CN105869047A - Multidimensional method for evaluating mobile banking user credibility - Google Patents

Abstract

The invention discloses a multidimensional method for evaluating mobile banking client credibility. The method includes the following steps: building a mobile banking client direct credibility model, building a mobile banking client indirect credibility model, building a mobile banking finance service credibility model based on direct credibility and indirect credibility between mobile users and mobile banking, building a mobile banking credit deviation degree model, building a mobile banking client credit punishment value model; based on near-term credit values and long-term credit values between mobile users and mobile banking, acquiring a credibility evaluation result. The method establishes a multidimensional credit mobile banking finance risk evaluation index system, integrates the relation between mobile banking payment frequency and trust, the mobile banking credit deviation degree, the client credit punishment value, and the like into a credit level evaluation. Compared with prior art, the method has a more comprehensive evaluation, and has a more accurate assessment result.

Description

A kind of Mobile banking customer trust degree evaluation method of various dimensions

Technical field

The invention belongs to field of information security technology, more particularly, to Mobile banking's customer trust of a kind of various dimensions Degree evaluation method.

Background technology

Mobile banking is also referred to as mobile banking (Mobile Bank), is made by mobile communication and internet platform, makees with mobile phone For terminal, provide a kind of finance service means of bank service to client.Along with the development of the 4G communication technology, each bank opens one after another Lead to mobile banking service.Mobile banking service, while providing the user great convenience, there is also potential business risk. The factor affecting Mobile banking's security includes, operational risk, risks of trust, information asymmetry risk, liquidity risk etc..

In prior art, many employing Mobile bankings reliability rating evaluation method sets up trust model, determines that Mobile banking uses The creditworthiness at family；By the consumption habit of Mobile banking consumer, user's liveness and viscosity, distributed letter based on prestige Appoint administrative model, the credibility of quantitative evaluation node；For the relation of Mobile banking's payment frequency Yu trust, Mobile banking believes Departure degree, the factor such as customer trust penalty value is appointed not yet to evaluate in view of reliability rating.

Summary of the invention

For disadvantages described above or the Improvement requirement of prior art, the invention provides the Mobile banking client of a kind of various dimensions Degree of belief evaluation method, its object is to improve the degree of accuracy that Mobile banking's reliability rating is evaluated.

For achieving the above object, according to one aspect of the present invention, it is provided that the Mobile banking client letter of a kind of various dimensions Appoint degree evaluation method, comprise the steps:

(1) according to the financial payment number of times in business hours section, the direct trust metric model of Mobile banking client is set up, specifically As follows:

$D_n(i,j)=\left\{\begin{array}{c}\frac{\underset{k=1}{\overset{m}{\Σ}}f(i,j)}{m},\\ 0,m=0.\end{array}\right.m\≠0;$

Wherein, D_n(i j) refers to the direct degree of belief between cellphone subscriber i and Mobile banking j in the n-th business hours；m Referring to, in n-th business, carry out the total degree of financial payment between cellphone subscriber i and Mobile banking j, k is financial payment time Number numbering；(i j) is the trust propensity value between cellphone subscriber i and Mobile banking j to f；

(2) the indirect trust metric model of Mobile banking client is set up, specific as follows:

$R_n(i,j)=\frac{1}{n}\left(\underset{k=1}{\overset{m}{\Σ}}{D}_n\left(i,j\right)\·{\mathrm{IC}}_j\right)$

Wherein, R_n(i, j) refers to the indirect degree of belief between cellphone subscriber i and Mobile banking j in the n-th business hours, IC_jBe in the kth time financial payment got by the indirect trust table of kth time financial payment Mobile banking j to cellphone subscriber i The confidence level indirectly trusted；

(3) according to the direct degree of belief between cellphone subscriber and Mobile banking and indirect degree of belief, Mobile banking's gold is set up Melt business trust metric model, specific as follows:

PT_n(i, j)=α D_n(i,j)+(1-α)R_n(i,j),α∈[0,1]；

Wherein, PT_n(i, j) refers to cellphone subscriber i and the financial business degree of belief in Mobile banking j n-th service interaction, α is degree of belief regulatory factor；

(4) according to trusting offset relation between cellphone subscriber and Mobile banking, set up Mobile banking and trust departure degree mould Type, specific as follows:

$P_{ij}=\sqrt{\left(\underset{k=1}{\overset{\max TZ}{\Σ}}\left\{g_k{\left(D_k\left(i,j\right)-R_k\left(i,j\right)\right)}^2\right\}\right)/\underset{k=1}{\overset{\max TZ}{\Σ}}{g}_k};$

