本文關(guān)鍵詞：網(wǎng)絡(luò)借貸業(yè)務(wù)個(gè)人信用評(píng)價(jià)方法研究　出處：《合肥工業(yè)大學(xué)》2016年博士論文　論文類(lèi)型：學(xué)位論文

  更多相關(guān)文章： 網(wǎng)絡(luò)借貸 信用評(píng)價(jià) 社會(huì)資本 信用特征選擇 信用評(píng)價(jià)模型

【摘要】：網(wǎng)絡(luò)借貸信用評(píng)價(jià)能夠有效地緩解借貸雙方間的信息不對(duì)稱(chēng)性,降低違約風(fēng)險(xiǎn)和交易成本。然而,網(wǎng)絡(luò)借貸業(yè)務(wù)中借款人的財(cái)務(wù)信息難以獲取和驗(yàn)證,給傳統(tǒng)的基于財(cái)務(wù)信息的信用評(píng)價(jià)方法帶來(lái)巨大的困難。事實(shí)上,網(wǎng)絡(luò)環(huán)境下,借款人的信用相關(guān)數(shù)據(jù)不僅包括財(cái)務(wù)信息,也包括非財(cái)務(wù)信息。這些非財(cái)務(wù)信息廣泛分布在不同的網(wǎng)絡(luò)平臺(tái)中,具有體量大、價(jià)值密度低和質(zhì)量參差不齊等特點(diǎn),給網(wǎng)絡(luò)借貸信用評(píng)價(jià)研究帶來(lái)新的困難。為此,本文在綜述信用評(píng)價(jià)相關(guān)理論與方法的基礎(chǔ)上,結(jié)合網(wǎng)絡(luò)借貸業(yè)務(wù)的特點(diǎn),從信用評(píng)價(jià)的數(shù)據(jù)預(yù)處理、信用特征選擇和信用評(píng)價(jià)模型構(gòu)建三個(gè)方面,對(duì)網(wǎng)絡(luò)借貸業(yè)務(wù)的信用評(píng)價(jià)問(wèn)題展開(kāi)研究。研究的主要內(nèi)容和相關(guān)結(jié)論如下。(1)網(wǎng)絡(luò)借貸個(gè)人信用評(píng)價(jià)數(shù)據(jù)預(yù)處理方法網(wǎng)絡(luò)借貸業(yè)務(wù)信用評(píng)價(jià)數(shù)據(jù)的質(zhì)量參差不齊,缺失值和異常值現(xiàn)象嚴(yán)重。在缺失值處理方面,針對(duì)多重填補(bǔ)法難以對(duì)包含類(lèi)別變量的數(shù)據(jù)集進(jìn)行缺失值填補(bǔ)的問(wèn)題,提出一種分類(lèi)多重填補(bǔ)法。該方法利用類(lèi)別變量與連續(xù)變量數(shù)學(xué)特征間的關(guān)系,根據(jù)類(lèi)別變量信息對(duì)連續(xù)變量的相關(guān)數(shù)據(jù)特征進(jìn)行估計(jì),提高缺失值填補(bǔ)效果。在異常值處理方面,針對(duì)單一信用特征的異常值處理問(wèn)題,提出一種基于KNN的異常值糾偏方法,該方法能夠利用近鄰樣本的相關(guān)信用特征對(duì)異常值進(jìn)行糾正。針對(duì)密度分布不均勻空間中的異常樣本檢測(cè)的距離閾值難以確定問(wèn)題,提出一種基于DBSCAN和相對(duì)密度的異常樣本檢測(cè)方法。該方法首先利用DBSCAN算法將密度分布不均勻空間分成若干個(gè)密度分布均勻的類(lèi),然后在每個(gè)類(lèi)中運(yùn)用相對(duì)密度方法確定異常樣本。最后,在拍拍貸平臺(tái)上進(jìn)行信用數(shù)據(jù)預(yù)處理實(shí)驗(yàn),結(jié)果表明經(jīng)過(guò)預(yù)處理的數(shù)據(jù)能夠顯著增強(qiáng)信用評(píng)價(jià)模型的性能。(2)網(wǎng)絡(luò)借貸個(gè)人信用特征選擇方法網(wǎng)絡(luò)借貸業(yè)務(wù)信用信息的體量大,價(jià)值密度低。需要結(jié)合相關(guān)理論與方法,對(duì)信用特征進(jìn)行定性初選和定量篩選。在信用特征定性初選階段,結(jié)合信用所具有的資本性和社會(huì)資本理論,從結(jié)構(gòu)維度、關(guān)系維度和認(rèn)知維度三個(gè)方面,分析網(wǎng)絡(luò)借貸業(yè)務(wù)中借款人的社會(huì)資本,并結(jié)合借款人的個(gè)人信息、借款歷史信息和身份驗(yàn)證信息等,研究融合社會(huì)資本的信用特征定性初選方法。在信用特征的定量篩選階段,考慮到信用特征的變量類(lèi)型多樣且與信用狀態(tài)變量間的關(guān)系復(fù)雜,提出一種基于綜合定量分析的信用特征篩選方法,該方法運(yùn)用相關(guān)分析、卡方統(tǒng)計(jì)量分析、信息增益分析和支持向量回歸分析等定量分析方法,篩選與信用狀態(tài)變量具有線性和非線性關(guān)系的定類(lèi)與定距信用特征。最后,在拍拍貸平臺(tái)上的實(shí)驗(yàn)結(jié)果表明,提出的信用特征初選和篩選方法能夠全面獲取多變量類(lèi)型和多關(guān)系類(lèi)型的信用特征。(3)網(wǎng)絡(luò)借貸個(gè)人信用評(píng)價(jià)模型現(xiàn)有的信用評(píng)價(jià)模型在網(wǎng)絡(luò)借貸環(huán)境下的應(yīng)用效果不佳,需要對(duì)現(xiàn)有模型進(jìn)行改進(jìn)并結(jié)合網(wǎng)絡(luò)借貸環(huán)境下信用表現(xiàn)出的相關(guān)特性,設(shè)計(jì)新的信用評(píng)價(jià)模式和模型。現(xiàn)有的Adaboost集成學(xué)習(xí)模型僅根據(jù)誤分類(lèi)率調(diào)整基分類(lèi)器的樣本權(quán)重,忽略了分歧度和誤分類(lèi)成本等因素對(duì)于樣本權(quán)重的影響,造成集成后的模型精確度下降。為此,提出一種基于分歧度與誤分代價(jià)的Adaboost信用評(píng)價(jià)模型,該模型能夠?qū)Ψ诸?lèi)困難樣本和誤分代價(jià)高的樣本進(jìn)行有針對(duì)性的學(xué)習(xí),提高信用評(píng)價(jià)結(jié)果的準(zhǔn)確性。在拍拍貸平臺(tái)上的實(shí)驗(yàn)結(jié)果表明,基于分歧度和誤分代價(jià)的Adaboost信用評(píng)價(jià)模型的性能顯著優(yōu)于傳統(tǒng)的Adaboost模型。此外,網(wǎng)絡(luò)借貸環(huán)境下,根據(jù)信用表現(xiàn)出的全息性,設(shè)計(jì)一種Peer-to-Peer協(xié)同信用分析機(jī)制,獲取并集成評(píng)價(jià)對(duì)象在多個(gè)網(wǎng)絡(luò)平臺(tái)上的相關(guān)信用特征,建立基于協(xié)同分析模式的跨業(yè)務(wù)的信用評(píng)價(jià)模型,從而對(duì)借款人的信用做出更加全面的評(píng)價(jià)。在相關(guān)的網(wǎng)絡(luò)借貸平臺(tái)、電子商務(wù)平臺(tái)和社會(huì)網(wǎng)絡(luò)平臺(tái)上的實(shí)驗(yàn)結(jié)果表明,基于協(xié)同分析的跨業(yè)務(wù)信用評(píng)價(jià)模型能夠有效地提升信用評(píng)價(jià)結(jié)果的有效性。
[Abstract]:The evaluation of the network credit lending can effectively alleviate the information asymmetry between the two sides to reduce the default risk and transaction cost. However, the network lending business in the financial information is difficult to obtain and verify, to the traditional credit evaluation method based on financial information has brought great difficulties. In fact, under the network environment, the borrower credit related data not only include financial information, including non financial information. These non financial information are widely distributed in different network platform, has the characteristics of large amount of low value density and uneven quality, bring new difficulties to the evaluation of the network lending credit. Therefore, based on the review of credit evaluation theory and method. Combined with the characteristics of the network lending business, from data preprocessing of credit evaluation, the construction of three aspects of credit feature selection and credit evaluation model of network Research on the credit evaluation problem. Lending business the main research contents and conclusions are as follows. (1) the network lending personal credit evaluation data preprocessing method of network lending business credit evaluation data quality uneven, missing values and outliers seriously. Processing value in the absence, for the multiple imputation