CN110730235A - Access request response method and device - Google Patents
Access request response method and device Download PDFInfo
- Publication number
- CN110730235A CN110730235A CN201910983592.6A CN201910983592A CN110730235A CN 110730235 A CN110730235 A CN 110730235A CN 201910983592 A CN201910983592 A CN 201910983592A CN 110730235 A CN110730235 A CN 110730235A
- Authority
- CN
- China
- Prior art keywords
- request
- processing
- data
- data processing
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
- H04L67/61—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources taking into account QoS or priority requirements
Abstract
The application provides an access request response method and device, after a data processing request of a user end is received, the user end and current request processing items belong to the same processing item category, and statistical behavior data of historical request processing items in a latest preset time period are determined, whether the data processing request needs to be responded is determined according to the statistical behavior data of the historical request processing items and the request behavior data of the current request processing items, when the data processing request needs to be responded, a processing mode and a processing priority for responding the data processing request can be flexibly selected according to the emergency degree of the current request processing items, and the data processing request is responded according to the selected processing mode and the processing priority, so that the emergency request processing items can be processed timely.
Description
Technical Field
The application relates to the technical field of computer information, in particular to an access request response method and device.
Background
At present, due to rapid development of the internet and popularization of intelligent terminals, users can perform various operations on the internet, such as online payment, account transfer, personal dynamic release, video watching, game playing and the like. These operations generally require the user side to send an access request to the server, and the server, after receiving the access request, will determine whether the access request is legal or not, and respond to the access request when the access request is legal.
Due to the large number of access requests and the limited resources of the server, the access requests need to be put into a task queue and responded in sequence. After the access request needs to be responded, in the prior art, a newly received access request is usually directly placed at the tail of a task queue, and the access request is processed according to the time sequence of receiving the access request, however, request processing items corresponding to different access requests have different urgency degrees, and in the existing access request response method, urgent request processing items cannot be processed in time.
Disclosure of Invention
In view of the above, an object of the embodiments of the present application is to provide an access request response method and apparatus, which can flexibly select a processing manner and a processing priority for responding to an access request according to an urgency level of a current request handling transaction, so that the urgent request handling transaction can be handled in time.
In a first aspect, the present application provides an access request response method, including:
receiving a data processing request of a user side, wherein the data processing request comprises current request processing items of data which are requested to be processed currently;
acquiring statistical behavior data of historical request processing items, which belong to the same processing item class and are within a latest preset time period, of the user side and the current request processing items based on the current request processing items requested by the data processing request;
determining whether the data processing request needs to be responded according to the statistical behavior data of the historical request processing items and the request behavior data of the current request processing items;
after the data processing request needs to be responded, determining a processing mode and a processing priority aiming at the current request processing item according to the emergency degree corresponding to the current request processing item, and responding to the data processing request according to the processing mode and the processing priority;
and after determining that the data processing request does not need to be responded, feeding back illegal processing reminding information to the user side.
In one possible embodiment, determining a processing mode and a processing priority for the current request transaction according to the urgency level corresponding to the current request transaction includes:
determining a processing mode and a processing priority matched with the emergency degree grade according to the emergency degree grade corresponding to the current request processing item; the higher the urgency degree corresponding to the urgency degree level is, the higher the processing priority of the matching is, and the more efficient the matching processing mode is.
In one possible implementation, determining whether the data processing request needs to be responded according to the statistical behavior data of the historical request processing items and the request behavior data of the current request processing item comprises:
determining a first feature vector corresponding to request behavior data of a current request processing item and a second feature vector corresponding to statistical behavior data of a historical request processing item;
splicing the first feature vector and the second feature vector to obtain a third feature vector;
inputting the third feature vector into a detection model obtained by pre-training, and calculating to obtain a risk probability value of the current request handling item;
determining whether a response to the data processing request is required based on the risk probability value.
In one possible embodiment, determining a second feature vector corresponding to the statistical behavior data of the historical request processing item includes:
extracting historical operation behavior characteristic vectors corresponding to a plurality of historical request processing items respectively;
and respectively carrying out processing calculation on the extracted multiple historical operation behavior characteristic vectors under different processing algorithms to obtain multiple characteristic vectors, and splicing the multiple characteristic vectors to obtain the second characteristic vector.
In a possible implementation manner, the processing calculation performed on the extracted multiple historical operation behavior feature vectors under different processing algorithms respectively to obtain multiple feature vectors includes:
and respectively carrying out averaging, maximum value taking, minimum value taking and weighted summation calculation on the extracted multiple historical operation behavior characteristic vectors to obtain the multiple characteristic vectors.