Wherein, P_ijReferring to the trust deviation value between cellphone subscriber i and Mobile banking j, max TZ is for calculating trust partially The maximum time siding-to-siding block length of difference, its upper limit is the whole mobile banking service period；D_k(i j) refers to kth time financial payment Direct degree of belief, g_kRefer to the trust pad value of kth time financial payment, R_k(i j) refers to the indirect letter of kth time financial payment Ren Du；

(5) set up Mobile banking's customer trust penalty value model, be expressed as follows:

PT_ij=β PT_ij–(1-β)(μP_ij+vQ_ij)

Wherein, β refers to penalty value coefficient, 0 < β < 1；

Q_ijFor trusting abuse value,

Wherein, u increases Studying factors for trusting, and v reduces Studying factors, u < v for trusting, and the speed ratio i.e. trusting reduction increases The speed added is fast；P_ijShow to trust departure degree, (μ P_ij+vQ_ij) it is the Mobile banking j penalty value to cellphone subscriber i；Use hand Machine bank client trusts penalty value model, and the Mobile banking's financial business to behavior upheaval change plays punishment effect；

(6) according to the recent trust value between cellphone subscriber i and Mobile banking j and long-term trust value, obtain degree of belief and comment Valency result；Trust evaluation result T_n(i j) takes minimum of a value T in recent trust value and long-term both trust values_n+1(i j), represents As follows:

T_n+1(i, j)=min (ST_n+1(i,j),LT_n+1(i,j))；

Wherein, LT_n+1(i j) refers to the long-term trust value after cellphone subscriber i and Mobile banking's j (n+1) secondary business；

LT_n+1(i, j)=(LT_n(i,j)n+PT_n+1(i,j))/(n+1)；

Wherein, ST_n+1(i j) refers to the recent trust value after cellphone subscriber i and Mobile banking's j (n+1) secondary business；

${\mathrm{ST}}_{n+1}\left(i,j\right)=\left\{\begin{array}{cc}\left(1-u\right){\mathrm{ST}}_n\left(i,j\right)+{\mathrm{uPT}}_{n+1}\left(i,j\right),& {\mathrm{PT}}_{n+1}\left(i,j\right)-{\mathrm{ST}}_n\left(i,j\right)\&GreaterEqual;-\mathrm{\&epsiv;}\\ \left(1-v\right){\mathrm{ST}}_n\left(i,j\right)+{\mathrm{vPT}}_{n+1}\left(i,j\right),& otherwise\end{array}\right.$

Wherein, ε is to trust threshold values in the recent period, 0 < ε < 1.

Preferably, in Mobile banking's customer trust degree evaluation method of above-mentioned various dimensions, cellphone subscriber i and Mobile banking j Between carry out financial payment trust propensity value f (i, j) according to below equation obtain:

F (i, j)=a₁*v₁+a₂*v₂+a₃*v₃+a₄*v₄+a₅*v₅+a₆*v₆；

Wherein, a₁、a₂、a₃、a₄、a₅、a₆It is adjustment factor, and a₁+a₂+a₃+a₄+a₅+a₆=1；

Wherein, v₁It is direct degree of belief, v₂It is indirect degree of belief, v₃It is financial business degree of belief, v₄It is to trust irrelevance, v₅ It is to trust penalty value, v₆It it is recent trust value.

Preferably, Mobile banking's customer trust degree evaluation method of above-mentioned various dimensions, also include the step building trust vector Suddenly, specific as follows:

(7) according to above-mentioned degree of belief evaluation result T_n+1(i j), divides and carries out finance between cellphone subscriber i and Mobile banking j The reliability rating paid；

(8) according to degree of belief membership function being subordinate at each reliability rating of main body between Mobile banking and cellphone subscriber The vector that degree is constituted, obtains trust vector；

Trust vector X is used to represent the degree of belief between cellphone subscriber i and Mobile banking j, wherein, X={T_n(i,j)}； By the form of trust vector, evaluate Mobile banking's reliability rating of various dimensions.

