Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide one kind can shorten air control is time-consuming, reduce data call atThis transaction data processing method, device, computer equipment and storage medium.
A kind of transaction data processing method, which comprises
Obtain trading activity to be discriminated；
Determine decision in the face of risk tree corresponding with the transaction categories of the trading activity；
First first kind decision index system is extracted from the decision in the face of risk tree, and obtains the first kind from the first sourceCorresponding first decision data of decision index system；
Run the first decision rule of the decision in the face of risk tree, the first decision rule to first decision data intoRow verification；
If the first decision rule miss is extracted posterior second class decision from the decision in the face of risk tree and is referred toMark, and corresponding second decision data of the second class decision index system is obtained from second source；
Run the decision in the face of risk tree in rear decision rule, it is described rear decision rule to second decision data intoRow verification obtains differentiating result.
In one embodiment, before acquisition trading activity to be discriminated, further includes:
It receives decision in the face of risk tree and generates request, carry transaction categories in the request；
The corresponding decision index system of the transaction categories is obtained, the decision index system includes first kind decision index system and the second classDecision index system；
The decision index system is sent to terminal, so that the terminal shows configuration circle for being associated with the decision index systemFace；
The rule configuration information that the configuration interface uploads is received, at least one institute is generated according to the rule configuration informationState the corresponding decision in the face of risk tree of transaction categories.
In one embodiment, the method also includes:
The corresponding multiple decision in the face of risk trees of the transaction categories carry out risk anticipation to positive negative sample, obtain multiple institutesState the corresponding risk anticipation result of decision in the face of risk tree；
Calculate the matching degree of the positive and negative attribute of the risk anticipation result and the positive negative sample；
Determine that the corresponding target risk of the transaction categories is determined from multiple decision in the face of risk trees according to the matching degreePlan tree.
In one embodiment, described to determine the transaction class from multiple decision in the face of risk trees according to the matching degreeNot corresponding target risk decision tree, comprising:
The corresponding first arrangement accounting of multiple decision in the face of risk trees is calculated, the first arrangement accounting is the first kindThe first arrangement specific gravity of the corresponding decision rule of decision index system；
The transaction class is determined from multiple decision in the face of risk trees according to the matching degree and the first arrangement accountingNot corresponding target risk decision tree；
Wherein, the decision rule in first decision rule comprising the second class decision index system is fewer, and second classThe decision rule of decision index system sorts more rearward in the first decision rule, and the first arrangement accounting is higher.
In one embodiment, the method also includes:
If the first decision rule hits any first decision data, the decision in the face of risk tree output differentiates knotFruit, Exit Decision-making engine.
A kind of transaction data processing unit, described device include:
Trading activity obtains module, for obtaining trading activity to be discriminated；
Decision in the face of risk tree determining module, for determining decision in the face of risk tree corresponding with the transaction categories of the trading activity；
First decision data obtains module, for extracting first first kind decision index system from the decision in the face of risk tree,And corresponding first decision data of the first kind decision index system is obtained from the first source；
First correction verification module, for running the first decision rule of the decision in the face of risk tree, the first decision rule pairFirst decision data is verified；
Second decision data obtains module, if the first decision rule miss is used for, from the decision in the face of risk treeIt is middle to extract posterior second class decision index system, and the corresponding second decision number of the second class decision index system is obtained from second sourceAccording to；
Second correction verification module, for run the decision in the face of risk tree in rear decision rule, it is described in rear decision rule pairSecond decision data is verified, and obtains differentiating result.
In one embodiment, described device further include:
Decision in the face of risk tree generation module generates request for receiving decision in the face of risk tree, carries transaction categories in the request；The corresponding decision index system of the transaction categories is obtained, the decision index system includes that first kind decision index system and the second class decision refer toMark；The decision index system is sent to terminal, so that the terminal shows the configuration interface for being associated with the decision index system；It receivesThe rule configuration information that the configuration interface uploads, generates at least one described transaction categories pair according to the rule configuration informationThe decision in the face of risk tree answered.
