Information processing method and apparatus, device, medium, and product

Abstract

One or more of the scenarios herein provide an information processing method and apparatus, device, medium and product, the method comprising: receiving session credential information, the session credential information being sent by a first service end, the session credential information being used for data transmission between a first client and the first service end; obtaining a first feature, the first feature representing feature information of the first client, and the first feature being related to data transmission between the first client and the first service end; processing the first feature by using a first model to obtain first information, the first information indicating a caching strategy of the first client for the session credential information; and processing the session credential information based on the caching strategy indicated by the first information. In view of the security and resource consumption problems caused by full caching of session credential information, a flexible caching strategy is implemented for the session credential information, the security attack risk caused by full caching is reduced, and the storage resource overhead caused by full caching is reduced.

Description

Technical Field

[0001] One or more of the present invention relates to an information processing method, an information processing apparatus, an electronic device, a computer-readable storage medium, and a computer program product. Background Technology

[0002] Applications (APPs) typically adopt a client-server architecture (C / S architecture). During the operation of the application, the client can initiate data transmission requests to the server, and the server can process the data transmission requests and return response data to the client.

[0003] Data can be transmitted between the client and the server based on network transmission protocols. Therefore, it is particularly important to achieve reliable and efficient data transmission based on network transmission protocols. Summary of the Invention

[0004] This content section is provided to provide a brief overview of the concepts, which will be described in detail in the subsequent Detailed Description section. This content section is not intended to identify key or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0005] This document provides an information processing method for at least one scenario, comprising: receiving session credential information, wherein the session credential information is sent by a first server and is used for data transmission between a first client and the first server; obtaining a first feature, wherein the first feature characterizes feature information of the first client and is related to data transmission between the first client and the first server; processing the first feature using a first model to obtain first information, wherein the first information indicates a caching strategy for the session credential information by the first client; and processing the session credential information based on the caching strategy indicated by the first information.

[0006] This document provides at least one type of information processing apparatus, comprising: a communication module configured to: receive session credential information, wherein the session credential information is sent by a first server and is used for data transmission between a first client and the first server; an acquisition module configured to: acquire a first feature, wherein the first feature characterizes feature information of the first client and is related to data transmission between the first client and the first server; the communication module is further configured to: process the first feature using a first model to obtain first information, wherein the first information indicates a caching strategy for the session credential information by the first client; and a processing module configured to: process the session credential information based on the caching strategy indicated by the first information.

[0007] At least one aspect of this document provides an electronic device comprising: at least one processor; and at least one memory including one or more computer program instructions; wherein the one or more computer program instructions are executed by the processor to perform the information processing method provided by at least one aspect of this document.

[0008] At least one aspect of this document provides a computer-readable storage medium that non-transitory stores computer-readable instructions, wherein the information processing method provided by at least one aspect of this document is implemented when the computer-readable instructions are executed by a processor.

[0009] At least one aspect of this document provides a computer program product, including a computer program that, when executed by a processor, implements the information processing method provided by at least one aspect of this document.

[0010] In one of the information processing methods provided in this paper, for the session credential information received by the client, a first model is used, combined with a first feature related to the current data transmission between the client and the server, to make a real-time decision on the caching strategy for the session credential information. In this way, the client can implement a flexible caching strategy for the session credential information, reduce the security attack risk that may be caused by full caching, reduce the storage resource overhead caused by full caching, and improve the reliability of data transmission between the client and the server. Attached Figure Description

[0011] The above and other features, advantages, and aspects of the various scenarios described herein will become more apparent when taken in conjunction with the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0012] Figure 1 This illustration shows an application scenario diagram of at least one of the information processing methods provided in this paper; Figure 2 The schematic diagram illustrates a flowchart of an information processing method provided in at least one scenario of this article; Figure 3 The schematic diagram illustrates the structure of a first model provided in at least one of the cases described herein; Figure 4 The schematic diagram illustrates a flowchart of an information processing method provided in at least one scenario of this article; Figure 5 The schematic diagram illustrates the structure of an information processing apparatus provided in at least one of the embodiments herein; and Figure 6 A schematic diagram of the structure of an electronic device suitable for implementing at least one of the situations described herein is shown. Detailed Implementation

[0013] One or more scenarios described herein will now be described in more detail with reference to the accompanying drawings. While some scenarios are shown in the drawings, it should be understood that this document can be implemented in various forms and should not be construed as limited to the scenarios set forth herein; rather, these scenarios are provided to provide a more thorough and complete understanding of this document. It should be understood that the accompanying drawings and scenarios are for illustrative purposes only and are not intended to limit the scope of this document.