method is difficult to contain type variables the data sets fill problem of missing values, this paper proposed a classification of multiple imputation. The method of categorical variables and continuous variables, the mathematical relation between features, estimated according to the data characteristics of continuous variables and class variables, improve the effect of missing values missing. In treatment of abnormal, the abnormal value for a single credit characteristic to deal with the problem, this paper proposes a novel value correction method based on KNN, this method can correct the abnormal value of the credit characteristics of neighbor samples Is the distance threshold. According to the abnormal samples of uneven density distribution in space detection is difficult to determine the problem, this paper proposes a novel sample DBSCAN and relative density detection method based on using the method of DBSCAN algorithm with non-uniform density space into a plurality of uniform density distribution, and then use the method to determine the relative density of abnormal samples in each class. Finally, experiments of credit data preprocessing in a pat on the loan platform. The results show that the preprocessed data can significantly enhance the performance of credit evaluation model. (2) lending network personal credit feature selection method of network lending business credit information of the large, low value density. According to the related theory and method the credit characteristics, qualitative and quantitative screening. In the primary credit characteristics of qualitative theory of primary stage, with the combination of credit capital and social capital, From the three aspects of structure dimension, relational dimension and cognitive dimension, network analysis of the borrower lending business in social capital, and combined with the borrower's personal information, borrowing history information and authentication information, the credit characteristics of primary qualitative study on fusion method of social capital. In the quantitative characteristics of credit screening stage, taking into account the characteristics of credit variables various types of credit and the relationship with the state variables is complex, a screening method of credit characteristics of comprehensive quantitative analysis based on the method of using correlation analysis, chi square statistic analysis method to analyze the information gain analysis and support vector regression analysis, quantitative screening and credit of state variables is linear and nonlinear relationship with from the characteristics of credit. Finally, in a pat on the loan on the platform. The experimental results show that the proposed method can select the characteristics of credit primaries and full access to more than The characteristics of credit variable types and multiple relation types. (3) the application effect of network lending personal credit evaluation model of credit model in the existing network lending environment is poor, need to be improved and combined with the characteristics of the network lending environment credit showed on the existing model, design a new credit evaluation model and Adaboost model. The existing ensemble learning model only based on sample weight classification error rate adjustment of base classifiers, ignoring the influence of divergence and misclassification cost factors for sample weight, decrease the model accuracy after integration. Therefore, this paper proposes a model for divergence and misclassification cost Adaboost credit evaluation based on the model to classification difficult samples and high misclassification cost for targeted learning, improve the accuracy of credit evaluation results. In a pat on the loan on the platform. The experimental results show that the base The Adaboost model in credit evaluation of Adaboost divergence and misclassification cost model significantly outperforms the traditional. In addition, the network lending environment, according to the holographic credit the credit analysis mechanism of collaborative design of a Peer-to-Peer, acquisition and integration evaluation objects in multiple network platform related credit characteristics, the establishment of credit evaluation cross business collaborative analysis model based on the model, make the evaluation more comprehensive and credit of the borrower. The related network lending platform, e-commerce platform and social network platform. The experimental results indicate that the evaluation of the effectiveness of cross business credit cooperative analysis model can effectively improve the credit evaluation based on the results.