In one embodiment, described device further include:
Decision in the face of risk tree screening module, for the corresponding multiple decision in the face of risk trees of the transaction categories to positive negative sampleRisk anticipation is carried out, the corresponding risk anticipation result of multiple decision in the face of risk trees is obtained；Calculate risk anticipation result withThe matching degree of the positive and negative attribute of the positive negative sample；The friendship is determined from multiple decision in the face of risk trees according to the matching degreeThe corresponding target risk decision tree of easy classification.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processingThe step of device realizes method described above when executing the computer program.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processorThe step of method described above is realized when row.
Above-mentioned transaction data processing method, device, computer equipment and storage medium, the decision rule in decision in the face of risk treeIt is the decision rule of corresponding decision index system, decision rule verifies the corresponding decision data of decision index system, according to decision numberAccording to separate sources, decision index system is divided into first kind decision index system and the second class decision index system, when carrying out decision in the face of risk,The corresponding decision data of first kind decision index system only is obtained from the first source, decision in the face of risk tree only determines to first kind decision index systemPlan data are verified, only when decision data does not just obtain the second class decision from second source by first decision rule hitThe corresponding decision data of index avoids unnecessary decision data and obtains, and reduces data call cost, what hit was exitedDecision mode improves trading activity identification effect.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understoodThe application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, notFor limiting the application.
Transaction data processing method provided by the present application can be applied in application environment as shown in Figure 1.Wherein, eventuallyEnd 102 is communicated with server 104 by network by network.Terminal triggers trading activity, and transaction request is sent toServer, server carry out risk differentiation to transaction request by the corresponding decision in the face of risk tree of transaction predetermined, obtain windDanger differentiates result.Wherein, terminal 102 can be, but not limited to be various personal computers, laptop, smart phone, plateComputer and portable wearable device, server 104 can use the service of the either multiple server compositions of independent serverDevice cluster is realized.
In one embodiment, as shown in Fig. 2, providing a kind of transaction data processing method, it is applied to Fig. 1 in this wayIn server for be illustrated, comprising the following steps:
Step 202, trading activity to be discriminated is obtained, decision in the face of risk tree corresponding with the transaction categories of trading activity is searched.
Trading activity includes the financial transaction behavior of user terminal triggering, as loan transaction, personal acquisition of debenture, stock,Enterprise investment etc..In the present embodiment, mainly differentiate financial transaction behavior with the presence or absence of transaction the differentiation of financial transaction behaviorRisk, such as the various risk cases of transaction swindling.
Multiple transaction categories are marked off to each financial transaction behavior in transaction platform in advance.Multiple transaction categories are distinguishedTrading object, trade mode.It is that individual, enterprise, government are divided into not by trading object so that financial transaction behavior is loan as an exampleIt is generic, financial instrument is exchanged into financial instrument, financial instrument Exchange Service, financial instrument exchange material object and is divided into different classesNot.
Pre-define the corresponding decision tree of each transaction categories.Multiple decision rules are mutually connected according to the decision logic of settingIt connects and constitutes decision tree.Each decision rule includes the decision condition of decision index system, decision index system, each decision condition in decision treeCorresponding result of decision item.
In the present embodiment, server obtains trading activity to be discriminated, determines the trading activity according to the classification of trading activityDecision in the face of risk tree.Wherein, a decision in the face of risk tree can be determined for trading activity according to trading activity classification, can also determinedMultiple corresponding decision in the face of risk trees out.
Step 204, first first kind decision index system is extracted from decision in the face of risk tree, and obtains the first kind from the first sourceCorresponding first decision data of decision index system.
Decision in the face of risk tree is made of the decision rule being connected from top to bottom, and the corresponding decision of each decision rule refers toMark.In the present embodiment, decision index system includes first kind decision index system and the second class decision index system, wherein first kind decision index systemCorresponding decision rule is located at decision in the face of risk tree in the upstream of decision in the face of risk tree, the corresponding decision rule of the second class decision index systemDownstream.That is, first kind decision index system is first decision index system, the second class decision index system is in rear decision index system.
Server extracts first first kind decision index system or server from decision in the face of risk tree and obtains decision in the face of risk treeAll decision index systems of corresponding ordered arrays, first first kind the sequence of decision index is preceding.Server obtains first theCorresponding first decision data of a kind of decision index system.