In one possible embodiment, the detection model is obtained by training in the following way:
obtaining a training sample set, wherein the training sample set comprises a positive sample pair and a negative sample pair, each positive sample pair comprises request behavior data of a data processing request which corresponds to the ith time in a plurality of times and can respond, and statistical behavior data within a preset time period before the ith time, and each negative sample pair comprises request behavior data of a data processing request which corresponds to the jth time and cannot respond, and statistical behavior data within a preset time period before the jth time;
extracting feature vectors corresponding to the two types of samples in each positive sample pair respectively, and splicing the extracted feature vectors corresponding to the two types of samples respectively to obtain a first spliced feature vector sample; extracting feature vectors corresponding to the two types of samples in each negative sample pair respectively aiming at each negative sample pair, and splicing the extracted feature vectors corresponding to the two types of samples respectively to obtain a second spliced feature vector sample;
respectively inputting first splicing feature vector samples respectively corresponding to different positive sample pairs and second splicing feature vector samples respectively corresponding to different negative sample pairs into the detection model to obtain a first detection result for each positive sample pair and a second detection result for each negative sample pair;
calculating an error value of the training of the current round based on the first detection result of each positive sample pair and a preset first correct result, the second detection result of each negative sample pair and a preset second correct result;
and when the calculated error value is larger than a set value, adjusting the model parameters of the detection model, and performing the next round of training process by using the adjusted detection model until the calculated error value is not larger than the set value, and determining that the training is finished.
In one possible embodiment, determining whether the data processing request needs to be responded to based on the risk probability value includes:
and when the risk probability value is smaller than a set threshold value, determining that the data processing request needs to be responded, otherwise, determining that the data processing request does not need to be responded.
In a second aspect, an embodiment of the present application further provides an access request responding apparatus, including:
the receiving module is used for receiving a data processing request of a user side, wherein the data processing request comprises current request processing items of data which are requested to be processed currently;
the acquisition module is used for acquiring the statistical behavior data of the historical request processing items, which belong to the same processing item class and are within the latest preset time period, of the user side and the current request processing items based on the current request processing items requested by the data processing request;
a first determining module, configured to determine whether the data processing request needs to be responded according to the statistical behavior data of the historical request processing item and the request behavior data of the current request processing item;
a second determining module, configured to determine, after it is determined that the data processing request needs to be responded, a processing manner and a processing priority for the current request handling item according to the urgency level corresponding to the current request handling item, and respond to the data processing request according to the processing manner and the processing priority;
and the feedback module is used for feeding back illegal processing reminding information to the user side after determining that the data processing request does not need to be responded.
In a possible implementation manner, when determining the processing manner and the processing priority for the current request transaction according to the urgency level corresponding to the current request transaction, the second determining module is specifically configured to:
determining a processing mode and a processing priority matched with the emergency degree grade according to the emergency degree grade corresponding to the current request processing item; the higher the urgency degree corresponding to the urgency degree level is, the higher the processing priority of the matching is, and the more efficient the matching processing mode is.
In one possible implementation, the first determining module includes:
the first determining unit is used for determining a first feature vector corresponding to request behavior data of a current request processing item and a second feature vector corresponding to statistical behavior data of a historical request processing item;
the splicing unit is used for splicing the first feature vector and the second feature vector to obtain a third feature vector;
the calculating unit is used for inputting the third feature vector into a detection model obtained by pre-training and calculating to obtain a risk probability value of the current request processing item;
a second determining unit, configured to determine whether the data processing request needs to be responded based on the risk probability value.
In a possible implementation manner, when determining the second feature vector corresponding to the statistical behavior data of the history request processing item, the first determining unit is specifically configured to:
extracting historical operation behavior characteristic vectors corresponding to a plurality of historical request processing items respectively;
and respectively carrying out processing calculation on the extracted multiple historical operation behavior characteristic vectors under different processing algorithms to obtain multiple characteristic vectors, and splicing the multiple characteristic vectors to obtain the second characteristic vector.
In a possible implementation manner, when the first determining unit performs processing calculation on the extracted multiple historical operation behavior feature vectors under different processing algorithms to obtain multiple feature vectors, the first determining unit is specifically configured to:
and respectively carrying out averaging, maximum value taking, minimum value taking and weighted summation calculation on the extracted multiple historical operation behavior characteristic vectors to obtain the multiple characteristic vectors.
In a possible implementation, the access request responding apparatus further includes a model training module, and the model training module is configured to:
obtaining a training sample set, wherein the training sample set comprises a positive sample pair and a negative sample pair, each positive sample pair comprises request behavior data of a data processing request which corresponds to the ith time in a plurality of times and can respond, and statistical behavior data within a preset time period before the ith time, and each negative sample pair comprises request behavior data of a data processing request which corresponds to the jth time and cannot respond, and statistical behavior data within a preset time period before the jth time;
extracting feature vectors corresponding to the two types of samples in each positive sample pair respectively, and splicing the extracted feature vectors corresponding to the two types of samples respectively to obtain a first spliced feature vector sample; extracting feature vectors corresponding to the two types of samples in each negative sample pair respectively aiming at each negative sample pair, and splicing the extracted feature vectors corresponding to the two types of samples respectively to obtain a second spliced feature vector sample;
respectively inputting first splicing feature vector samples respectively corresponding to different positive sample pairs and second splicing feature vector samples respectively corresponding to different negative sample pairs into the detection model to obtain a first detection result for each positive sample pair and a second detection result for each negative sample pair;
calculating an error value of the training of the current round based on the first detection result of each positive sample pair and a preset first correct result, the second detection result of each negative sample pair and a preset second correct result;
and when the calculated error value is larger than a set value, adjusting the model parameters of the detection model, and performing the next round of training process by using the adjusted detection model until the calculated error value is not larger than the set value, and determining that the training is finished.
