Access request processing method, electronic equipment, storage medium and program product
The router automatically identifies network addressing identifiers by using the identifier classification model trained through fine-tuning of a large model, solving the problems of complex and poor real-time performance of network resource blacklist filtering operations, and improving the timeliness and accuracy of network security.
Patent Information
- Application Number
- CN202510977910.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-16
- Publication Date
- 2025-09-30
AI Technical Summary
In the prior art, the network resource blacklist filtering function requires manual setting by the user, which is complicated to operate and has poor real-time performance and cannot be updated in time, resulting in the inability to filter illegal network resources in time.
The router receives access requests from the client and uses the identification classification model based on large model fine-tuning training to automatically identify network addressing identification, dynamically classify illegal network resources, and achieve automatic interception.
It improves the timeliness and accuracy of network security protection, reduces the burden of user management and maintenance, and reduces the threat of illegal network resources.
Smart Images
Figure CN120729602A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of communication technologies, and in particular to an access request processing method, electronic device, storage medium, and program product. Background Art
[0002] With the rapid development of mobile networking, mobile broadband (MBB) has become one of the primary ways for users to access the Internet. In home and enterprise networks, routers, as devices connecting local area networks (LANs) and wide area networks (WANs), typically provide basic network security functions, such as domain name filtering.
[0003] However, in related technologies, when implementing the network resource blacklist filtering function, users are required to log in to the webUI (Web User Interface) to set the blacklist. The operation is complicated and has poor real-time performance. Newly emerging illegal network resources cannot be updated to the webUI in time, resulting in filtering failure. Summary of the Invention
[0004] The embodiments of the present application provide an access request processing method, electronic device, storage medium, and program product, which can solve the problem of network resource blacklist filtering failure caused by the inability to update illegal network resources in a timely manner. The technical solution is as follows.
[0005] In one aspect, a method for processing an access request is provided, the method being executed by a router, the method comprising: Receive a target access request sent by a client; the target access request includes a target network addressing identifier; the target network addressing identifier is used to indicate the network resource accessed by the target access request; Sending the target network addressing identifier to an identifier classification model for identifier classification to obtain an identifier classification result of the target network addressing identifier; the identifier classification result is used to indicate whether the target network addressing identifier is a blacklist identifier or a whitelist identifier; the identifier classification model is obtained by fine-tuning training based on a large model; The target access request is processed based on the identification classification result of the target network addressing identification.
[0006] On the other hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the above-mentioned access request processing method.
[0007] On the other hand, a computer-readable storage medium is provided, wherein at least one computer program is stored in the computer-readable storage medium, and the computer program is loaded and executed by a processor to implement the above-mentioned access request processing method.
[0008] On the other hand, a computer program product is provided, comprising a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program comprises program instructions, which, when executed by a computer, cause the computer to execute to implement the access request processing method provided in the various optional implementations described above.
[0009] The access request processing method provided in the embodiment of the present application is that the router extracts the target network addressing identifier from the target access request sent by the client, sends the target network addressing identifier to the identifier classification model for identifier classification and identification, obtains the identifier classification result of the target network addressing identifier, and processes the target access request based on the identifier classification result of the target network addressing identifier, wherein the identifier classification model is obtained by fine-tuning and training based on a large model; by adopting the identifier classification model fine-tuned based on the large model, the router can automatically classify and identify the target network addressing identifier dynamically, and can automatically identify and intercept various threats including illegal network resources without relying on human intervention, while improving the timeliness and accuracy of network security protection, and significantly reducing the user management and maintenance burden.
[0010] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0012] Figure 1 A flowchart of a method for processing an access request provided by an exemplary embodiment of the present application is shown; Figure 2 A flowchart of a method for processing an access request provided by another exemplary embodiment of the present application is shown; Figure 3 A schematic diagram showing a method for processing an access request provided by an exemplary embodiment of the present application is shown; Figure 4 A schematic diagram of an access request processing method provided by another exemplary embodiment of the present application is shown. DETAILED DESCRIPTION
[0013] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numbers in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present application. Rather, they are merely examples of methods consistent with certain aspects of the present application, as detailed in the appended claims.