As shown in figure 3, decision index system 1- decision index system 3 is first first kind decision index system, decision index system 1 is obtained firstCorresponding decision data, the corresponding decision data of decision index system 2 and the corresponding decision data of decision index system 3.First kind decisionIndex can be name, and the corresponding decision data of name is the name of instantiation, such as Zhang San.First kind decision index system can also beAge, age corresponding decision data are the age of instantiation, such as 35 years old.
First decision data source of first kind decision index system is identical, the second decision data source of the second class decision index systemIt is identical.That is, server obtains corresponding first decision data of first first kind decision index system from the first source.?In one embodiment, the first source is internal data, and second source is the external data of external platform.First source and secondSource can also be that direct sources and secondary source, direct sources refer to that decision data can be extracted directly from trading activity dataIt arrives, secondary source refers to needs from external data or needs to do what further data operation can just access, such as needs to adjustWith just getable data are secondary source after interface operation.
Step 206, first decision rule in operation risk decision tree, first decision rule carry out school to the first decision dataIt tests.
First decision rule refers to the first decision rule that sorts in generating decision tree, can be according in decision in the face of risk tree theThe quantity of one decision index system and the second decision index system decision rule more first than determination and the cut-point in rear decision rule.This implementationIn example, first decision rule is the rule of corresponding first decision index system, if decision in the face of risk tree includes 10 decision rules, wherein havingCorresponding first decision index system of 7 decision rules, then first decision rule, which refers to, is arranged in preceding 7 decision rules for being.
It should be noted that may formerly have corresponding second decision index system of a small amount of rule in decision rule, even if correspondingSecond decision index system, as long as it is arranged in front of first posterior cut-point, decision index system is exactly first decision rule, is being dividedIt is exactly in rear decision rule after point.
Decision engine in server is according to the decision rule in top-down sequence one by one operational decisions tree, prerequisitePlan rule verifies the first decision data, and any decision data is hit by corresponding first decision rule, then exports the result of decision, move backDecision engine out.
Specifically, decision rule the first in operation risk decision tree, obtains the first corresponding decision of decision rule and refers toThe first corresponding decision data of mark, decision rule verifies first decision data, if decision rule miss, operationNext decision rule obtains corresponding first decision data of next decision rule, until first first kind decision index systemCorresponding decision rule verifies completion, as in Fig. 3 until in the corresponding decision rule 3 of the last one first kind decision index systemComplete verification.
When in first decision rule including the decision rule of corresponding second class decision index system, in step 204, it is also necessary to mentionThe first corresponding decision data of the second class decision index system is taken, and it is verified by corresponding first decision rule.WhenIt then also include corresponding first in rear decision rule when including the decision rule of corresponding second class decision index system in first decision ruleThe decision rule of decision index system, in step 204, the decision data in the first kind decision index system of acquisition includes the posterior first kindThe corresponding decision data of decision index system.That is, what is obtained in step 204 is first kind decision all in decision in the face of risk treeThe corresponding decision data of index.
Step 208, if first decision rule miss, posterior second class decision is extracted from decision in the face of risk tree and is referred toMark, and obtain corresponding second decision data of the second class decision index system.
If the first corresponding decision rule of first kind decision index system has verified, regular hit not yet.Server obtainsPosterior second class decision index system in decision in the face of risk tree, and by obtaining the in the corresponding second source of the second class decision index systemCorresponding second decision data of two class decision index systems.
Step 210, posterior decision rule in operation risk decision tree, decision rule carry out school to the second decision dataIt tests, obtains differentiating result.
Posterior decision rule in decision engine operation risk decision tree, as run the decision rule 4 in Fig. 3, rearDecision rule verify corresponding second decision data, until rule hit, output differentiates as a result, Exit Decision-making engine, orUntil the second all decision rules has verified, if hitting still without rule, differentiation result is exported.
The differentiation result of above-mentioned rule hit is " trading activity to be discriminated is risk trade behavior ", regular complete verificationThe differentiation result of complete still miss is " trading activity to be discriminated is arm's length dealing behavior ".Server will be determined as risk trade rowFor transaction request be assigned to risk processing terminal, carry out risk request processing.The transaction for being determined as arm's length dealing behavior is askedIt asks and is back to service process terminal, carry out normal service request processing.