[0014] It should be understood that the steps described in at least one of the embodiments herein may be performed in different orders and / or in parallel. Furthermore, at least one of the embodiments herein may include additional steps and / or omit the steps shown. The scope of this document is not limited in this respect.

[0015] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one situation" means "at least one situation"; the term "another situation" means "at least one additional situation"; the term "some situations" means "at least some situations". Definitions of other terms will be given in the following description.

[0016] It should be noted that the concepts of "first" and "second" mentioned in this article are only used to distinguish different devices, modules or units, and are not used to limit the order of the functions performed by these devices, modules or units or their interdependencies.

[0017] It should be noted that the terms "one" and "more" used in this document are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0018] The names of messages or information exchanged between multiple devices in at least one scenario described herein are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0019] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition, use, storage or deletion of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0020] It is understood that before using the technical solutions disclosed in each scenario in this article, relevant users should be informed of the type, scope of use, and usage scenarios of the information involved in this article and their authorization should be obtained through appropriate means in accordance with relevant laws and regulations. Relevant users may include any type of rights holder, such as individuals, enterprises, or groups.

[0021] For example, in response to receiving an active request from a user, a prompt message is sent to the relevant user to clearly inform the user that the requested operation will require obtaining and using the user's information, thereby enabling the relevant user to choose whether to provide information to the software or hardware such as electronic devices, applications, servers, or storage media that perform the operation of any of the technical solutions described herein.

[0022] As an optional but non-restrictive implementation, in response to a user's active request, a prompt message can be sent to the user, such as a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide information to the electronic device.

[0023] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation method described in this article. Other methods that comply with relevant laws and regulations may also be applied to the implementation method described in this article.

[0024] Distributed network applications (APPs) typically adopt a client-server architecture (C / S architecture), meaning that an application can be divided into a client and a server.

[0025] A client can be understood as an application instance that faces end users. For example, a client can be deployed on a terminal device and is responsible for receiving user input, rendering the interactive interface, assembling business requests, and initiating data transmission requests to the server. The server can run continuously on a remote host or server cluster and is responsible for receiving data transmission requests sent by the client, processing the data transmission requests, executing business logic, and returning response data to the client after processing.

[0026] During application runtime, the client can generate numerous data transmission requests. These requests need to be forwarded to the server via various network intermediary devices such as switches, routers, and gateways. For example, the client can generate different types of data transmission requests, such as those related to interactive commands, video signaling, log synchronization, and file uploads. All of these different types of data transmission requests need to be forwarded to the server via various network intermediary devices such as switches, routers, and gateways.

[0027] Data transmission between the client and the server can be based on network transmission protocols. Network transmission protocols can be understood as standardized rules that define end-to-end data segmentation, transmission control, error management, traffic scheduling, and session management between the client and the server. For example, network transmission protocols can include Transmission Control Protocol (TCP) and User Datagram Protocol (UDP).

[0028] Normally, a client needs to perform a handshake process with the server to establish a session connection before it can send data to the server. To improve the efficiency of end-to-end data transmission, some network transmission protocols support the reuse of session credential information. For example, after a successful communication connection between the client and the server, the client can cache the session credential information locally. In subsequent data transmission with the server, the client can reuse the locally cached session credential information to send application data directly without having to establish a communication connection again (such as at least one of handshake or key negotiation), thus reducing the handshake process with the server.

[0029] For example, the Quick UDP Internet Connections (QUIC) protocol provides two types of communication procedures: 1-round trip time (1-RTT) communication procedures and 0-round trip time (0-RTT) communication procedures.