[0014] In order to improve the security of users' Internet access and enhance the protection effect against illegal network resources, the present application provides an access request processing method. Figure 1 A flowchart of an access request processing method provided by an exemplary embodiment of the present application is shown. The method can be executed by a router, such as Figure 1 As shown, the method may include the following steps.
[0015] Step 110: Receive a target access request sent by a client; the target access request includes a target network addressing identifier; the target network addressing identifier is used to indicate the network resource to be accessed by the target access request.
[0016] In an embodiment of the present application, when the router is in an online state, it receives a target access request sent by the client, that is, the device first performs an online operation after being turned on. After the router is successfully connected to the network, the router monitors the target access request sent by the client in real time and extracts the target network addressing identifier used to indicate the network resource accessed by the target access request.
[0017] Step 120, the target network addressing identifier is sent to the identifier classification model for identifier classification to obtain the identifier classification result of the target network addressing identifier; the identifier classification result is used to indicate whether the target network addressing identifier is a blacklist identifier or a whitelist identifier; the identifier classification model is obtained by fine-tuning training based on a large model.
[0018] In one possible implementation, the identification classification model can be deployed on a computer device, and the router can establish a data connection with the computer device to send the network addressing identification to the computer device. After receiving the network addressing identification, the computer device classifies the network addressing identification through the identification classification model and feeds back the obtained identification classification result to the router.
[0019] In one possible implementation, the router can send the target network addressing identifier to the identifier classification model in real time for identifier classification to determine whether the network resource currently accessed by the client is an illegal network resource. When the identifier classification result indicates that the target network addressing identifier is a blacklist identifier, the network resource currently accessed is determined to be an illegal network resource. When the identifier classification result indicates that the target network addressing identifier is a whitelist identifier, the network resource currently accessed is determined to be a legal network resource.
[0020] In another possible implementation, the router may first cache the target network addressing identifier, and after a certain period, send the cached network addressing identifier to the identifier classification model for batch identifier classification to obtain the identifier classification results of each network addressing identifier. After receiving the same network addressing identifier again, the corresponding access request is processed based on the known identifier classification results.
[0021] Among them, the identification classification model can be obtained by fine-tuning the large model through the training sample set, and the training sample set can include network addressing identification samples and identification classification labels corresponding to the network addressing identification samples. By fine-tuning the large model, the large model can have identification classification capabilities. The large model can be a known large language model. This application does not limit the type of large model. In principle, the fine-tuning training process can be implemented as follows: the network addressing identification sample and the prompt word indicating identification classification are input into the large model to obtain the predicted identification classification result output by the large model, and the loss function is calculated based on the predicted identification classification result and the identification classification label. The parameters of the large model are updated based on the function value of the loss function. The above process is repeated until the training completion condition is met to obtain the identification classification model, and the training completion condition can be one of the following: the number of training times reaches the number threshold, and the function value of the loss function converges.
[0022] In a possible implementation, the identification classification model can be updated according to a preset period, and the recognition effect of variant blacklist identification can be improved through continuous learning.
[0023] Step 130: Process the target access request based on the identification classification result of the target network addressing identification.
[0024] If the identification classification result of the target network addressing identifier indicates that the target access request is an illegal request, the target access request is intercepted; if the identification classification result of the target network addressing identifier indicates that the target access request is a legal request, the target access request is forwarded.
[0025] To summarize, the access request processing method provided in the embodiment of the present application is that the router extracts the target network addressing identifier from the target access request sent by the client, sends the target network addressing identifier to the identifier classification model for identifier classification, obtains the identifier classification result of the target network addressing identifier, and processes the target access request based on the identifier classification result of the target network addressing identifier, wherein the identifier classification model is obtained by fine-tuning and training based on a large model; by adopting the identifier classification model fine-tuned based on the large model, the router can automatically and dynamically classify and identify the target network addressing identifier, and can automatically identify and intercept various threats including illegal network resources without relying on human intervention, while improving the timeliness and accuracy of network security protection, and significantly reducing the user management and maintenance burden.