In the present embodiment, the decision rule in decision in the face of risk tree is the decision rule of corresponding decision index system, decision rule pairThe corresponding decision data of decision index system is verified, and according to the separate sources of decision data, decision index system is divided into the first kindDecision index system and the second class decision index system are that the corresponding decision rule of first kind decision index system and the second class are determined according to data sourceThe corresponding decision rule of plan index defines different priority, and the corresponding decision rule of first kind decision index system has higher priorGrade, the upstream configured in decision in the face of risk tree, correspondingly, the corresponding decision rule configuration of the second class decision index system is determined in riskThe downstream of plan tree.When carrying out decision in the face of risk, the corresponding decision data of first kind decision index system is only obtained, decision in the face of risk tree is only rightThe decision data of first kind decision index system is verified, only when decision data is by first decision rule hit just acquisition theThe corresponding decision data of two class decision index systems avoids unnecessary decision data and obtains, reduce data call cost, hitsThe decision mode exited improves trading activity identification effect.
In one embodiment, as shown in figure 4, providing a kind of decision in the face of risk tree generation method, this method is applied in Fig. 1In server, specifically comprise the following steps:
Step 402, it receives decision in the face of risk tree and generates request, transaction categories are carried in request, it is corresponding to obtain transaction categoriesDecision index system, decision index system include first kind decision index system and the second class decision index system.
The decision in the face of risk tree that server receiving terminal is sent generates request, and according to the transaction categories in request, searches pre-The decision index system being first associated with transaction categories.Each associated decision index system of transaction categories includes first kind decision index system andTwo class decision index systems.In the present embodiment, set first kind decision index system as non-external decision index system, be meant that, it is non-it is external certainlyThe corresponding decision data of plan index can be got from itself platform data；The second class decision index system is set to refer to as external decisionMark, is meant that, the corresponding decision data of external decision index system is needed by calling external platform data that can get.
Step 404, decision index system is sent to terminal, terminal shows the configuration interface for being associated with decision index system, and reception is matchedThe rule configuration information for setting interface upload, generates the corresponding decision in the face of risk of at least one transaction categories according to rule configuration informationTree.
Decision index system corresponding with transaction categories is sent to terminal by server, and terminal is based on the decision index system to for matchingThe configuration interface for setting decision in the face of risk tree is updated configuration, i.e., is associated with decision index system under the corresponding input frame of configuration interfaceIt draws in list.Configuration decision in the face of risk tree is can to choose decision index system by drop-down list, and match to the decision index system of selectionIt sets, server generates decision rule to the configuration information of decision index system.
Server generates multiple decision rules, by configuring boundary by the configuration information of the decision index system of configuration interfaceRule connection configuration information in face, connects multiple decision rules and generates decision in the face of risk tree.Terminal can be configured by configuration interfaceMultiple rule connects configuration information, generates multiple decision in the face of risk trees based on multiple rule connection configuration information.It can also be serviceDevice automatically generates multiple decision in the face of risk trees using different connection types.
Step 406, the corresponding multiple decision in the face of risk trees of transaction categories carry out risk anticipation to positive negative sample, obtain multiple windThe corresponding risk anticipation of dangerous decision tree as a result, the positive and negative attribute of calculation risk anticipation result and positive negative sample matching degree.
After generating multiple decision in the face of risk trees, server tests the decision precision of multiple decision in the face of risk trees.Specifically, historical transactional information all under transaction categories is obtained, the Transaction Information that arm's length dealing is chosen from historical trading data is madeFor positive sample, the transaction request being rejected is as negative sample.
When verification, decision index system is extracted from decision in the face of risk tree to be tested first, each sample is then collected and corresponds to thisThe verification data of a little decision index systems, the verification data of sample are input in decision in the face of risk tree and obtain decision in the face of risk result.RiskPrejudging result includes risk sample and normal sample.
