【摘要】：信息時(shí)代計(jì)算機(jī)網(wǎng)絡(luò)技術(shù)的快速發(fā)展,人們已經(jīng)在逐步脫離純?nèi)斯さ膭趧?dòng),逐漸使用計(jì)算機(jī)來(lái)代替人們做一些高效、快速、海量的數(shù)據(jù)挖掘和分析。政府是為人民服務(wù)的,政府頒布的政策與法令需要得到廣大人民的認(rèn)可,所以政府需要了解人民的意向,于是就出現(xiàn)了民意調(diào)查。對(duì)于一個(gè)國(guó)家的人口來(lái)說(shuō),人口基數(shù)十分龐大如果單純依靠人力進(jìn)行民意調(diào)查的審查、統(tǒng)計(jì)和分析,就會(huì)浪費(fèi)巨大的人力與物力。以前,國(guó)家對(duì)于民意調(diào)查這方面就是采用純?nèi)斯ぶ?往往在統(tǒng)計(jì)分析的過(guò)程中,會(huì)錯(cuò)過(guò)一部分調(diào)查,會(huì)出現(xiàn)一些人工不必要的錯(cuò)誤。信息時(shí)代人們的工作方式逐漸向網(wǎng)絡(luò)上轉(zhuǎn)義,這個(gè)給予政府一個(gè)很大的契機(jī),民意調(diào)查這些動(dòng)輒上億的數(shù)據(jù)量,全部交給計(jì)算機(jī)來(lái)做,雖然信息量巨大,但還是可以接受的。本文根據(jù)政府對(duì)于民意調(diào)查的需要,設(shè)計(jì)開(kāi)發(fā)了政府民意調(diào)研分析系統(tǒng)。論文應(yīng)用Jsp設(shè)計(jì)頁(yè)面,主體部分采用JAVA設(shè)計(jì)語(yǔ)言,使用J2EE組件模型和SSH框架,利用JDBC連接數(shù)據(jù)庫(kù),使用最新的ORACLE 11g作為數(shù)據(jù)庫(kù)。完成了用戶前臺(tái)模塊和管理員后臺(tái)模塊的設(shè)計(jì),用戶前臺(tái)模塊又分為問(wèn)卷作答子模塊、歷史調(diào)研問(wèn)卷結(jié)果查看子模塊、用戶注冊(cè)子模塊和個(gè)人信息子模塊,問(wèn)卷作答子模塊包括調(diào)研問(wèn)卷查看、投票、IP過(guò)濾和意見(jiàn)填寫(xiě)四個(gè)部分,用戶注冊(cè)子模塊包括信息填寫(xiě)和實(shí)名認(rèn)證,個(gè)人信息子模塊包括查看、信息修改和密碼三個(gè)部分。管理員后臺(tái)模塊分為創(chuàng)建調(diào)研問(wèn)卷子模塊、查看調(diào)研結(jié)果子模塊、根據(jù)分析子模塊和用戶管理子模塊,查看調(diào)研結(jié)果包括正在進(jìn)行的問(wèn)卷結(jié)果、歷史結(jié)果和數(shù)據(jù)分析結(jié)果,數(shù)據(jù)分析模塊包括人員分析、職業(yè)分析、經(jīng)濟(jì)分析、參與度分析和意見(jiàn)分析,用戶管理模塊包括注冊(cè)用戶審核和臨時(shí)用戶分配。其中對(duì)用戶前臺(tái)模塊在IP過(guò)濾這里使用了 RFC(RecursiveFlowClassfication)算法,通過(guò)對(duì)IP的遞歸過(guò)濾來(lái)實(shí)現(xiàn)IP過(guò)濾的安全性,管理員后臺(tái)模塊這里使用了 k-means算法,通過(guò)對(duì)于大數(shù)據(jù)集中的不同關(guān)鍵詞的聚類(lèi)分析,生成不同的聚集,從而為后面的數(shù)據(jù)分析做準(zhǔn)備。政府民意調(diào)研分析系統(tǒng)隨著信息時(shí)代的到來(lái)而誕生,是在充分研究了計(jì)算機(jī)技術(shù)的發(fā)展模式、計(jì)算機(jī)的普及度等問(wèn)題之后推出的一個(gè)基于計(jì)算機(jī)的網(wǎng)絡(luò)調(diào)研分析系統(tǒng),實(shí)現(xiàn)了用計(jì)算機(jī)代替純?nèi)斯ぶ?減輕了人們的工作量,加強(qiáng)了調(diào)研結(jié)果的處理能力,提高了調(diào)研結(jié)果的處理速度和分析速度,大大加快了調(diào)研進(jìn)程,節(jié)約了大量的人力物力。
[Abstract]:With the rapid development of computer network technology in the information age, people have gradually separated from pure manual labor and gradually used computers to do some efficient, rapid and massive data mining and analysis instead of people. The government serves the people, and the policies and decrees issued by the government need to be recognized by the broad masses of the people, so the government needs to understand the intentions of the people, so there are opinion polls. For the population of a country, if the population base is very large, if we rely solely on manpower for the examination, statistics and analysis of public opinion polls, we will waste a lot of manpower and material resources. In the past, the state used pure manual system for public opinion polls, often in the process of statistical analysis, will miss some of the survey, there will be some unnecessary mistakes. The way people work in the information age is gradually turning to the Internet, which gives the government a great opportunity. Opinion polls, which are prone to hundreds of millions of data, are all handed over to computers. Although the amount of information is huge, it is still acceptable. According to the needs of the government for the public opinion survey, this paper designs and develops the government public opinion research and analysis system. In this paper, Jsp is used to design the page, the main part is JAVA design language, J2EE component model and SSH framework are used, JDBC is used to connect the database, and the latest ORACLE 11g is used as the database. The design of the user foreground module and the administrator background module is completed. The user foreground module is divided into the questionnaire answer sub-module, the history questionnaire result inspection sub-module, the user registration sub-module and the personal information sub-module. The sub-module of questionnaire answer includes four parts: questionnaire viewing, voting, IP filtering and opinion filling. The user registration sub-module includes information filling and real name authentication, and the personal information sub-module includes three parts: viewing, information modification and password. The administrator background module is divided into the sub-module of creating research questions, the sub-module of viewing the results of the research, the sub-module of the analysis and the sub-module of user management, and the results of the survey, including the results of the ongoing questionnaire, the historical results and the results of the data analysis. The data analysis module includes personnel analysis, career analysis, economic analysis, participation analysis and opinion analysis, and user management module includes registered user audit and temporary user allocation. The RFC (RecursiveFlowClassfication) algorithm is used for the user foreground module in IP filtering, and the security of IP filtering is realized by recurrent filtering of IP. The administrator background module uses k-means algorithm here. Through the cluster analysis of different keywords in big data concentration, different aggregates are generated, so as to prepare for the later data analysis. With the advent of the information age, the government public opinion research and analysis system was born, which is a computer-based network research and analysis system after fully studying the development mode of computer technology, the popularity of computer and so on. The computer is used instead of the pure manual system, which reduces the workload of people, strengthens the processing ability of the investigation results, improves the processing speed and analysis speed of the investigation results, greatly speeds up the investigation process, and saves a lot of manpower and material resources.
【學(xué)位授予單位】：東北大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類(lèi)號(hào)】：TP311.52

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 張玉芳,毛嘉莉,熊忠陽(yáng);一種改進(jìn)的K-means算法[J];計(jì)算機(jī)應(yīng)用;2003年08期

，

本文編號(hào)：2487354