[0026] In order to improve the classification efficiency of network addressing identifiers, in a possible implementation, a local database can be maintained in the router, which can record known network addressing identifiers and corresponding identifier classification results. Before interacting with the identifier classification model to perform identifier classification, the router device can query the identifier classification results in the local database to perform subsequent processing based on the query results. Based on this, Figure 2 A flowchart of an access request processing method provided by another exemplary embodiment of the present application is shown. The method can be executed by a router, such as Figure 2 As shown, the method may include the following steps.
[0027] Step 210: Receive a target access request sent by a client; the target access request includes a target network addressing identifier; the target network addressing identifier is used to indicate the network resource to be accessed by the target access request.
[0028] The target access request is any one of the access requests received by the router. The access request may be a DNS (Domain Name System) request, and correspondingly, the target network addressing identifier in the DNS request is the access domain name, or it may be a direct access request based on IP (Internet Protocol), and correspondingly, the target network addressing identifier in the direct access request based on IP is the access IP address; or, the target access request may be an application layer protocol request, and correspondingly, the target network addressing identifier of the application layer protocol request includes the access domain name and the access IP address, that is, the network addressing identifier includes at least one of the following: the access domain name and the access IP address.
[0029] The router can obtain the network addressing identifier in the access request by parsing the received access request.
[0030] Step 220: query the target network addressing identifier in the local database to obtain the query result of the target network addressing identifier; the local database is used to record the network addressing identifier and the corresponding identifier classification result.
[0031] The sources of the network addressing identifiers and corresponding identifier classification results recorded in the local database may include at least one of the following: Relevant personnel manually add known network addressing identifiers and corresponding identifier classification results to the local database; The router obtains the historical network addressing identifiers in the user's historical access requests in advance, obtains the identifier classification results of each historical network addressing identifier through interaction with the identifier classification model, and stores them in the local database; The router updates the identification classification results of the network addressing identification obtained in real time or periodically into the local database; The user sets the identification classification of the network addressing identification through the interactive interface; by providing an interactive interface to enable user customization, flexibility can be increased and the occurrence of misjudgment can be reduced.
[0032] After receiving the user's target access request, the router first searches the local database to see whether the target network addressing identifier in the target access request is included. If the target network addressing identifier has been recorded in the local database, the identification classification result of the target network addressing identifier can be directly obtained from the local database. Otherwise, the identification classification result of the target network addressing identifier is determined by interacting with the identification classification model.
[0033] Step 230: Process the target access request based on the query result of the target network addressing identifier.
[0034] Step 231 : In response to the query of the target network addressing identifier, the target access request is processed based on the identifier classification result of the target network addressing identifier recorded in the local database.
[0035] If the target network addressing identifier recorded in the local database is a whitelist identifier, the target access request is forwarded; if the target network addressing identifier recorded in the local database is a blacklist identifier, the target access request is intercepted.
[0036] Step 232: In response to not finding the target network addressing identifier, the target network addressing identifier is sent to an identifier classification model for identifier classification to obtain an identifier classification result of the target network addressing identifier.
[0037] In one possible implementation, after obtaining the network addressing identifier, the router may interact with the identifier classifier in real time to determine the identifier classification result of the network addressing identifier in real time. In this case, the process may be implemented as follows: The target network addressing identifier is sent to the identifier classification model in real time for identifier classification, and an identifier classification result of the target network addressing identifier is obtained, so as to process the target network addressing identifier in real time based on the identifier classification result of the target network addressing identifier.
[0038] Schematically, when interacting with the identification classification model, the router may encapsulate the network addressing identification into an identification classification instruction and send it to the identification classification model. The identification classification instruction is used to instruct the identification classification model to perform identification classification on the network addressing identification it carries.