After each decision in the face of risk tree to be tested is obtained to the decision in the face of risk result of each sample, the risk of each sample is determinedPlan result is matched with the positive and negative attribute of sample itself.Wherein, risk sample matches with negative sample attribute, normal sample withPositive sample attribute phase configuration.If sample is positive sample, and decision in the face of risk tree is risk sample to the risk anticipation result of the sample,Then the positive and negative attribute of risk anticipation result and sample does not match that, calculates matching degree according to the accounting for the sample set that mismatches.NoThe accounting of the sample to match is bigger, and the matching degree of the positive and negative attribute of the result of decision and sample that decision in the face of risk tree is made is smaller,The decision accuracy of decision in the face of risk tree is smaller.
Step 408, the corresponding target risk decision tree of transaction categories is determined from multiple decision in the face of risk trees according to matching degree.
According to step 406, the matching degree of the corresponding multiple decision in the face of risk trees of transaction categories is calculated, by corresponding maximum matching degreeDecision in the face of risk tree as the corresponding target risk decision tree of transaction categories.
In the present embodiment, a variety of decision index systems are defined for every kind of transaction categories, the corresponding decision rule of decision index system is matchedInterface is set, decision rule can be realized by configuration interface flexibly and fast to be configured and adjust, and rule is disposed and property is improved.TogetherSample, the connection type of multiple decision rules is configured by configuration interface, and multiple decisions in the face of risk are then generated according to connection typeTree, realizes the various decision in the face of risk trees of generation according to demand flexibly and fast.
In addition, also being screened to multiple decision in the face of risk trees of generation, in the present embodiment to determine decision accuracy mostHigh decision in the face of risk tree realizes the screening of optimal decision in the face of risk tree as the corresponding decision in the face of risk tree of transaction categories.
Further, when screening decision in the face of risk tree, the first arrangement for also calculating multiple decision in the face of risk trees to be screened is accounted forThan what this formerly arranged accounting characterization is that the corresponding decision rule of first kind decision index system (non-external decision rule) is determined in riskFirst arrangement situation in plan tree, accounting of formerly arranging is bigger, and the preceding quantity of external decision rule arrangement is fewer, and is arranged inPreceding external decision rule sequence is more rearward.When formerly arrangement accounting is 1, meaning is all non-external regular arrays in instituteBefore having external rule.
In the first arrangement accounting of the non-external decision rule of calculation risk decision tree, decision in the face of risk tree can be divided intoA part and second part, wherein the regular quantity of first part is the quantity of non-external decision index system, the rule of second partQuantity is the quantity of external decision index system, and the quantity comprising the corresponding decision rule of external decision index system is fewer in first part,External decision index system first part sequence more rearward, non-external decision index system specific gravity of formerly arranging is higher.
In the present embodiment, the preferably non-first arrangement higher decision in the face of risk tree of accounting of external rule is based on preferred riskWhen decision tree carries out trading activity differentiation, can preferentially the relevant non-external data of checkout transaction behavior, non-external data it obtainIt is more convenient, only when non-external data miss, the relevant external data of trading activity is just called to do further verification, untilIt obtains differentiating result.Therefore, formerly arrangement accounting is higher for external rule, and data call cost is lower when trading activity differentiates.
It should be understood that although each step in the flow chart of Fig. 2 and Fig. 4 is successively shown according to the instruction of arrow,But these steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly state otherwise herein, theseThere is no stringent sequences to limit for the execution of step, these steps can execute in other order.Moreover, in Fig. 2 and Fig. 4At least part step may include that perhaps these sub-steps of multiple stages or stage are not necessarily same to multiple sub-stepsOne moment executed completion, but can execute at different times, and the execution in these sub-steps or stage sequence is also not necessarilyBe successively carry out, but can at least part of the sub-step or stage of other steps or other steps in turn orAlternately execute.
In one embodiment, as shown in figure 5, providing a kind of transaction data processing unit, comprising:
Trading activity obtains module 502, for obtaining trading activity to be discriminated.
Decision in the face of risk tree determining module 504, for searching decision in the face of risk corresponding with the transaction categories of the trading activityTree.
First decision data obtains module 506, refers to for extracting first first kind decision from the decision in the face of risk treeMark, and corresponding first decision data of the first kind decision index system is obtained from the first source.
First correction verification module 508, for running the first decision rule of the decision in the face of risk tree, the first decision ruleFirst decision data is verified.