In a possible implementation manner, the second determining unit is specifically configured to:
and when the risk probability value is smaller than a set threshold value, determining that the data processing request needs to be responded, otherwise, determining that the data processing request does not need to be responded.
In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating via the bus when the electronic device is operating, the machine-readable instructions when executed by the processor performing the steps of the access request response method as described above.
In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the access request response method as described above.
The access request response method and device provided by the embodiment of the application firstly receive a data processing request of a user side, wherein the data processing request comprises current request processing items of data which are requested to be processed currently; then, acquiring statistical behavior data of historical request processing items, which belong to the same processing item class and are within a latest preset time period, of the user side and the current request processing items based on the current request processing items requested by the data processing request; then, determining whether the data processing request needs to be responded according to the statistical behavior data of the historical request processing items and the request behavior data of the current request processing items; after the data processing request needs to be responded, determining a processing mode and a processing priority aiming at the current request processing item according to the emergency degree corresponding to the current request processing item, and responding to the data processing request according to the processing mode and the processing priority; and after determining that the data processing request does not need to be responded, feeding back illegal processing reminding information to the user side. Compared with the prior art, the method and the device can flexibly select the processing mode and the processing priority for responding the access request aiming at the emergency degree of the current request processing items, so that the emergency request processing items can be processed in time.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
Fig. 1 is a flowchart illustrating an access request response method according to an embodiment of the present application;
FIG. 2 illustrates the steps of training a detection model in an access request response method according to an embodiment of the present application;
fig. 3 is a schematic structural diagram illustrating an access request responding apparatus according to an embodiment of the present application;
fig. 4 is a second schematic structural diagram of an access request responding apparatus according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
Research shows that due to the large number of access requests and the limited resources of the server, the access requests need to be put into a task queue and responded in sequence. After the access request needs to be responded, in the prior art, a newly received access request is usually directly placed at the tail of a task queue, and the access request is processed according to the time sequence of receiving the access request, however, request processing items corresponding to different access requests have different urgency degrees, and in the existing access request response method, urgent request processing items cannot be processed in time.
Based on this, the embodiment of the present application provides an access request response method, which can flexibly select a processing mode and a processing priority for responding to an access request according to the urgency of the current request processing event, so that the urgent request processing event can be processed in time.
Referring to fig. 1, fig. 1 is a flowchart of an access request response method according to an embodiment of the present disclosure. As shown in fig. 1, the method may be executed by a server, and includes the specific steps of:
s101, receiving a data processing request of a user side, wherein the data processing request comprises current request processing items of data which are requested to be processed currently.
In this step, the user side may send a data processing request to the server according to an instruction input by the user, where the current request processing item may include transfer, message sending, login, query, and the like, the current request processing data may include request time, request content, authentication information, and the like, and the server may process the data request according to the data corresponding to the data processing request, thereby completing processing of the current request processing item.
S102, acquiring statistical behavior data of historical request processing items, of which the user side and the current request processing items belong to the same processing item class and within the latest preset time period, based on the current request processing items requested by the data processing request.
Specifically, if the current request handling item is login, statistical behavior data of the user login in the last week can be obtained, and the statistical behavior data can include login time, login place, authentication information during login, equipment used during login and the like of the user; if the current request processing item is ordinary transfer, statistical behavior data of all transfers of the user in the last week can be obtained, such as real-time transfer, ordinary transfer, cross-bank transfer and the like.
S103, determining whether the data processing request needs to be responded according to the statistical behavior data of the historical request processing items and the request behavior data of the current request processing items.
In this step, it may be determined whether the current data processing request is a normal data processing request according to a difference between statistical behavior data of the historical request processing items and request behavior data of the current request processing item of the user, and a response is required if the current data processing request is a normal data processing request, and no response is required if the data processing request is an abnormal data processing request.
Therefore, whether the data processing request is an abnormal data processing request or not is judged, and whether the data processing request needs to be responded or not is further judged, so that the safety of access request response can be improved, and the account number of the user is prevented from being illegally stolen.
S104, after the data processing request is determined to need to be responded, determining a processing mode and a processing priority aiming at the current request processing item according to the emergency degree corresponding to the current request processing item, and responding to the data processing request according to the processing mode and the processing priority.