[0039] Taking an application scenario as an example, after the router is powered on and dials successfully, it monitors the access requests sent by the client in real time. Every time an access request is monitored, it performs an identification classification on the identification classification model. If the identification classification result indicates that the network addressing identifier is a whitelist identifier, the corresponding access request is forwarded. If the identification classification result indicates that the network addressing identifier is a blacklist identifier, the corresponding access request is intercepted and discarded.
[0040] By classifying the received network addressing identifiers in real time, timeliness can be improved, thereby improving the interception effect of illegal network resources. However, since it is necessary to interact with the identifier classification model for classification before processing, the processing of access requests is time-consuming, resulting in a delay in the processing of access requests. The user side may experience an unsmooth and stuck web page experience.
[0041] To improve the user experience, in another possible implementation, the router may periodically interact with the identifier classification model to batch determine the types of received network addressing identifiers. In this case, the process may be implemented as follows: Forwarding target access requests and caching target network addressing identifiers; In response to reaching the target period, each network addressing identifier cached within the target period is sent to the identifier classification model for identifier classification, and an identifier classification result of each network addressing identifier is obtained to update the local database based on the identifier classification result of each network addressing identifier.
[0042] Taking an application scenario as an example, if the target period is 1 day, after the router is powered on and dialed successfully, it monitors the access requests sent by the client in real time, and records the network addressing identifiers in all access requests locally, and makes a classification request to the identifier classification model once a day. During this period, since the router has not recorded the blacklist and whitelist status of the network addressing identifier, the network addressing identifiers within this period are all whitelist identifiers, that is, all access requests within this period are forwarded, and users can access all network resources during this period; after 1 day, the router obtains the blacklist and whitelist status of the network addressing identifiers within the previous day from the identifier classification model. If subsequent access requests carry network addressing identifiers with recorded blacklist and whitelist status, they can be processed based on the query results in the local database. If it is a whitelist identifier, the corresponding access request is forwarded; if it is a blacklist identifier, the corresponding access request is intercepted and discarded.
[0043] That is to say, in the case of periodic interaction, the router can forward the access request in real time after receiving the access request, and the classification of the network addressing identifier in the access request will be performed periodically. Therefore, it will not cause lag on the user side. However, compared with real-time interactive classification, when the user accesses for the first time, the local database does not record the network addressing identifier in the access request and does not obtain the identifier classification result in real time, which may cause the user to access illegal network resources.
[0044] It should be noted that the interaction method with the identification classification model can be selected based on actual needs. If the user's Internet experience is prioritized, periodic interaction can be used. If access security is prioritized, real-time interaction can be used.
[0045] In one possible implementation, after the router device obtains the identification classification result of the network addressing identifier, such as obtaining the network addressing identifier in real time, or periodically obtaining the network addressing identifier, the router can update the local database. Taking the target network addressing identifier as an example, the router can update the local database based on the target network addressing identifier and the identification classification result of the target network addressing identifier, so that after receiving the target network addressing identifier again, the identification classification result of the target network addressing identifier can be determined through the local database in priority to process the target access request corresponding to the target network addressing identifier, thereby improving the processing efficiency of the target access request.
[0046] Since the types of large models used in different regions may be different, in one possible implementation, the type of identification classification model may be determined based on the region to which the client belongs. The process of determining the identification classification model may be implemented as follows: Obtaining network access feature data of the client, the network access feature data including at least one of the following: an IP address, an MCC (Mobile Country Code) and an MNC (Mobile Network Code) of a SIM (Subscriber Identity Module) card; Determine the region to which the client belongs based on network access feature data; Determine the identification classification model based on the region to which the client belongs.
[0047] In another possible implementation, geographical location information of the client may be obtained to determine the region to which the client belongs.
[0048] When determining the identification classification model based on the region to which the client belongs, the router can determine the identification classification model that can be interacted with in the client's current region based on the pre-configured correspondence between the identification classification model and the region, thereby ensuring the feasibility of interaction between the router and the identification classification model.