Second decision data obtains module 510, if the first decision rule miss is used for, from the decision in the face of riskPosterior second class decision index system is extracted in tree, and obtains corresponding second decision of the second class decision index system from second sourceData.
Second correction verification module 512, for run the decision in the face of risk tree in rear decision rule, it is described in rear decision ruleSecond decision data is verified, obtains differentiating result.
In one embodiment, transaction data processing unit further include:
Decision in the face of risk tree generation module generates request for receiving decision in the face of risk tree, carries transaction categories in the request；The corresponding decision index system of the transaction categories is obtained, the decision index system includes that first kind decision index system and the second class decision refer toMark；The decision index system is sent to terminal, so that the terminal shows the configuration interface for being associated with the decision index system；It receivesThe rule configuration information that the configuration interface uploads, generates at least one described transaction categories pair according to the rule configuration informationThe decision in the face of risk tree answered.
In one embodiment, transaction data processing unit further include: decision in the face of risk tree screening module is used for the transactionThe corresponding multiple decision in the face of risk trees of classification carry out risk anticipation to positive negative sample, obtain multiple decision in the face of risk trees and correspond toRisk prejudge result；Calculate the matching degree of the positive and negative attribute of the risk anticipation result and the positive negative sample；According to describedMatching degree determines the corresponding target risk decision tree of the transaction categories from multiple decision in the face of risk trees.
In one embodiment, it is corresponding to be also used to calculate multiple decision in the face of risk trees for decision in the face of risk tree screening moduleFormerly arrangement accounting, the first arrangement accounting are the first arrangement ratio of the corresponding decision rule of the first kind decision index systemWeight；Determine that the transaction categories are corresponding from multiple decision in the face of risk trees according to the matching degree and the first arrangement accountingTarget risk decision tree；Wherein, the decision rule in first decision rule comprising the second class decision index system is fewer, and instituteThe decision rule for stating the second class decision index system sorts more rearward in the first decision rule, and the first arrangement accounting is got overIt is high.
In one embodiment, the first correction verification module, if being also used to the first decision rule hit any described firstDecision data, then the decision in the face of risk tree output differentiates as a result, Exit Decision-making engine.
Specific about transaction data processing unit limits the limit that may refer to above for transaction data processing methodFixed, details are not described herein.Modules in above-mentioned transaction data processing unit can fully or partially through software, hardware and itsCombination is to realize.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be withIt is stored in the memory in computer equipment in a software form, in order to which processor calls the above modules of execution correspondingOperation.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junctionComposition can be as shown in Figure 6.The computer equipment include by system bus connect processor, memory, network interface andDatabase.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipmentInclude non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and dataLibrary.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculatingThe database of machine equipment is for storing decision in the face of risk tree.The network interface of the computer equipment is used to pass through net with external terminalNetwork connection communication.To realize a kind of transaction data processing method when the computer program is executed by processor.
It will be understood by those skilled in the art that structure shown in Fig. 6, only part relevant to application scheme is tiedThe block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipmentIt may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, which is stored withComputer program, which performs the steps of when executing computer program obtains trading activity to be discriminated；Lookup and instituteState the corresponding decision in the face of risk tree of transaction categories of trading activity；First first kind decision is extracted from the decision in the face of risk tree to refer toMark, and corresponding first decision data of the first kind decision index system is obtained from the first source；Run the decision in the face of risk treeFirst decision rule, the first decision rule verify first decision data；If the first decision rule is notHit then extracts posterior second class decision index system from the decision in the face of risk tree, and obtains second class from second sourceCorresponding second decision data of decision index system；Run the decision in the face of risk tree in rear decision rule, it is described in rear decision ruleSecond decision data is verified, obtains differentiating result.
In one embodiment, it is also performed the steps of when processor executes computer program and receives the life of decision in the face of risk treeAt request, transaction categories are carried in the request；The corresponding decision index system of the transaction categories is obtained, the decision index system includesFirst kind decision index system and the second class decision index system；The decision index system is sent to terminal, so that terminal display associationThere is the configuration interface of the decision index system；The rule configuration information that the configuration interface uploads is received, is configured according to the ruleInformation generates the corresponding decision in the face of risk tree of at least one described transaction categories.