[0030] In 1-RTT communication, the client first sends a raw data packet to the server. Then, the server sends a response data packet to the client, completing key negotiation between the client and server in one UDP round trip. The client can then use the negotiated key for data transmission, such as sending application data. The response data packet returned by the server includes session credentials, which the client caches locally for use in subsequent 0-RTT communication.

[0031] In 0-RTT communication, the client first constructs a raw data packet and an application data packet. The raw data packet includes session credential information cached locally on the client, and the application data packet includes the application data to be sent. This application data packet can be encrypted using a pre-shared key derived from the session credential information. The client sends both the raw data packet and the application data packet to the server. The server verifies the session credential information in the raw data packet and then decrypts the application data packet using the pre-shared key derived from the session credential information to process the application data and return the corresponding response data packet to the client. Thus, in 0-RTT communication, the client directly sends application data based on the session credential information, without needing to perform a handshake and key negotiation process with the server before sending the application data.

[0032] However, in practical applications, the method of fully caching session credentials on the client side has at least the following drawbacks: First, since session credentials can derive a pre-shared key, and the derived pre-shared key is fixed, if an attacker detects the application data packet sent by the client during 0-RTT communication, they can repeatedly send the application data packet to the server, causing the server to repeatedly process the same application data, resulting in the risk of replay attacks; Second, the client needs to fully cache session credentials locally, increasing the consumption of storage resources.

[0033] To at least partially solve the above-mentioned technical problem, this paper provides an information processing method in at least one scenario, comprising: receiving session credential information, the session credential information being sent by a first server, the session credential information being used for data transmission between a first client and a first server; obtaining a first feature, the first feature representing feature information of a first client, and the first feature being related to data transmission between the first client and the first server; processing the first feature using a first model to obtain first information, the first information instructing the first client on a caching strategy for the session credential information; and processing the session credential information based on the caching strategy indicated by the first information.

[0034] Based on the information processing method provided in at least one of the embodiments herein, at least one of the embodiments herein also provides an information processing apparatus, an electronic device, a computer-readable storage medium, and a computer program product.

[0035] In one of the information processing methods provided in this paper, for the session credential information received by the client, a first model is used, combined with a first feature related to the current data transmission between the client and the server, to make a real-time decision on the caching strategy for the session credential information. In this way, the client can implement a flexible caching strategy for the session credential information, reduce the security attack risk that may be caused by full caching, reduce the storage resource overhead caused by full caching, and improve the reliability of data transmission between the client and the server.

[0036] The following detailed description, with reference to the accompanying drawings, illustrates one or more scenarios and some examples thereof.

[0037] Figure 1 The illustration shows an application scenario diagram of an information processing method provided in at least one of the cases described in this article.

[0038] like Figure 1 As shown, the application scenario provided in this case can include client 101 and server 102. For example, the client can run on a terminal device, which can be various electronic devices capable of running applications, such as smart wearable devices, smart appliances, smart cars, mobile phones, tablets, laptops, or desktop computers.

[0039] Client 101 can be a client program of an application; that is, client 101 can be understood as an application client. For example, client 101 can provide human-computer interaction functions, such as displaying interactive pages to users and receiving user operations on terminal devices. Client 101 can also perform local basic processing on terminal devices, such as rendering pages and caching data locally on the terminal devices. Client 101 can also assemble business requests, such as generating data transmission requests and initiating communication outwards (e.g., to server 102).

[0040] Server 102 may be the server of the same application as client 101. Server 102 may be deployed in the cloud, such as a cloud server or cloud server cluster; that is, server 102 can be understood as an application server. Server 102 may have the business processing capabilities of the application. For example, server 102 may be responsible for business logic processing, unified data storage and management. In addition, server 102 may also receive data transmission requests sent by client 101, process the data transmission requests, and return corresponding response data to client 101.

[0041] The one or more scenarios described herein do not limit the applications corresponding to client 101 and server 102. For example, the application can be a content platform, a content sharing platform, a shopping platform, etc.

[0042] In one or more of the scenarios described herein, an information processing system 103 is deployed in client 101. For example, the information processing system 103 may be integrated into client 101 in the form of cloud services, plugins, middleware, etc. The information processing system 103 can be understood as a software system with information processing capabilities. For example, the information processing system 103 may decide the caching strategy of client 101 for session credential information.