[0049] To improve the accuracy of identity classification by the representation classification model, in another possible implementation, the router may interact with multiple identity classification models. Different identity classification models may be constructed based on different large models. The identity classification results based on the multiple identity classification models are comprehensively used to determine the identity classification result of the network addressing identity. Taking the target network addressing identity as an example, the process of determining the identity classification result of the network addressing identity can be implemented as follows: Sending the target network addressing identifier to at least two identifier classification models for identifier classification, respectively, to obtain at least two candidate identifier classification results; at least two identifier classification models are obtained by fine-tuning training based on different large models; Based on the at least two candidate identifier classification results and the weights corresponding to the at least two identifier classification models, an identifier classification result of the target network addressing identifier is determined.
[0050] Among them, when comprehensively determining the identification classification result of the network addressing identifier, in a possible implementation, at least two candidate identification classification results can be weightedly scored based on the weights corresponding to different identification classification models to determine the identification classification result of the target network addressing identifier based on the weighted score; schematically, if the candidate identification classification result is a whitelist, the score is 1, and if the candidate identification classification result is a blacklist, the score is 0. After weighted scoring according to the weights corresponding to different identification classification models, if the score value is greater than or equal to 0.5, it is determined to be a whitelist identifier, and if the score value is less than 0.5, it is determined to be a blacklist identifier.
[0051] In another possible implementation, the router can also determine the identification classification result of the network addressing identifier based on the number of candidate identification classification results corresponding to the black and white lists in at least two candidate identification classification results. For example, if the router interacts with 5 different identification classification results and obtains 5 candidate identification classification results, if 3 of them indicate that the network addressing identifier is a whitelist identifier and 2 indicate that the network addressing identifier is a blacklist identifier, then the network addressing identifier is determined to be a whitelist identifier and its corresponding access request is forwarded.
[0052] Furthermore, if multiple identification classification results are inconsistent for the same network addressing identifier, after processing the access request based on the identification classification result of the comprehensively determined network addressing identifier, the access request can also be sent to relevant personnel for further determination, so as to update the local database based on the judgment result feedback from the relevant personnel. That is, after determining that the current network addressing identifier is a whitelist identifier based on at least two candidate identification classification results and forwarding its corresponding access request, if at least two candidate identification classification results are inconsistent, it needs to be sent to relevant personnel for further confirmation. If the judgment result feedback from the relevant personnel is that the network addressing identifier is a blacklist identifier, then when recording the network addressing identifier in the local database, its corresponding identification classification result will be recorded as a blacklist identifier. When the access request is received again, the access request will be intercepted.
[0053] Step 233: Process the target access request based on the identification classification result of the target network addressing identification.
[0054] In a possible implementation, the network addressing identifier includes at least one of the following: an access domain name and an access IP address; The process of processing the target access request based on the identification classification result of the target network addressing identifier can be implemented as follows: In response to the target network addressing identifier including a blacklist identifier, intercepting the target access request; In response to the target network addressing identifier not including the blacklist identifier, the target access request is forwarded.
[0055] That is, when the target network addressing identifier is a domain name identifier or an IP identifier, if the target network addressing identifier is a whitelist identifier, the target access request is forwarded; if the target access addressing identifier is a blacklist identifier, the target access request is intercepted.
[0056] When the target network addressing identifier includes a domain name identifier or an IP identifier, if the identification classification result of either the domain name identifier or the IP identifier is a blacklist identifier, the target access request is intercepted; if both the domain name identifier and the IP identifier are whitelist identifiers, the target access request is forwarded.
[0057] Furthermore, if the identification classification results of the domain name identifier and the IP identifier are different, that is, one of them is a blacklist identifier and the other is a whitelist identifier, after intercepting the corresponding target access request, the target access request can also be sent to relevant personnel for further judgment, and the target network addressing identifier and the corresponding identification classification result are updated to the local database based on the identification classification result indicated by the judgment result of the relevant personnel. That is to say, if the relevant personnel determines that the target access request is a legal request, the target network access identifier and the corresponding identification classification result indicating the whitelist identifier are stored in the local database, so that after receiving the target access request next time, the target access request can be forwarded based on the query result of the local database.