In this step, after determining that the data processing request needs to be responded, the urgency level of the currently requested processing item may be determined according to the type and specific content of the currently requested processing item, whether the currently requested processing item has an urgency tag, the expected completion time, and other information.
Specifically, if the current request processing item is a real-time transfer, the urgency level of the request processing item is higher, and if the current request processing item is a normal transfer, the urgency level of the request processing item is lower; similarly, if the request behavior data of the current request transaction carries an urgency tag, the urgency of the request transaction is increased. The urgency level may be determined by a preset model, or the data processing request itself carries the urgency level of the current request handling transaction.
Further, the specific content and the urgency of the current request handling item may be combined to determine the handling manner and the handling priority corresponding to the current request handling item. Specifically, if the current request processing item is the real-time transfer and the urgency level thereof is the secondary urgency level, the processing priority level of the current request processing item can be set as the secondary priority level corresponding to the secondary urgency level, the current request processing item is placed at the position corresponding to the secondary priority level in the task queue, and the data processing request is responded according to the processing mode corresponding to the real-time transfer; similarly, if the current request transaction is a general transfer and the urgency level thereof is a first-level urgency level, the processing priority level of the current request transaction may be set to a first-level priority level corresponding to the first-level urgency level, the current request transaction may be placed at a position in the task queue corresponding to the first-level priority level, and the data processing request may be responded according to the processing mode of arriving at the account the next day.
In this way, the request processing items with high urgency can be processed preferentially by adopting a processing mode with a high processing speed, and the request processing items with low urgency can be processed at a later time by adopting a processing mode with low resource consumption, so that the probability that the request processing items with high urgency cannot be processed in time can be reduced.
And S105, after determining that the data processing request does not need to be responded, feeding back illegal processing reminding information to the user side.
In the step, the user can be informed that the current data request is illegal through modes of short messages, telephone calls, pushing, mails and the like, so that the user can confirm the terminal or the account number of the user in time, and the loss of information or property is prevented.
In one possible embodiment, determining a processing mode and a processing priority for the current request transaction according to the urgency level corresponding to the current request transaction includes:
determining a processing mode and a processing priority matched with the emergency degree grade according to the emergency degree grade corresponding to the current request processing item; the higher the urgency degree corresponding to the urgency degree level is, the higher the processing priority of the matching is, and the more efficient the matching processing mode is.
In this step, the processing manner and the processing priority matched with the urgency level may be determined according to the mapping relationship between the urgency level of the currently requested processing item and the processing manner and the processing priority, each urgency level may correspond to at least one processing manner and processing priority, and when the urgency level corresponds to a plurality of processing manners and processing priorities, the processing manner and processing priority may be selected according to the specific content and processing conditions of the currently requested processing item.
In one possible implementation, determining whether the data processing request needs to be responded according to the statistical behavior data of the historical request processing items and the request behavior data of the current request processing item comprises:
determining a first feature vector corresponding to request behavior data of a current request processing item and a second feature vector corresponding to statistical behavior data of a historical request processing item; splicing the first feature vector and the second feature vector to obtain a third feature vector; inputting the third feature vector into a detection model obtained by pre-training, and calculating to obtain a risk probability value of the current request handling item; determining whether a response to the data processing request is required based on the risk probability value.
In this step, key data of different dimensions in the request behavior data of the current request processing item may be extracted, and the key data of all dimensions may be converted into the first feature vector together. The same is done for the statistical behavior data of the historical request handling transactions.
Specifically, the non-digital data in the request behavior data of the current request handling transaction may be represented by using a specific number, the original digital form data is not changed, and the request behavior data is represented in the form of a vector; and removing damaged data in the request behavior data expressed by the vector to obtain a first feature vector and a second feature vector.
Further, the detection model can determine a risk probability value of the current request processing item according to the current request behavior data in the third feature vector and the feature data in the historical statistical behavior data, the risk probability value represents a probability that the current request processing item is an abnormal request, the risk probability value is obtained by combining the feature data in the current request behavior data and the historical statistical behavior data and the association between the feature data and the historical statistical behavior data, and the risk probability value has higher reliability compared with the risk probability value in the prior art.
Further, after the risk probability value is obtained, whether the data processing request needs to be responded or not can be judged according to the risk probability value, and specifically, if the risk probability value is greater than or equal to a preset threshold value, it can be determined that the data processing request does not need to be responded.
In one possible embodiment, determining a second feature vector corresponding to the statistical behavior data of the historical request processing item includes:
extracting historical operation behavior characteristic vectors corresponding to a plurality of historical request processing items respectively;
and respectively carrying out processing calculation on the extracted multiple historical operation behavior characteristic vectors under different processing algorithms to obtain multiple characteristic vectors, and splicing the multiple characteristic vectors to obtain the second characteristic vector.
Specifically, the average value of the extracted historical operation behavior feature vectors may be calculated first, then the maximum value and the minimum value in the historical operation behavior feature vectors are determined, and finally the weighted average value of the historical operation behavior feature vectors is determined.