【學(xué)位授予單位】：合肥工業(yè)大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2016
【分類(lèi)號(hào)】：F832.4;F724.6

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 黃秋_g;史小康;;個(gè)人信用風(fēng)險(xiǎn)評(píng)分的指標(biāo)選擇研究[J];新疆財(cái)經(jīng)大學(xué)學(xué)報(bào);2015年03期

2 李少波;魏中賀;孟偉;;基于距離的數(shù)據(jù)流在線檢測(cè)算法研究[J];計(jì)算機(jī)應(yīng)用研究;2015年12期

3 王莉君;何政偉;馮平興;;基于ICA的異常數(shù)據(jù)挖掘算法研究[J];電子科技大學(xué)學(xué)報(bào);2015年02期

4 郭昱;馬翻翻;鄭超文;;我國(guó)小微企業(yè)信用評(píng)價(jià)指標(biāo)體系的構(gòu)建[J];金融經(jīng)濟(jì);2015年02期

5 陳運(yùn)森;;社會(huì)網(wǎng)絡(luò)與企業(yè)效率:基于結(jié)構(gòu)洞位置的證據(jù)[J];會(huì)計(jì)研究;2015年01期

6 高敬陽(yáng);陳程立詔;朱群雄;;基于爭(zhēng)議度的Boosting集成網(wǎng)絡(luò)樣本權(quán)值調(diào)整算法[J];中南大學(xué)學(xué)報(bào)(自然科學(xué)版);2012年11期

7 武森;馮小東;單志廣;;基于不完備數(shù)據(jù)聚類(lèi)的缺失數(shù)據(jù)填補(bǔ)方法[J];計(jì)算機(jī)學(xué)報(bào);2012年08期

8 姜明輝;謝行恒;王樹(shù)林;溫瀟;;個(gè)人信用評(píng)估的Logistic-RBF組合模型[J];哈爾濱工業(yè)大學(xué)學(xué)報(bào);2007年07期

9 任亮;;社會(huì)資本理論的五個(gè)命題[J];探索;2007年03期

10 肖文兵;費(fèi)奇;;基于支持向量機(jī)的個(gè)人信用評(píng)估模型及最優(yōu)參數(shù)選擇研究[J];系統(tǒng)工程理論與實(shí)踐;2006年10期

相關(guān)博士學(xué)位論文 前3條

1 向暉;個(gè)人信用評(píng)分組合模型研究與應(yīng)用[D];湖南大學(xué);2011年

2 曾勇;電子商務(wù)信用風(fēng)險(xiǎn)機(jī)理研究[D];武漢理工大學(xué);2005年

3 沈翠華;基于支持向量機(jī)的消費(fèi)信貸中個(gè)人信用評(píng)估方法研究[D];中國(guó)農(nóng)業(yè)大學(xué);2005年

相關(guān)碩士學(xué)位論文 前2條

1 周韻然;基于流形學(xué)習(xí)的A股上市公司抽樣的信用評(píng)價(jià)[D];電子科技大學(xué);2014年

2 石麗;多重插補(bǔ)在成分?jǐn)?shù)據(jù)缺失值補(bǔ)全中的應(yīng)用[D];山西大學(xué);2012年

，

本文編號(hào)：1431723