[0043] For example, the information processing system 103 can integrate a first model (not shown in the figure), and the information processing system 103 can use the first model to process the session credential information received by the client 101 and decide on the caching strategy of the client 101 for the session credential information.

[0044] The information processing methods provided in one or more scenarios described herein can be implemented in software, hardware, firmware, or any combination thereof.

[0045] For example, the information processing methods provided in one or more scenarios herein are applicable to client 101, such as information processing system 103 deployed in client 101. Client 101 (e.g., information processing system 103 deployed in client 101) can load and execute the information processing method. For example, client 101 (e.g., information processing system 103 deployed in client 101) may include a central processing unit (CPU), graphics processing unit (GPU), digital signal processor (DSP), neural network processing unit (NPU), or other forms of processing units with data processing capabilities and / or instruction execution capabilities, storage units, etc. Client 101 (e.g., information processing system 103 deployed in client 101) may also have an operating system and various types of application programming interfaces installed on it, implementing the information processing methods provided in one or more scenarios herein by running code or instructions.

[0046] The following will combine Figures 2 to 4 This paper provides a detailed description of an information processing method for at least one scenario.

[0047] Figure 2 The schematic diagram illustrates a flowchart of an information processing method provided in at least one of the scenarios described herein.

[0048] like Figure 2 As shown, the information processing method for this situation includes steps S201 to S203. The steps of this information processing method are explained below: Step S201: Receive session credential information.

[0049] In one or more contexts herein, the first client can be understood as any application client. For example, the first client can be an application client deployed on the first device, or it can be one or more of the software program, desktop client, and mini-program client installed on the first device.

[0050] The first client can correspond to the same application as the first server. That is, the first server can be understood as the application server, and the first client and the first server are the application client and application server of the same application, respectively.

[0051] Session credential information can be sent by the first server; for example, session credential information can be sent by the first server to the first client.

[0052] Session credentials can be used for data transmission between the first client and the first server. In other words, session credentials can be used for the first client to send application data to the first server. That is, when the first client has a need to transmit data (such as application data) to the first server, for example, when the first client generates a request to transmit application data, the first client can use session credentials to send application data to the first server.

[0053] Application data can be understood as the request data that the first client needs to send to the first server. Application data can come from the business requests of the first client. For example, application data can come from business requests generated by interactive events in the first client, business requests generated by scheduled tasks in the first client, business requests generated by system callbacks or framework callbacks, or business requests generated by the response results of preceding business processes.

[0054] There can be multiple correspondences between application data and business requests. For example, one business request can correspond to a segment of request data (such as application data); another example is that one business request can correspond to multiple segments of request data, which may include application data; yet another example is that multiple business requests can also correspond to a segment of request data (such as application data).

[0055] In other words, the first client can use session credentials to transmit data with the first server without a handshake. Without waiting for a handshake response from the first server or engaging in key negotiation with the first server, the client can directly transmit data (e.g., send application data to the first server) without going through a handshake or key negotiation process.

[0056] For example, session credential information may include a session ticket, and a key may be derived from the session credential information, such as a pre-shared key; for example, the session credential information may encapsulate a key, which the first client can use to encrypt the application data when it needs to send application data to the first server.

[0057] In addition, the session credential information may also include information for verifying the validity of the session credential information, such as the validity period of the session credential information.

[0058] In some cases, receiving session credential information includes: receiving session credential information during the first communication process.

[0059] The first communication process can be used to establish a session connection between the first client and the first server, and the session credential information can be used by the first client to transmit data to the first server during the second communication process.

[0060] Both the first and second communication processes can be understood as the process of data transmission (e.g., sending application data) between the first client and the first server. The first and second communication processes can be two types of communication processes under the same network transmission protocol. The difference between the first and second communication processes is that in the first communication process, the first client needs to establish a session connection with the first server first (e.g., perform a handshake and key exchange) before transmitting data to the first server (e.g., sending application data). In the second communication process, the first client does not need to establish a session connection with the first server first, but can directly transmit data to the first server (e.g., sending application data).