In general, by the contemplated above technical scheme of the present invention compared with prior art, it is possible to show under acquirement Benefit effect:

(1) Mobile banking's customer trust degree evaluation method of the various dimensions that the present invention provides, establishes a kind of multidimensional and trusts Mobile banking's financial risks assessment indicator system；Can be used for evaluating the trusting relationship between Mobile banking and cellphone subscriber, tool There are enhancing user and Mobile banking's beneficial effect to its service interaction degree of belief；Mobile phone bank-user is had with Mobile banking The strongest reference significance；

(2) Mobile banking's customer trust degree evaluation method of the various dimensions that the present invention provides, by Mobile banking's payment frequency With the relation trusted, the factors such as Mobile banking trusts departure degree, customer trust penalty value are all brought in reliability rating evaluation； For compared with prior art, evaluating more comprehensively, the evaluation result therefore obtained is more accurate；

(3) Mobile banking's customer trust degree evaluation method of the various dimensions that the present invention provides, it is provided that Mobile banking client Trusting penalty value model, Mobile banking's financial business that this model can be used to change behavior upheaval plays punishment effect；

(4) Mobile banking's customer trust degree evaluation method of the various dimensions that the present invention provides, processes in conjunction with big data, permissible Set up Mobile banking's customer trust rating database that a kind of multidimensional is trusted, preferably cellphone subscriber is carried out trust evaluation, its Evaluation result can be used for reference as Credit Rank Appraisal；Cellphone subscriber and mobile banking service provider all can examine according to this result Careful finance interactive service for the treatment of, and then reduction payment risk.

Accompanying drawing explanation

Fig. 1 is the flow process signal of Mobile banking's customer trust degree evaluation method of the various dimensions that the embodiment of the present invention provides Figure.

Detailed description of the invention

In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, right The present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, and It is not used in the restriction present invention.If additionally, technical characteristic involved in each embodiment of invention described below The conflict of not constituting each other just can be mutually combined.

Mobile banking's customer trust degree evaluation method of the various dimensions that the embodiment of the present invention provides, its flow process is as shown in Figure 1 , specifically include following steps:

(1) according to the financial payment number of times in business hours section, the direct trust metric model of Mobile banking client is set up, by hand Direct degree of belief D between machine user i and Mobile banking j_n(i, j) is specifically expressed as follows:

$D_n(i,j)=\left\{\begin{array}{c}\frac{\underset{k=1}{\overset{m}{\&Sigma;}}f(i,j)}{m},\\ 0,m=0.\end{array}\right.m\&NotEqual;0;$

Wherein, m represents in n-th business, carries out the total degree of financial payment between cellphone subscriber i and Mobile banking j, K is financial payment number of times numbering；(i j) is the trust propensity value between cellphone subscriber i and Mobile banking j to f；

In embodiment, f (i, j) obtains according to below equation:

F (i, j)=a₁*v₁+a₂*v₂+a₃*v₃+a₄*v₄+a₅*v₅+a₆*v₆；

Wherein, a₁、a₂、a₃、a₄、a₅、a₆It is adjustment factor, and a₁+a₂+a₃+a₄+a₅+a₆=1；

Wherein, v₁Refer to direct degree of belief, v₂Refer to indirect degree of belief, v₃Refer to financial business degree of belief, v₄Refer to trust Irrelevance, v₅Refer to trust penalty value, v₆Refer to recent trust value；0<v₁<1,0<v₂<1,0<v₃<1,0<v₄<1,0<v₅<1,0<v₆ <1；

(2) the indirect trust metric model of Mobile banking client is set up, by the indirect letter between cellphone subscriber i and Mobile banking j Appoint degree R_n(i, j) is specifically expressed as follows:

$R_n(i,j)=\frac{1}{n}\left(\underset{k=1}{\overset{m}{\&Sigma;}}{D}_n\left(i,j\right)\&CenterDot;{\mathrm{IC}}_j\right)$

Wherein, D_n(i, is j) directly to trust angle value between cellphone subscriber i and Mobile banking j in the n-th business hours, IC_jBe in the kth time financial payment got by the indirect trust table of kth time financial payment Mobile banking j to cellphone subscriber i The confidence level indirectly trusted；

(3) according to the direct degree of belief between cellphone subscriber and Mobile banking and indirect degree of belief, Mobile banking's gold is set up Melt business trust metric model, be expressed as follows:

PT_n(i, j)=α D_n(i,j)+(1-α)R_n(i,j),α∈[0,1]；

Wherein, PT_n(i j) refers to cellphone subscriber i and the finance in Mobile banking j n-th service interaction in n-th business Business degree of belief, α is degree of belief regulatory factor；In embodiment, 0≤α≤1；

(4) according to trusting offset relation between cellphone subscriber and Mobile banking, set up Mobile banking and trust departure degree mould Type；By the trust deviation value P between cellphone subscriber i and Mobile banking j_ijIt is expressed as follows:

$P_{ij}=\sqrt{\left(\underset{k=1}{\overset{\max TZ}{\&Sigma;}}\left\{g_k{\left(D_k\left(i,j\right)-R_k\left(i,j\right)\right)}^2\right\}\right)/\underset{k=1}{\overset{\max TZ}{\&Sigma;}}{g}_k};$