And finally, splicing the multiple calculated feature vectors to obtain a second feature vector. The second feature vector obtained by the method can effectively embody the relevant information in the historical request behavior data, and further improves the reliability of the risk probability value.
In a possible implementation manner, the processing calculation performed on the extracted multiple historical operation behavior feature vectors under different processing algorithms respectively to obtain multiple feature vectors includes:
and respectively carrying out averaging, maximum value taking, minimum value taking and weighted summation calculation on the extracted multiple historical operation behavior characteristic vectors to obtain the multiple characteristic vectors.
Referring to fig. 2, fig. 2 is a flowchart illustrating steps of training a detection model in an access request response method according to an embodiment of the present disclosure. In one possible embodiment, as shown in fig. 2, the detection model is obtained by training in the following way:
s201, a training sample set is obtained, wherein the training sample set comprises a positive sample pair and a negative sample pair, each positive sample pair comprises request behavior data of a data processing request which corresponds to the ith time in a plurality of times and can respond, and statistical behavior data in a preset time period before the ith time, and each negative sample pair comprises request behavior data of a data processing request which corresponds to the jth time and cannot respond, and statistical behavior data in a preset time period before the jth time.
The data processing method comprises the steps that request behavior data of data processing requests capable of responding indicate that the data processing requests corresponding to the request behavior data are normal data processing requests; the request behavior data of the data processing request which cannot be responded indicates that the data processing request is an abnormal data processing request.
Further, in the embodiments of the present application, the time indicated by i and j is not limited. The preset time period can be determined step by step through the early training process of the detection model.
S202, extracting the feature vectors corresponding to the two types of samples in each positive sample pair, and splicing the extracted feature vectors corresponding to the two types of samples to obtain a first spliced feature vector sample; and extracting the feature vectors corresponding to the two types of samples in each negative sample pair aiming at each negative sample pair, and splicing the extracted feature vectors corresponding to the two types of samples respectively to obtain a second spliced feature vector sample.
In the step, firstly, the processing can be carried out according to the steps of obtaining the first characteristic vector and the second characteristic vector, after damaged data are removed, the data are enhanced, the number of positive samples and negative samples is balanced, finally, the data are screened and standardized, the data with low correlation are removed, the data are reduced in dimension and are mapped to the same numerical value range, and the model training speed and the model identification accuracy rate are further improved.
S203, inputting the first splicing feature vector samples respectively corresponding to different positive sample pairs and the second splicing feature vector samples respectively corresponding to different negative sample pairs into the detection model respectively to obtain a first detection result for each positive sample pair and a second detection result for each negative sample pair.
In this step, after the detection model obtains the risk probability threshold, the obtained risk probability threshold may be compared with a preset threshold, and when the comparison result is that the risk probability threshold is greater than the preset threshold, a first detection result with a value of 0 may be output, otherwise, a first detection result with a value of 1 may be output.
S204, calculating an error value of the training of the current round based on the first detection result of each positive sample pair and a preset first correct result, the second detection result of each negative sample pair and a preset second correct result.
Specifically, the preset first correct result of each positive sample pair may be set to 1, and the preset second correct result of each negative sample pair may be set to 0. Comparing the first detection result of each positive sample pair with a preset first correct result, comparing the second detection result of each negative sample pair with a preset second correct result, and synthesizing the comparison results of the first detection result and the preset first correct result and the comparison results of the second detection result and the preset second correct result to obtain an error value.
S205, when the calculated error value is larger than a set value, adjusting the model parameters of the detection model, and performing the next round of training process by using the adjusted detection model until the calculated error value is not larger than the set value, and determining that the training is finished.
In one possible embodiment, determining whether the data processing request needs to be responded to based on the risk probability value includes:
and when the risk probability value is smaller than a set threshold value, determining that the data processing request needs to be responded, otherwise, determining that the data processing request does not need to be responded.
Based on the same inventive concept, an access request responding apparatus is further provided in the embodiments of the present application, and the principle of solving the problem with the apparatus in the embodiments of the present application is similar to that of the access request responding method in the embodiments of the present application, so the implementation of the apparatus may refer to the implementation of the method, and repeated details are not repeated.
Referring to fig. 3 and fig. 4, fig. 3 is a diagram of one access request responding apparatus according to an embodiment of the present application, and fig. 4 is a diagram of another access request responding apparatus according to an embodiment of the present application. As shown in fig. 3, the access request responding apparatus 300 includes:
a receiving module 310, configured to receive a data processing request from a user side, where the data processing request includes a current request transaction item for data currently requested to be processed;
an obtaining module 320, configured to obtain statistical behavior data of historical request processing events within a latest preset time period, where the user side and the current request processing event belong to a same processing event category, based on a current request processing event requested by the data processing request;
a first determining module 330, configured to determine whether to respond to the data processing request according to the statistical behavior data of the historical request transaction and the request behavior data of the current request transaction;
a second determining module 340, configured to determine, after it is determined that the data processing request needs to be responded, a processing manner and a processing priority for the current request processing item according to the urgency level corresponding to the current request processing item, and respond to the data processing request according to the processing manner and the processing priority;
a feedback module 350, configured to feed back an illegal processing reminding message to the user side after determining that the data processing request does not need to be responded.