[0061] For example, the first communication process can be the 1-RTT process in the QUIC protocol, and the second communication process can be the 0-RTT process in the QUIC protocol.

[0062] In other words, during the first communication process where the first client establishes a session connection with the first server (e.g., performs a handshake and key exchange), the first server sends session credential information to the first client. The first client can then use this session credential information in the subsequent second communication process for transmitting data to the first server (e.g., sending application data).

[0063] For example, a second communication process can be a communication process following a first communication process. Session credential information can be used by a first client to send application data to a first server in one or more second communication processes; that is, session credential information can be reused in multiple second communication processes.

[0064] Thus, based on the different types of communication processes provided by the network transmission protocol, the information processing method provided in at least one scenario of this paper can be adapted to the original communication process between the first client and the first server, thereby improving the applicability of session credential information.

[0065] Step S202: Obtain the first feature.

[0066] In one or more scenarios described herein, the first feature can be used to characterize the feature information of the first client; that is, the first feature can be understood as a client-side feature associated with the first client. The first feature is related to the data transmission between the first client and the first server. For example, the first feature can characterize the session link parameters between the first client and the first server. In other words, the first feature can be used to characterize the actual data transmission status between the first client and the first server at the current moment (i.e., the moment when the first client receives the session credential information).

[0067] This document does not limit the method of obtaining the first feature in one or more scenarios. For example, the first feature may be obtained from at least one of the following: the network communication link between the first client and the first server, the first device where the first client is located, the first server, and the network intermediate device (e.g., a switch, router, gateway, etc.) between the first client and the first server.

[0068] In other words, for the session credential information sent from the first server to the first client, a first feature related to the data transmission status of the application data transmitted between the first client and the first server is obtained. The first feature is used to represent the actual data transmission status between the first client and the first server. Subsequently, the caching strategy of the session credential information is determined using the first feature.

[0069] In some cases, the first feature may include at least one of the following: request feature, historical behavior feature, and communication quality feature.

[0070] The request features may include characteristic information of the request information, which may be initiated by the first client to the first server. For example, the request information can be understood as a transmission request from the first client to the first server for data transmission. In other words, the request features can characterize the request attributes between the first client and the first server; that is, the request features can be understood as inherent request attribute features unrelated to the application data actually sent by the first client. For example, the request features may be request attribute features related to the entity accessed by the first client to the first server (e.g., a domain name). For example, the request features may include one or more of the following: transport protocol type, idempotency flag, request body attributes, domain name priority, domain name hash, and request method.

[0071] Historical behavioral characteristics may include information on the historical communication behavior between the first client and the first server. That is, historical behavioral characteristics can be understood as the communication behavior characteristics between the first client and the first server within a historical time period. For example, historical behavioral characteristics may include one or more of the following: the number of times the first client accesses the first server's access entity (e.g., a domain name) within a historical time period, and the last time the first client transmitted data to the first server.

[0072] Communication quality characteristics can characterize the communication quality between the first client and the first server. That is, communication quality characteristics can be understood as the transmission quality characteristics of the data transmission from the first client to the first server (e.g., sending application data). For example, communication quality characteristics may include characteristics related to the network communication link (e.g., the current network quality rating, the current TCP round-trip time, the packet loss rate in the most recent set period, etc.) and characteristics related to the first device where the first client is located (e.g., the current memory usage, the current central processing unit (CPU) load, the network type, etc.).

[0073] In this way, the first feature is obtained from multiple different dimensions such as request attributes, historical communication behavior, and communication quality. Furthermore, the first feature can reflect the actual data transmission status (such as the session link status) between the first client and the first server when the first client receives the session credential information, thereby improving the quality and diversity of the first feature and thus improving the accuracy and effectiveness of the decision-making session credential information caching strategy.

[0074] Step S203: Process the first feature using the first model to obtain the first information.

[0075] In one or more of the scenarios described herein, the first information may instruct the first client on its caching strategy for session credential information; that is, the first information may be understood as decision information on whether the first client should cache session credential information.

[0076] The first model can be understood as an artificial intelligence model with the ability to make caching strategy decisions for session credential information. That is, the first model can determine the caching strategy for session credential information of the first client based on the first feature.