Wherein, max TZ is the maximum time siding-to-siding block length that need to calculate and trust deviation value, and its upper limit is whole Mobile banking The business period；D_k(i j) refers to the direct degree of belief of kth time financial payment, g_kRefer to the trust pad value of kth time financial payment, R_k(i j) refers to the indirect degree of belief of kth time financial payment；

(5) set up Mobile banking's customer trust penalty value model, be expressed as follows:

PT_ij=β PT_ij–(1-β)(μP_ij+vQ_ij)

Wherein, β refers to penalty value coefficient；In embodiment, 0 < β < 1；

Q_ijFor trusting abuse value,

Wherein, u increases Studying factors for trusting, and v reduces Studying factors, u < v for trusting, and the speed ratio i.e. trusting reduction increases The speed added is fast；P_ijShow to trust departure degree, (μ P_ij+vQ_ij) it is the Mobile banking j penalty value to cellphone subscriber i；

(6) according to the recent trust value between cellphone subscriber i and Mobile banking j and long-term trust value, Mobile banking is set up Customer trust evaluation model, by degree of belief evaluation result T_n+1(i, j) is expressed as follows:

T_n+1(i, j)=min (ST_n+1(i,j),LT_n+1(i,j))；

Wherein, LT_n+1(i j) refers to the long-term trust value after cellphone subscriber i and Mobile banking's j (n+1) secondary business；

LT_n+1(i, j)=(LT_n(i,j)n+PT_n+1(i,j))/(n+1)；

Wherein, ST_n+1(i j) refers to the recent trust value after cellphone subscriber i and Mobile banking's j (n+1) secondary business；

${\mathrm{ST}}_{n+1}\left(i,j\right)=\left\{\begin{array}{cc}\left(1-u\right){\mathrm{ST}}_n\left(i,j\right)+{\mathrm{uPT}}_{n+1}\left(i,j\right),& {\mathrm{PT}}_{n+1}\left(i,j\right)-{\mathrm{ST}}_n\left(i,j\right)\&GreaterEqual;-\mathrm{\&epsiv;}\\ \left(1-v\right){\mathrm{ST}}_n\left(i,j\right)+{\mathrm{vPT}}_{n+1}\left(i,j\right),& otherwise\end{array}\right.$

Wherein, ε is to trust threshold values in the recent period；In embodiment, 0 < ε < 1；

(7) according to above-mentioned degree of belief evaluation result T_n+1(i j), divides and carries out finance between cellphone subscriber i and Mobile banking j The reliability rating paid；

(8) according to degree of belief membership function being subordinate at each reliability rating of main body between Mobile banking and cellphone subscriber The vector that degree is constituted, obtains trust vector；

Trust vector X is used to represent the degree of belief between cellphone subscriber i and Mobile banking j, wherein, X={T_n(i,j)}； By the form of trust vector, evaluate Mobile banking's reliability rating of various dimensions.

As it will be easily appreciated by one skilled in the art that and the foregoing is only presently preferred embodiments of the present invention, not in order to Limit the present invention, all any amendment, equivalent and improvement etc. made within the spirit and principles in the present invention, all should comprise Within protection scope of the present invention.

Claims (3)

1. Mobile banking's customer trust degree evaluation method of various dimensions, it is characterised in that comprise the steps:

(1) according to the financial payment number of times in business hours section, the direct trust metric model of Mobile banking client is set up, the most such as Under:

$D_n(i,j)=\left\{\begin{array}{c}\frac{\underset{k=1}{\overset{m}{\&Sigma;}}f(i,j)}{m},m\&NotEqual;0;\\ 0,m=0.\end{array}\right.$

Wherein, D_n(i j) is the n-th business hours interior direct degree of belief referred between cellphone subscriber i and Mobile banking j；M refers to In n-th business, between cellphone subscriber i and Mobile banking j, carry out the total degree of financial payment；K is that financial payment number of times is compiled Number；(i j) refers to the trust propensity value between cellphone subscriber i and Mobile banking j to f；

(2) the indirect trust metric model of Mobile banking client is set up, specific as follows:

$R_n(i,j)=\frac{1}{n}\left(\underset{k=1}{\overset{m}{\mathrm{\&Sigma;}}}{D}_n\right(i,j)\&CenterDot;{\mathrm{IC}}_j)$

Wherein, R_n(i j) refers to the indirect degree of belief between cellphone subscriber i and Mobile banking j, IC_jIt is by kth time finance The Mobile banking j confidence level indirectly trusted to cellphone subscriber i in the kth time financial payment that the indirect trust table paid gets；