[0058] Among them, when forwarding the target access request, the router can forward the target access request to the downstream server, such as the recursive server. By determining the legitimacy of the access request at the router node, unnecessary interactive signaling can be reduced and the processing pressure of subsequent servers can be reduced.
[0059] To summarize, the access request processing method provided in the embodiment of the present application is that the router extracts the target network addressing identifier from the target access request sent by the client, sends the target network addressing identifier to the identifier classification model for identifier classification, obtains the identifier classification result of the target network addressing identifier, and processes the target access request based on the identifier classification result of the target network addressing identifier, wherein the identifier classification model is obtained by fine-tuning and training based on a large model; by adopting the identifier classification model fine-tuned based on the large model, the router can automatically and dynamically classify and identify the target network addressing identifier, and can automatically identify and intercept various threats including illegal network resources without relying on human intervention, while improving the timeliness and accuracy of network security protection, and significantly reducing the user management and maintenance burden.
[0060] In addition, before interacting with the identification classification model to determine the identification classification result of the network addressing identification, querying the local database first can improve the efficiency of determining the identification classification result.
[0061] For example, a router interacts with an identification classification model in real time, and the network addressing identification of the access request is a domain name identification. Figure 3 A schematic diagram of an access request processing method provided by an exemplary embodiment of the present application is shown, and the method is executed by a router, such as Figure 3 As shown, the process may include the following steps.
[0062] S301, the router is turned on and connected to the Internet.
[0063] S302: Monitor the access request sent by the client in real time, and extract the resolved domain name identifier therefrom.
[0064] S303, query the local database.
[0065] S304, determine whether the domain name identifier is recorded, if so, execute S305, otherwise, execute S306.
[0066] S305, determine whether it is a blacklist identifier, if so, execute S309, otherwise execute S310.
[0067] S306: Send a domain name classification request to the identifier classification model, where the domain name classification request includes the domain name identifier.
[0068] S307: Receive the identification classification result fed back by the identification classification model.
[0069] S308: Update the local database.
[0070] S309: intercept the access request.
[0071] S310: Forward the access request.
[0072] For example, a router interacts with the identification classification model periodically, and the network addressing identification of the access request is a domain name identification. Figure 4 A schematic diagram of an access request processing method provided by another exemplary embodiment of the present application is shown, and the method is executed by a router, such as Figure 4 As shown, the process may include the following steps.
[0073] S401, the router is turned on and connected to the Internet.
[0074] S402: Monitor the access request sent by the client in real time, and extract the resolved domain name identifier therefrom.
[0075] S403, query the local database.
[0076] S404, determine whether the domain name identifier is recorded, if so, execute S405, otherwise, execute S408.
[0077] S405, determine whether it is a blacklist identifier, if so, execute 406, otherwise execute S407.
[0078] S406: intercept the access request.
[0079] S407: Forward the access request.
[0080] S408: Cache the domain name identifier.
[0081] S409: Forward the access request.
[0082] S410: After the target period is reached, a domain name classification request is sent to the identifier classification model. The domain name classification request includes the domain name identifiers cached within the target period.
[0083] S411, receiving the identification classification results of each domain name identification fed back by the identification classification model.
[0084] S412, update the local database.
[0085] In an exemplary embodiment, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement all or part of the steps in the access request processing method described above. The electronic device may be implemented as a router.
[0086] In an exemplary embodiment, a computer-readable storage medium is further provided. The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to implement all or part of the steps in the access request processing method described above. For example, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, or an optical data storage device.
[0087] In an exemplary embodiment, a computer program product is further provided. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to execute to implement the above-mentioned Figures 1 to 4 All or part of the steps of any embodiment shown in any embodiment.