[0077] One or more scenarios in this paper do not restrict the form of the first model. For example, the first model can be a deep neural network model including at least one of feedforward neural networks, convolutional neural networks, and recurrent neural networks; for another example, the first model can also be a transformer architecture model based on self-attention mechanism; and for yet another example, the first model can also be a generative model.

[0078] For example, processing the first feature using the first model to obtain the first information includes: sending the first feature to the first model and receiving the first information output by the first model.

[0079] In other words, the first model can be an end-to-end model, in which the first model directly outputs the first information that instructs the first client on the caching strategy for session credential information, thereby improving the decision-making efficiency of the caching strategy for session credential information.

[0080] For example, the caching strategy for session credential information indicated by the first information may include not caching the session credential information and caching the session credential information. When the first information indicates that the first client should cache the session credential information, the first information may further indicate the caching time of the session credential information.

[0081] Step S204: Process the session credential information based on the caching strategy indicated by the first information.

[0082] After using the first model to determine the caching strategy for session credential information, the first client can execute the caching strategy and process the session credential information in accordance with the caching strategy.

[0083] In this way, for the session credential information received by the first client at different times, the caching strategy of the first client for the session credential information is dynamically adjusted according to the current actual data transmission status (such as the session link status between the first client and the first server). Compared with the method of caching the session credential information in its entirety, this method can achieve a balance between improving the efficiency of subsequent data transmission and reducing storage resource overhead, thereby improving the overall benefits of the first client under different data transmission statuses and different session link statuses.

[0084] In some cases, the session credential information is processed based on the caching strategy indicated by the first information, including: in response to the first information indicating the first caching strategy, caching the session credential information on the first client based on the caching time indicated by the first information; or in response to the first information indicating the second caching strategy, discarding the session credential information.

[0085] In other words, the first caching strategy can be understood as a strategy for caching session credential information, while the second caching strategy can be understood as a strategy for not caching session credential information. Under the first caching strategy, since session credential information needs to be cached, the caching time of session credential information can also be specified. That is, within the cached information, the first client can use the session credential information to send application data to the first server.

[0086] Thus, in the case where the first information indicates the first caching strategy, the session credential information is cached according to the caching time so that the first client can use the locally cached session credential information to start the data transmission process with the first server, thereby improving data transmission efficiency; in the case where the first information indicates the second caching strategy, the session credential information is discarded in a timely manner, thereby reducing the storage resources occupied by the first client.

[0087] Furthermore, in response to the first information indicating the first caching policy, and in response to the caching time not exceeding the first information indicating the first caching time, a first message can be generated based on the session credential information and the first application data, and the first message can be sent to the first server.

[0088] If the cache time indicated by the first information has not been exceeded, it can be understood that the session credential information is in a valid state. That is, if the cache time indicated by the first information has not been exceeded, the first client can use the cached session credential information to send one or more application data to the first server.

[0089] The first application data can be understood as the request data that the first client needs to send to the first server, and the first message can be understood as an Internet Protocol (IP) data packet constructed based on session credential information and the first application data. In other words, due to the existence of session credential information, the first client can directly send the first message to the first client, which includes the specific application data that the first client needs to transmit, without first establishing a session connection with the first server.

[0090] For example, the first message may include raw data packets and application data packets from the 0-RTT process of the QUIC protocol. The raw data packets include session credential information, and the application data packets include first application data.

[0091] In this way, the first client does not need to establish a session connection with the first server first, such as handshake and key exchange. It can send the first application data directly to the first server based on the session credential information, thereby improving the data transmission efficiency between the first client and the first server.

[0092] In some cases, the first feature is processed using a first model to obtain first information, including: fusing the first feature to obtain a fused feature, classifying the fused feature to obtain a probability distribution, and outputting the first information based on the probability distribution.

[0093] The fusion of first features can be understood as a feature engineering operation that aggregates multiple feature vectors from different modalities, branches, and representation spaces through normalization operations to generate a unified representation with stronger discriminative power. In other words, compared with the first feature, the fusion of features can eliminate the information limitations of each individual first feature and achieve multi-source information complementarity.