(3) according to the direct degree of belief between described cellphone subscriber and Mobile banking and indirect degree of belief, Mobile banking's gold is set up Melt business trust metric model, specific as follows:

PT_n(i, j)=α D_n(i,j)+(1-α)R_n(i,j),α∈[0,1]；

Wherein, PT_n(i, j) refers to cellphone subscriber i and the financial business degree of belief in Mobile banking j n-th service interaction, and α is letter Appoint degree regulatory factor；

(4) according to trusting offset relation between cellphone subscriber and Mobile banking, set up Mobile banking and trust departure degree model, tool Body is as follows:

$P_{ij}=\sqrt{\left(\underset{k=1}{\overset{\max TZ}{\mathrm{\&Sigma;}}}\left\{g_k{\left(D_k\left(i,j\right)-R_k\left(i,j\right)\right)}^2\right\}\right)/\underset{k=1}{\overset{\max TZ}{\mathrm{\&Sigma;}}}{g}_k};$

Wherein, P_ijReferring to the trust deviation value between cellphone subscriber i and Mobile banking j, max TZ is for calculating trust deviation value Maximum time siding-to-siding block length；D_k(i j) refers to the direct degree of belief of kth time financial payment, g_kRefer to kth time financial payment Trust pad value, R_k(i j) refers to the indirect degree of belief of kth time financial payment；

(5) Mobile banking's customer trust penalty value model is set up, specific as follows:

PT_ij=β PT_ij–(1-β)(μP_ij+vQ_ij)

Wherein, β refers to penalty value coefficient, 0 < β < 1；

Q_ijFor trusting abuse value,

Wherein, u increases Studying factors for trusting, and v reduces Studying factors, u < v for trusting；(μP_ij+vQ_ij) it is Mobile banking j opponent The penalty value of machine user i；

(6) according to the recent trust value between cellphone subscriber i and Mobile banking j and long-term trust value, degree of belief evaluation knot is obtained Really T_n+1(i, j), is expressed as follows:

T_n+1(i, j)=min (ST_n+1(i,j),LT_n+1(i,j))；

Wherein, LT_n+1(i j) refers to the long-term trust value after cellphone subscriber i and Mobile banking's j (n+1) secondary business；

LT_n+1(i, j)=(LT_n(i,j)n+PT_n+1(i,j))/(n+1)；

Wherein, ST_n+1(i j) refers to the recent trust value after cellphone subscriber i and Mobile banking's j (n+1) secondary business；

${\mathrm{ST}}_{n+1}\left(i,j\right)=\left\{\begin{array}{cc}\left(1-u\right){\mathrm{ST}}_n\left(i,j\right)+{\mathrm{uPT}}_{n+1}\left(i,j\right),& {\mathrm{PT}}_{n+1}\left(i,j\right)-{\mathrm{ST}}_n\left(i,j\right)\&GreaterEqual;-\mathrm{\&epsiv;}\\ \left(1-v\right){\mathrm{ST}}_n\left(i,j\right)+{\mathrm{vPT}}_{n+1}\left(i,j\right),& otherwise\end{array}\right.$

Wherein, ε is to trust threshold values in the recent period, 0 < ε < 1.

2. Mobile banking as claimed in claim 1 customer trust degree evaluation method, it is characterised in that after described step (6), Also include the step building trust vector, specific as follows:

(7) according to described degree of belief evaluation result T_n+1(i j), divides and carries out financial payment between cellphone subscriber i and Mobile banking j Reliability rating；

(8) according to the degree of membership institute at each reliability rating of the degree of belief membership function of main body between Mobile banking and cellphone subscriber The vector constituted, obtains trust vector；

Trust vector X is used to represent the degree of belief between cellphone subscriber i and Mobile banking j, wherein, X={T_n(i,j)}；By letter Appoint the form of vector, evaluate Mobile banking's reliability rating of various dimensions.

3. Mobile banking as claimed in claim 1 or 2 customer trust degree evaluation method, it is characterised in that described cellphone subscriber i And carry out between Mobile banking j financial payment trust propensity value f (i, j) according to below equation obtain:

F (i, j)=a₁*v₁+a₂*v₂+a₃*v₃+a₄*v₄+a₅*v₅+a₆*v₆；

Wherein, a₁、a₂、a₃、a₄、a₅、a₆It is adjustment factor, a₁+a₂+a₃+a₄+a₅+a₆=1；v₁It is direct degree of belief, v₂It is indirect Degree of belief, v₃It is financial business degree of belief, v₄It is to trust irrelevance, v₅It is to trust penalty value, v₆It it is recent trust value.