[0088] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered merely as exemplary, and the true scope and spirit of the present application are indicated by the claims.
[0089] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.
Claims
1. A method for processing an access request, characterized in that: The method is performed by a router, and includes: Receive a target access request sent by a client; the target access request includes a target network addressing identifier; the target network addressing identifier is used to indicate the network resource accessed by the target access request; Sending the target network addressing identifier to an identifier classification model for identifier classification to obtain an identifier classification result of the target network addressing identifier; the identifier classification result is used to indicate whether the target network addressing identifier is a blacklist identifier or a whitelist identifier; the identifier classification model is obtained by fine-tuning training based on a large model; The target access request is processed based on the identification classification result of the target network addressing identification.
2. The method according to claim 1, characterized in that Before sending the target network addressing identifier to the identifier classification model for identifier classification, the method further includes: Querying the target network addressing identifier in a local database to obtain a query result of the target network addressing identifier; the local database is used to record the network addressing identifier and the corresponding identifier classification result; The target access request is processed based on the query result of the target network addressing identifier.
3. The method according to claim 2, characterized in that The processing of the target access request based on the query result of the target network addressing identifier includes: In response to querying the target network addressing identifier, processing the target access request based on the identifier classification result of the target network addressing identifier recorded in the local database; In response to the target network addressing identifier not being found, the target network addressing identifier is sent to an identifier classification model for identifier classification to obtain an identifier classification result of the target network addressing identifier.
4. The method according to claim 2 or 3, characterized in that The step of sending the target network addressing identifier to an identifier classification model for identifier classification to obtain an identifier classification result of the target network addressing identifier includes: forwarding the target access request, and caching the target network addressing identifier; In response to reaching the target cycle, each network addressing identifier cached within the target cycle is sent to the identifier classification model for identifier classification, and the identifier classification result of each network addressing identifier is obtained to update the local database based on the identifier classification result of each network addressing identifier.
5. The method according to claim 2 or 3, characterized in that The step of sending the target network addressing identifier to an identifier classification model for identifier classification to obtain an identifier classification result of the target network addressing identifier includes: The target network addressing identifier is sent to the identifier classification model in real time for identifier classification, and an identifier classification result of the target network addressing identifier is obtained, so as to process the target network addressing identifier in real time based on the identifier classification result of the target network addressing identifier.
6. The method according to claim 5, characterized in that The method further comprises: The local database is updated based on the target network addressing identifier and the identifier classification result of the target network addressing identifier.
7. The method according to claim 1, characterized in that The network addressing identifier includes at least one of the following: an access domain name and an access IP address; The processing of the target access request based on the identification classification result of the target network addressing identification includes: In response to the target network addressing identifier including a blacklist identifier, intercepting the target access request; In response to the target network addressing identifier not including a blacklist identifier, forwarding the target access request.
8. The method according to claim 1, characterized in that The method further comprises: Get the client's network access feature data, the network access feature data includes at least one of the following: IP address, user identity module SIM card mobile country code MCC and mobile network code MNC; Determining the region to which the client belongs based on the network access characteristic data; The identification classification model is determined based on the region to which the client belongs.
9. The method according to claim 1, characterized in that The step of sending the target network addressing identifier to an identifier classification model for identifier classification to obtain an identifier classification result of the target network addressing identifier includes: Sending the target network addressing identifier to at least two identifier classification models for identifier classification, respectively, to obtain at least two candidate identifier classification results; the at least two identifier classification models are obtained by fine-tuning training based on different large models; The identification classification result of the target network addressing identification is determined based on the at least two candidate identification classification results and the weights corresponding to the at least two identification classification models.
10. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the access request processing method according to any one of claims 1 to 9.
11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores at least one computer program, and the computer program is loaded and executed by a processor to implement the access request processing method according to any one of claims 1 to 9.
12. A computer program product, characterized in that The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by an electronic device, the electronic device is caused to execute to implement the access request processing method according to any one of claims 1 to 9.