In a possible implementation manner, when determining the processing manner and the processing priority for the current request transaction according to the urgency level corresponding to the current request transaction, the second determining module 340 is specifically configured to:
determining a processing mode and a processing priority matched with the emergency degree grade according to the emergency degree grade corresponding to the current request processing item; the higher the urgency degree corresponding to the urgency degree level is, the higher the processing priority of the matching is, and the more efficient the matching processing mode is.
As shown in fig. 4, in one possible implementation, the first determining module 330 includes:
a first determining unit 331, configured to determine a first feature vector corresponding to request behavior data of a current request processing item, and a second feature vector corresponding to statistical behavior data of the historical request processing item;
a splicing unit 332, configured to splice the first eigenvector and the second eigenvector to obtain a third eigenvector;
a calculating unit 333, configured to input the third feature vector into a detection model obtained through pre-training, and calculate a risk probability value of the current request transaction;
a second determining unit 334, configured to determine whether a response to the data processing request is required based on the risk probability value.
In a possible implementation manner, when determining the second feature vector corresponding to the statistical behavior data of the history request processing item, the first determining unit 331 is specifically configured to:
extracting historical operation behavior characteristic vectors corresponding to a plurality of historical request processing items respectively;
and respectively carrying out processing calculation on the extracted multiple historical operation behavior characteristic vectors under different processing algorithms to obtain multiple characteristic vectors, and splicing the multiple characteristic vectors to obtain the second characteristic vector.
In a possible implementation manner, when the first determining unit 331 performs processing calculation on the extracted multiple historical operation behavior feature vectors under different processing algorithms to obtain multiple feature vectors, specifically, the first determining unit is configured to:
and respectively carrying out averaging, maximum value taking, minimum value taking and weighted summation calculation on the extracted multiple historical operation behavior characteristic vectors to obtain the multiple characteristic vectors.
In a possible implementation, the access request responding apparatus 300 further includes a model training module 360, and the model training module 360 is configured to:
obtaining a training sample set, wherein the training sample set comprises a positive sample pair and a negative sample pair, each positive sample pair comprises request behavior data of a data processing request which corresponds to the ith time in a plurality of times and can respond, and statistical behavior data within a preset time period before the ith time, and each negative sample pair comprises request behavior data of a data processing request which corresponds to the jth time and cannot respond, and statistical behavior data within a preset time period before the jth time;
extracting feature vectors corresponding to the two types of samples in each positive sample pair respectively, and splicing the extracted feature vectors corresponding to the two types of samples respectively to obtain a first spliced feature vector sample; extracting feature vectors corresponding to the two types of samples in each negative sample pair respectively aiming at each negative sample pair, and splicing the extracted feature vectors corresponding to the two types of samples respectively to obtain a second spliced feature vector sample;
respectively inputting first splicing feature vector samples respectively corresponding to different positive sample pairs and second splicing feature vector samples respectively corresponding to different negative sample pairs into the detection model to obtain a first detection result for each positive sample pair and a second detection result for each negative sample pair;
calculating an error value of the training of the current round based on the first detection result of each positive sample pair and a preset first correct result, the second detection result of each negative sample pair and a preset second correct result;
and when the calculated error value is larger than a set value, adjusting the model parameters of the detection model, and performing the next round of training process by using the adjusted detection model until the calculated error value is not larger than the set value, and determining that the training is finished.
In a possible implementation manner, the second determining unit 334 is specifically configured to:
and when the risk probability value is smaller than a set threshold value, determining that the data processing request needs to be responded, otherwise, determining that the data processing request does not need to be responded.
The access request responding device provided by the embodiment of the application firstly receives a data processing request of a user side, wherein the data processing request comprises current request processing items of data which are requested to be processed currently; then, acquiring statistical behavior data of historical request processing items, which belong to the same processing item class and are within a latest preset time period, of the user side and the current request processing items based on the current request processing items requested by the data processing request; then, determining whether the data processing request needs to be responded according to the statistical behavior data of the historical request processing items and the request behavior data of the current request processing items; after the data processing request needs to be responded, determining a processing mode and a processing priority aiming at the current request processing item according to the emergency degree corresponding to the current request processing item, and responding to the data processing request according to the processing mode and the processing priority; and after determining that the data processing request does not need to be responded, feeding back illegal processing reminding information to the user side. Compared with the prior art, the method and the device can flexibly select the processing mode and the processing priority for responding the access request aiming at the emergency degree of the current request processing items, so that the emergency request processing items can be processed in time.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. As shown in fig. 5, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.