In one embodiment, it is corresponding that the transaction categories are also performed the steps of when processor executes computer programMultiple decision in the face of risk trees to positive negative sample carry out risk anticipation, it is pre- to obtain the corresponding risk of multiple decision in the face of risk treesSentence result；Calculate the matching degree of the positive and negative attribute of the risk anticipation result and the positive negative sample；According to the matching degree fromThe corresponding target risk decision tree of the transaction categories is determined in multiple decision in the face of risk trees.
In one embodiment, it is also performed the steps of when processor executes computer program and calculates multiple risksThe corresponding first arrangement accounting of decision tree, the first arrangement accounting is the corresponding decision rule of the first kind decision index systemFormerly arrangement specific gravity；The friendship is determined from multiple decision in the face of risk trees according to the matching degree and the first arrangement accountingThe corresponding target risk decision tree of easy classification；Wherein, the decision in first decision rule comprising the second class decision index system is advisedIt is then fewer, and the decision rule of the second class decision index system sorts more rearward in the first decision rule, it is described firstAccounting of arranging is higher.
In one embodiment, if processor also performs the steps of the first decision rule when executing computer programAny first decision data is then hit, then the decision in the face of risk tree output differentiates as a result, Exit Decision-making engine.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculatedMachine program performs the steps of when being executed by processor obtains trading activity to be discriminated；Search the friendship with the trading activityThe corresponding decision in the face of risk tree of easy classification；Extract first first kind decision index system from the decision in the face of risk tree, and from firstSource obtains corresponding first decision data of the first kind decision index system；The first decision rule of the decision in the face of risk tree is run,The first decision rule verifies first decision data；If the first decision rule miss, from describedPosterior second class decision index system is extracted in decision in the face of risk tree, and corresponding from second source acquisition the second class decision index systemSecond decision data；Run the decision in the face of risk tree in rear decision rule, it is described in rear decision rule to second decisionData are verified, and obtain differentiating result.
In one embodiment, it is also performed the steps of when computer program is executed by processor and receives decision in the face of risk treeRequest is generated, carries transaction categories in the request；Obtain the corresponding decision index system of the transaction categories, the decision index system packetInclude first kind decision index system and the second class decision index system；The decision index system is sent to terminal, so that terminal display is closedIt is associated with the configuration interface of the decision index system；The rule configuration information that the configuration interface uploads is received, is matched according to the ruleConfidence breath generates the corresponding decision in the face of risk tree of at least one described transaction categories.
In one embodiment, the transaction categories pair are also performed the steps of when computer program is executed by processorThe multiple decision in the face of risk trees answered carry out risk anticipation to positive negative sample, obtain the corresponding risk of multiple decision in the face of risk treesPrejudge result；Calculate the matching degree of the positive and negative attribute of the risk anticipation result and the positive negative sample；According to the matching degreeThe corresponding target risk decision tree of the transaction categories is determined from multiple decision in the face of risk trees.
In one embodiment, it is also performed the steps of when computer program is executed by processor and calculates multiple windThe corresponding first arrangement accounting of dangerous decision tree, the first arrangement accounting are the corresponding decision rule of the first kind decision index systemFirst arrangement specific gravity；Described in being determined from multiple decision in the face of risk trees according to the matching degree and the first arrangement accountingThe corresponding target risk decision tree of transaction categories；It wherein, include the decision of the second class decision index system in first decision ruleRule is fewer, and the decision rule of the second class decision index system sorts more rearward in the first decision rule, it is describedFirst arrangement accounting is higher.
In one embodiment, if also performing the steps of the first decision when computer program is executed by processorRule hits any first decision data, then the decision in the face of risk tree output differentiates as a result, Exit Decision-making engine.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be withRelevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computerIn read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,To any reference of memory, storage, database or other media used in each embodiment provided herein,Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may includeRandom access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancingType SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodimentIn each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lanceShield all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneouslyIt cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the artIt says, without departing from the concept of this application, various modifications and improvements can be made, these belong to the protection of the applicationRange.Therefore, the scope of protection shall be subject to the appended claims for the application patent.