基于Apriori算法的花呗舆情关联规则挖掘
摘 要
随着云计算、大数据、物联网、人工智能等新一代信息技术的飞速发展,我国的基本金融服务已经覆盖99%的人口,金融服务供给主体越来越多。互联网金融消费产品花呗,2014年进入互联网消费领域,凭借大数据平台和先进的技术手段优势发展速度很快,在市场中占据很大份额,互联网产品花呗在满足用户消费欲望的同时,同时也为用户带来债务。网民们习惯于在网络平台上发布信息,对社会热点事件和热点话题的讨论,使得网络上不断更新海量信息,最终成为网络舆情热点。
因此结合网络舆情研究互联网金融产品花呗可以分析互联网借贷产品,在满足年轻人消费消费欲望时,出现“消费一直爽,还款两行泪”现象以及金融行业面临监管的问题,有助于完善金融监管机构针对该行业的发展现状确立针对性的法律文件,对该行业的业务范围进行明确。保护用户的合法权益,以防止“钻空子”现象发生,引导该行业向着正确的方向发展,促进互联网金融产品健康有序的发展,从而推动互联网金融的进一步发展。
基于以上研究背景将花呗作为研究对象,本文首先依据微博爬取数据,对文本数据分词、过滤停止词、利用VSM模型转化为向量、数据降维处理采用TF-IDF方法抽取关键词:互联网金融消费;舆情分析;SVM;Apriori
SVM;Apriori
Abstract
With the rapid development of new information technologies such as cloud computing,big data, Internet of Things and artificial intelligence,China's basic financial services have covered 99% of the population, and there are more and more financial service providers. Internet financial consumer products flower gardenentered the field of Internet consumption in 2014. With the advantages of big data platform and advanced technical means, it developed rapidly and occupied a large share in the market. Internet products flower garden not only satisfied users' consumption desires, but also brought debts to users. Netizens are used to publishing information on the network platform,and discussing social hot events and hot topics makes the network constantly update massive information, which eventually becomes a hot spot of network public opinion.
Therefore, the study of Internet financial products based onInternet public opinion can analyze Internet lending products. When satisfying young people's desire for consumption,there is a phenomenon of "two tears in repayment" andthe financial industry is facing regulatory problems,which is helpful to improve the legal documents established by the financial regulatory agencies for the development status of the industry and clarify the business scope of the industry. To protect the legitimate rights and interests of users,so as to prevent the phenomenon of "exploiting loopholes", guide the industry to develop in the right direction,promote the healthy and orderly development of Internet financial products, and thus promote the further development of Internet finance.
Based on the above research background, the flower buds are taken as the research object.Firstly, according to the microblog crawling data, the text data is segmented,stopword is filtered, VSM model is used to transform into vector,and the data dimension reduction is processed by TF-IDF method toextract keywords, extract text feature words,and carry out word frequency statistical classification analysis; Secondly, the SVM model of public opinion theme classification is established to realize automatic classification of public opinion. Based on SVM classification results,the problems reflected by public opinion are analyzed and the causes of problems reflected by classification results are explored. Then, combined with the text data mining method, Apriori algorithm,the relationship between high-frequency keywords and its influencing factors are further explored Finally,based on the previous in-depth analysis of the specific public opinion content of the flower garden, we can better grasp the problems and influencing factors of the flower garden reflected by the public opinion published by users, and give relevant suggestions and measures based on the previous research and analysis results.
The specific suggestions are as follows: improve the supervision and management of governmentagencies, further improve the financial supervision system of this industry,and clarify the main body and scope of supervision duties;To establish and improve the Internet financial consumption credit information system, it is necessary to establish a targeted credit rating system. The credit information, repayment and default information of the flower garden are includedin the personal credit information record of the central bank;Ant Group adjusts the product model to adjust the user quota of young people, and opens a new function of self-management. Users can set the adjustment quota,consumption reminder and adjust thepayment order independently; Users need to "tailor-made" their personal consumption and choose the appropriate amount of advance consumption.
Key Words: Internet Financial Consumption; Public Opinion Analysis; SVM; Apriori
|
基于Apriori算法的花呗舆情关联规则挖掘
更新时间:2023-03-29
上一篇:基于GSM的智能指纹门禁系统
下一篇:基于强化学习的水下定位研究