The memory 520 stores machine-readable instructions executable by the processor 510, when the electronic device 500 runs, the processor 510 communicates with the memory 520 through the bus 530, and when the machine-readable instructions are executed by the processor 510, the steps of the access request response method in the method embodiments shown in fig. 1 and fig. 2 may be executed.
An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the access request response method in the method embodiments shown in fig. 1 and fig. 2 may be executed.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
Claims (10)
1. An access request response method, comprising:
receiving a data processing request of a user side, wherein the data processing request comprises current request processing items of data which are requested to be processed currently;
acquiring statistical behavior data of historical request processing items, which belong to the same processing item class and are within a latest preset time period, of the user side and the current request processing items based on the current request processing items requested by the data processing request;
determining whether the data processing request needs to be responded according to the statistical behavior data of the historical request processing items and the request behavior data of the current request processing items;
after the data processing request needs to be responded, determining a processing mode and a processing priority aiming at the current request processing item according to the emergency degree corresponding to the current request processing item, and responding to the data processing request according to the processing mode and the processing priority;
and after determining that the data processing request does not need to be responded, feeding back illegal processing reminding information to the user side.
2. The method of claim 1, wherein determining a processing manner and a processing priority for the current request transaction according to the urgency level corresponding to the current request transaction comprises:
determining a processing mode and a processing priority matched with the emergency degree grade according to the emergency degree grade corresponding to the current request processing item; the higher the urgency degree corresponding to the urgency degree level is, the higher the processing priority of the matching is, and the more efficient the matching processing mode is.
3. The method of claim 1, wherein determining whether the data processing request needs to be responded to based on the statistical behavior data of the historical request transaction and the request behavior data of the current request transaction comprises:
determining a first feature vector corresponding to request behavior data of a current request processing item and a second feature vector corresponding to statistical behavior data of a historical request processing item;
splicing the first feature vector and the second feature vector to obtain a third feature vector;
inputting the third feature vector into a detection model obtained by pre-training, and calculating to obtain a risk probability value of the current request handling item;
determining whether a response to the data processing request is required based on the risk probability value.
4. The method of claim 3, wherein determining a second feature vector corresponding to statistical behavior data for the historical request handling transactions comprises:
extracting historical operation behavior characteristic vectors corresponding to a plurality of historical request processing items respectively;
and respectively carrying out processing calculation on the extracted multiple historical operation behavior characteristic vectors under different processing algorithms to obtain multiple characteristic vectors, and splicing the multiple characteristic vectors to obtain the second characteristic vector.
5. The method according to claim 4, wherein the processing calculation of the extracted plurality of historical operating behavior feature vectors under different processing algorithms is performed to obtain a plurality of feature vectors, and the method comprises:
and respectively carrying out averaging, maximum value taking, minimum value taking and weighted summation calculation on the extracted multiple historical operation behavior characteristic vectors to obtain the multiple characteristic vectors.
6. The method of claim 3, wherein the detection model is trained by:
obtaining a training sample set, wherein the training sample set comprises a positive sample pair and a negative sample pair, each positive sample pair comprises request behavior data of a data processing request which corresponds to the ith time in a plurality of times and can respond, and statistical behavior data within a preset time period before the ith time, and each negative sample pair comprises request behavior data of a data processing request which corresponds to the jth time and cannot respond, and statistical behavior data within a preset time period before the jth time;
extracting feature vectors corresponding to the two types of samples in each positive sample pair respectively, and splicing the extracted feature vectors corresponding to the two types of samples respectively to obtain a first spliced feature vector sample; extracting feature vectors corresponding to the two types of samples in each negative sample pair respectively aiming at each negative sample pair, and splicing the extracted feature vectors corresponding to the two types of samples respectively to obtain a second spliced feature vector sample;
respectively inputting first splicing feature vector samples respectively corresponding to different positive sample pairs and second splicing feature vector samples respectively corresponding to different negative sample pairs into the detection model to obtain a first detection result for each positive sample pair and a second detection result for each negative sample pair;
calculating an error value of the training of the current round based on the first detection result of each positive sample pair and a preset first correct result, the second detection result of each negative sample pair and a preset second correct result;
and when the calculated error value is larger than a set value, adjusting the model parameters of the detection model, and performing the next round of training process by using the adjusted detection model until the calculated error value is not larger than the set value, and determining that the training is finished.
7. The method of claim 3, wherein determining whether the data processing request needs to be responded to based on the risk probability value comprises:
and when the risk probability value is smaller than a set threshold value, determining that the data processing request needs to be responded, otherwise, determining that the data processing request does not need to be responded.
