本文選題：企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警　切入點(diǎn)：支持向量機(jī)　出處：《吉林大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：隨著我國社會經(jīng)濟(jì)市場化和國際化程度越來越高，企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)管理和預(yù)警的重要性愈加凸顯。財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警體系，研究企業(yè)或者政府等的財(cái)務(wù)狀況，結(jié)合歷史數(shù)據(jù)和累積的經(jīng)驗(yàn)，對財(cái)務(wù)狀況是否會有風(fēng)險(xiǎn)做出預(yù)警。企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警涉及企業(yè)各方面海量的數(shù)據(jù)，不僅與企業(yè)內(nèi)部因素有關(guān)，也且與行業(yè)因素和外部大環(huán)境有關(guān)。進(jìn)行企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警的研究，首先要搞清楚以下幾個問題：第一，企業(yè)每天都產(chǎn)生了哪些數(shù)據(jù)，哪些數(shù)據(jù)對研究有用，哪些數(shù)據(jù)對研究沒用；第二，在企業(yè)產(chǎn)生的海量數(shù)據(jù)中，哪些與財(cái)務(wù)有關(guān)，哪些是主要因素，哪些是次要因素；第三，通過分析，判斷企業(yè)處于財(cái)務(wù)風(fēng)險(xiǎn)的哪個階段，應(yīng)該采取哪些措施，來盡量減少企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)給企業(yè)造成的可能的損失。 傳統(tǒng)上，人們在進(jìn)行財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警的時候，使用的方法多種多樣，但是，它們都要求較多的前提條件，這些限制條件往往影響了預(yù)測結(jié)果的準(zhǔn)確性和預(yù)警的可靠性與普遍性�，F(xiàn)在，支持向量機(jī)、神經(jīng)網(wǎng)絡(luò)等數(shù)據(jù)挖掘方法在預(yù)測方面被廣泛應(yīng)用，具有很高的可信度，，而且可以解決具有大量參數(shù)的難題。本文就采用支持向量機(jī)作為企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警的方法，當(dāng)然，本文建立的企業(yè)財(cái)務(wù)預(yù)警體系不僅可應(yīng)用于企業(yè)，還可應(yīng)用于政府等各種組織機(jī)構(gòu)。 針對傳統(tǒng)財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警領(lǐng)域的問題，本文提出了基于支持向量機(jī)的企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警模型，并將基于支持向量機(jī)的企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警模型和面向服務(wù)的體系架構(gòu)結(jié)合起來，實(shí)現(xiàn)了企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警服務(wù)與面向服務(wù)的體系架構(gòu)的組合，使得財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警的結(jié)果可以進(jìn)行及時的發(fā)布�？茖W(xué)可靠有效的企業(yè)財(cái)務(wù)風(fēng)險(xiǎn)預(yù)警體系，通過長期的觀測和分析，可以在前期就對可能出現(xiàn)的財(cái)務(wù)風(fēng)險(xiǎn)進(jìn)行預(yù)報(bào)和警示，并將預(yù)警結(jié)果進(jìn)行及時發(fā)布，使得企業(yè)可以及時采取措施，來減少企業(yè)損失。
[Abstract]:With the increasing degree of marketization and internationalization of social economy in our country, the importance of financial risk management and early warning of enterprises is becoming more and more prominent. The financial risk warning system studies the financial situation of enterprises or governments, etc. Combining with historical data and accumulated experience, we can make early warning on whether there will be risks in the financial situation. Enterprise financial risk early warning involves a large amount of data from all aspects of the enterprise, not only related to the internal factors of the enterprise. It is also related to the industry factors and the external environment. To carry out the research on early warning of enterprise financial risk, we should first find out the following questions: first, what data are generated by the enterprise every day and which data are useful for the research? Which data are useless for research; second, which of the vast amounts of data generated by enterprises are related to finance, which are major factors, and which are secondary factors; and third, through analysis, we can determine which stage of financial risk the enterprise is in. What measures should be taken to minimize the possible losses caused by financial risks. Traditionally, people have used a variety of methods for early warning of financial risks, but they all require more preconditions. These constraints often affect the accuracy of prediction results and the reliability and universality of early warning. Nowadays, support vector machines, neural networks and other data mining methods are widely used in prediction, and have high credibility. In this paper, support vector machine is adopted as the method of enterprise financial risk warning, of course, the enterprise financial early-warning system established in this paper can not only be applied to enterprises, It can also be applied to government and other organizations. Aiming at the problems in the traditional financial risk early warning field, this paper puts forward the enterprise financial risk early warning model based on support vector machine, and combines the enterprise financial risk warning model based on support vector machine with the service-oriented architecture. The combination of enterprise financial risk early warning service and service-oriented architecture is realized, and the results of financial risk warning can be released in time. The scientific, reliable and effective enterprise financial risk warning system, through long-term observation and analysis, The financial risk may be forecasted and warned in the early stage, and the early warning result can be issued in time, so that the enterprise can take measures in time to reduce the loss of the enterprise.
【學(xué)位授予單位】：吉林大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TP393.09;TP181

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 王恩德;;利用SOA構(gòu)建新一代企業(yè)管理信息系統(tǒng)[J];吉林大學(xué)學(xué)報(bào)(信息科學(xué)版);2006年03期

2 宋力,李晶;上市公司財(cái)務(wù)危機(jī)預(yù)警模型的實(shí)證研究[J];財(cái)經(jīng)論叢(浙江財(cái)經(jīng)學(xué)院學(xué)報(bào));2004年01期

3 陳良華,孫健;公司治理與財(cái)務(wù)困境:來自上海股票市場的證據(jù)[J];東南大學(xué)學(xué)報(bào)(哲學(xué)社會科學(xué)版);2005年05期

4 吳世農(nóng),盧賢義;我國上市公司財(cái)務(wù)困境的預(yù)測模型研究[J];經(jīng)濟(jì)研究;2001年06期

5 陳靜;上市公司財(cái)務(wù)惡化預(yù)測的實(shí)證分析[J];會計(jì)研究;1999年04期

6 張玲;財(cái)務(wù)危機(jī)預(yù)警分析判別模型[J];數(shù)量經(jīng)濟(jì)技術(shù)經(jīng)濟(jì)研究;2000年03期

7 楊保安,季海;基于人工神經(jīng)網(wǎng)絡(luò)的商業(yè)銀行貸款風(fēng)險(xiǎn)預(yù)警研究[J];系統(tǒng)工程理論與實(shí)踐;2001年05期

8 嚴(yán)太山;BP神經(jīng)網(wǎng)絡(luò)在學(xué)生素質(zhì)綜合測評系統(tǒng)中的應(yīng)用[J];中國科技信息;2005年13期

9 姜秀華,任強(qiáng),孫錚;上市公司財(cái)務(wù)危機(jī)預(yù)警模型研究[J];預(yù)測;2002年03期

本文編號：1609690