的理解。鉴于本学科在网络信息资源管理、Web数据挖掘方面的广泛研究,结合能够搜集到的网站相关数据,以创投类和实体店铺类网站为对象,选择从结构和内容两个角度股权众筹网站群体进行分析和挖掘,以期能够获得有价值的结论对网站的运营、股权众筹的发展、用户对网站的使用有所帮助。第一章绪论主要介绍本文的研究背景,国内外对股权众筹网站的研究现状,认为对股权众筹网站群体的全面分析与挖掘具有重要现实意义,结合本学科的相关研究、理论方法,确定了从结构和内容两个维度展开对股权众筹网站群体的实证分析。第二章相关理论研究进展,对本研究中涉及的链接分析、社会网络分析、聚类分析和关联规则挖掘这四种理论方法的来龙去脉作必要的阐述,并以此作为本次研究的基础。第三章介绍了数据搜集与预处理,由于没有完整准确的二手数据源,本研究需要对研究的40个股权众筹网站之间的URL共现数据和网站融资成功的项目数据进行抓取。再根据分析的需要将原始数据转化成规范的格式。第四章对股权众筹网站群体的网络结构进行社会网络分析,分别从整体、子网络和个体属性三个层面进行。第五章对股权众筹网站群体的核心内容,也就是网站融资成功的项目进行分析、挖掘,先后进行了项目的统计分析、网站的相似性分析和项目的关联规则挖掘。第六章总结与展望,本章首先总结全文的思路、研究方法和一些主要结论,分析研究中存在的不足以及进一步的研究方向。
[Abstract]:After more than 20 years of development, the World wide Web has become the largest public data source in the world. It is of great value to analyze and mine these massive and abundant network resources. As more and more organizations move their business to the Internet, all kinds of websites become the hot objects of analysis and mining, including government websites at all levels. Academic institutions website and numerous commercial websites. As a new means of corporate financing, crowdfunding refers to the publication of a project on the website, the enterprise raised a certain amount of funds and sold the corresponding equity. Investors pay more than the minimum amount of funds to obtain the corresponding equity, the site in the process of assuming the role of custodians of funds. And collect a certain proportion of commission. Stock raising to a certain extent alleviates the financing difficulties of small and medium-sized enterprises. After the rapid development of recent years, the current domestic stock raising industry growth slowed down, the heat declined. Industry development returns to rationality, as a new stock rights crowdfunding website has a lot of research and analysis of the place, domestic websites are mainly venture capital, physical stores and film and television categories of three categories. The author believes that venture capital and physical stores more close to the general public way of entrepreneurship, so choose these two types of websites as the object of this study. There are a lot of foreign research on stock financing, involving finance, law. Social benefits, user behavior and other aspects, and domestic related research is mainly focused on the financial, legal field. From the perspective of information science, there is little research. The development of domestic stock crowdfunding can only be learned through some crowdfunding industry reports, lack of a comprehensive and in-depth understanding. In view of the subject in the network information resource management. The extensive research on Web data mining, combined with the relevant data collected by the website, take the venture capital class and the physical store class as the object. Choose from the structure and content of the stock crowdfunding website community analysis and mining, in order to obtain valuable conclusions on the operation of the site, the development of equity crowdfunding. The first chapter mainly introduces the research background of this paper, the domestic and foreign research status of stock crowdfunding website. It is considered that the comprehensive analysis and mining of the stock rights crowdfunding website group has important practical significance, combined with the relevant research of this discipline, theoretical methods. Confirmed from the structure and content of the two dimensions of the stock rights website community empirical analysis. Chapter two related theoretical research progress, the link analysis involved in this study, social network analysis. Cluster analysis and association rules mining these four theoretical methods for the necessary elaboration of the background, and this as the basis of this study. Chapter three introduces the data collection and pretreatment. There is no complete and accurate secondary data source. This study needs to capture the URL co-occurrence data between 40 equity crowdfunding websites and the project data of successful website financing. Then, according to the need of analysis, the original data can be converted into a standard format. Chapter 4th. The network structure of stock capital crowdfunding website group is analyzed by social network. From the overall, sub-network and individual attributes of the three levels. 5th chapter on the core content of the stock crowdfunding website group, that is, the website financing success of the project analysis, mining. The statistical analysis of the project, the similarity analysis of the website and the mining of the association rules of the project have been carried out successively. 6th chapters are summarized and prospected. Firstly, this chapter summarizes the ideas, research methods and some main conclusions of the full text. Analysis of the shortcomings of the study and the direction of further research.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F49;F832.51

【相似文献】