8. An access request responding apparatus, comprising
The receiving module is used for receiving a data processing request of a user side, wherein the data processing request comprises current request processing items of data which are requested to be processed currently;
the acquisition module is used for acquiring the statistical behavior data of the historical request processing items, which belong to the same processing item class and are within the latest preset time period, of the user side and the current request processing items based on the current request processing items requested by the data processing request;
a first determining module, configured to determine whether the data processing request needs to be responded according to the statistical behavior data of the historical request processing item and the request behavior data of the current request processing item;
a second determining module, configured to determine, after it is determined that the data processing request needs to be responded, a processing manner and a processing priority for the current request handling item according to the urgency level corresponding to the current request handling item, and respond to the data processing request according to the processing manner and the processing priority;
and the feedback module is used for feeding back illegal processing reminding information to the user side after determining that the data processing request does not need to be responded.
9. An electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the electronic device is operating, the processor executing the machine-readable instructions to perform the steps of the access request response method according to any one of claims 1 to 7.
10. A computer-readable storage medium, having stored thereon a computer program for performing, when executed by a processor, the steps of the access request response method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910983592.6A CN110730235A (en) | 2019-10-16 | 2019-10-16 | Access request response method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910983592.6A CN110730235A (en) | 2019-10-16 | 2019-10-16 | Access request response method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110730235A true CN110730235A (en) | 2020-01-24 |
Family
ID=69221379
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910983592.6A Pending CN110730235A (en) | 2019-10-16 | 2019-10-16 | Access request response method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110730235A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101383996A (en) * | 2007-09-05 | 2009-03-11 | 华为技术有限公司 | Method and device for avoiding request collision in short data service |
| US20140094134A1 (en) * | 2012-09-28 | 2014-04-03 | Eddie Balthasar | Mechanism for facilitating dynamically prioritized control of calls over a network |
| CN107896180A (en) * | 2017-10-24 | 2018-04-10 | 北京小蓦机器人技术有限公司 | Equipment room cooperates with method, equipment, system and the storage medium of processing event |
| CN109191136A (en) * | 2018-09-05 | 2019-01-11 | 北京芯盾时代科技有限公司 | A kind of e-bank is counter to cheat method and device |
| CN109462634A (en) * | 2018-09-25 | 2019-03-12 | 郑州云海信息技术有限公司 | Message treatment method, device and equipment in a kind of distributed system |
-
2019
- 2019-10-16 CN CN201910983592.6A patent/CN110730235A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101383996A (en) * | 2007-09-05 | 2009-03-11 | 华为技术有限公司 | Method and device for avoiding request collision in short data service |
| US20140094134A1 (en) * | 2012-09-28 | 2014-04-03 | Eddie Balthasar | Mechanism for facilitating dynamically prioritized control of calls over a network |
| CN107896180A (en) * | 2017-10-24 | 2018-04-10 | 北京小蓦机器人技术有限公司 | Equipment room cooperates with method, equipment, system and the storage medium of processing event |
| CN109191136A (en) * | 2018-09-05 | 2019-01-11 | 北京芯盾时代科技有限公司 | A kind of e-bank is counter to cheat method and device |
| CN109462634A (en) * | 2018-09-25 | 2019-03-12 | 郑州云海信息技术有限公司 | Message treatment method, device and equipment in a kind of distributed system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107566358B (en) | Risk early warning prompting method, device, medium and equipment | |
| CN110096244B (en) | Information sharing method based on data processing and related equipment | |
| CN112801800A (en) | Behavior fund analysis system, behavior fund analysis method, computer equipment and storage medium | |
| CN110267215B (en) | Data detection method, equipment and storage medium | |
| CN110955581A (en) | Online software abnormity warning method and device, electronic equipment and storage medium | |
| CN111340574B (en) | Risk user identification method and device and electronic equipment | |
| CN111737055A (en) | Service processing method, device, equipment and computer readable storage medium | |
| CN112530074A (en) | Queuing and calling reminding method, device, equipment and storage medium | |
| CN110730235A (en) | Access request response method and device | |
| CN111738723B (en) | Online secure transaction method and device and readable storage medium | |
| CN115239450A (en) | Financial data processing method and device, computer equipment and storage medium | |
| CN114358543A (en) | Information processing method and device | |
| CN114363268A (en) | Message delay consumption method and related product | |
| CN114297735A (en) | Data processing method and related device | |
| CN113157911A (en) | Service verification method and device | |
| CN111309744A (en) | Object processing method and device, electronic equipment and computer readable storage medium | |
| CN109753599A (en) | The method and apparatus of recommendation business | |
| CN113905336B (en) | Text recognition-based short message sending method, device, equipment and storage medium | |
| CN117251769B (en) | Abnormal data identification method, device, equipment and medium based on monitoring component | |
| CN112422414B (en) | Session message processing method, device and computer readable storage medium | |
| CN109889590B (en) | Message processing method, device, client and computer readable medium | |
| CN114764713A (en) | Method and device for generating merchant patrol task, electronic equipment and storage medium | |
| CN116862605A (en) | Bill generation method, device, electronic equipment and storage medium | |
| CN114138972A (en) | Text type identification method and device | |
| CN111163125A (en) | Channel control method, device, electronic equipment and computer readable medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200124 |
|
| RJ01 | Rejection of invention patent application after publication |