[0094] In some possible implementations, the fusion of the first feature may include one or more of element-wise addition, element-wise multiplication, weighted fusion, and cross-attention fusion, and this paper does not limit the one or more cases.

[0095] The process of classifying the fused features to obtain the probability distribution can be understood as converting the feature vector of the fused features into a class probability distribution. This probability distribution can be used to describe the probability of multiple candidate caching strategies. That is, the probability distribution can be understood as the class posterior probability distribution of multiple candidate caching strategies.

[0096] Since the probability distribution can describe the probability of multiple candidate caching strategies, that is, the probability distribution can measure the likelihood of different candidate caching strategies for the first client's session credential information, the first information can be the candidate caching strategy with the highest probability in the probability distribution. In other words, the caching strategy for the session credential information indicated by the first information for the first client can be the candidate caching strategy with the highest probability.

[0097] Thus, by fully considering that the first feature can include features obtained from different dimensions, aspects, and sources, the first feature can be integrated into a more complete and discriminative fusion feature, thereby improving the modeling capability of the first model and thus enhancing the decision-making effect of the caching strategy for session credential information.

[0098] In some cases, the first feature may include discrete features and continuous features. For example, discrete features may include one or more of domain name hash and network type, while continuous features may include one or more of current TCP round-trip time, packet loss rate of the most recent set period, current memory usage, and current CPU load.

[0099] The first feature is processed using a first model to obtain first information, including: embedding discrete features to obtain a first feature vector, normalizing continuous features to obtain a second feature vector, fusing the first and second feature vectors to obtain a fused feature, classifying the fused feature to obtain a probability distribution, and outputting the first information based on the probability distribution.

[0100] Embedding can be understood as the process of mapping sparse discrete features to a continuous vector space (such as an embedding space). In other words, by embedding discrete features, the discrete features are transformed into dense vectors (i.e., the first feature vector).

[0101] Normalization can be understood as linearly mapping continuous features to a fixed interval, such as mapping continuous features to the range of 0 to 1. That is, by normalizing continuous features, the differences in the dimensions and numerical scales of features of different dimensions are eliminated, and a second feature vector is obtained.

[0102] In this way, by fully considering the discrete or continuous properties of the first feature, different preprocessing methods can be adopted for the first feature with different properties, thereby improving the efficiency and performance of model inference.

[0103] In some cases, discrete features may include at least two types of discrete features. By fusing a first feature vector and a second feature vector, a fused feature is obtained. This includes: performing feature cross processing on at least two types of discrete features based on the first feature vector to obtain a cross feature vector, and concatenating the first feature vector and the second feature vector to obtain a concatenated vector. Finally, the cross feature vector and the concatenated vector are fused to obtain the fused feature.

[0104] Feature crossing can be understood as the process of constructing a combined representation between the original features (i.e., feature vectors corresponding to at least two types of discrete features) to obtain the cross feature vector.

[0105] For example, at least two types of discrete features can include discrete features of domain name hash type and discrete features of network type. That is, based on the first feature vector corresponding to the discrete feature of domain name hash type and the first feature vector corresponding to the discrete feature of network type, feature cross processing is performed on the discrete features of domain name hash type and the discrete features of network type to obtain cross feature vector. In this way, the first model can more easily learn a high-frequency and effective caching strategy for session credential information.

[0106] The concatenation of the first and second feature vectors can be understood as a representation aggregation operation that stacks and merges the first and second feature vectors along the channel dimension or feature dimension.

[0107] Thus, cross feature vectors can capture joint effects that cannot be expressed by a single feature vector (a single first feature vector), improve the ability of a single model to model nonlinear associations, uncover potential association rules, enhance the feature discriminativeness of the first model, and thus improve the decision-making ability of caching strategies. Concatenating vectors can explore unknown combinations of feature vectors, such as enabling the first model to learn caching strategies that yield higher returns when the network environment is poor, thereby improving the inference accuracy of the first model.

[0108] Figure 3 The schematic diagram illustrates the structure of a first model provided in at least one of the cases described herein.

[0109] In some cases, considering that the first model can be deployed in the first client, and that edge inference has low latency and low power consumption performance requirements, the structure of the first model can be as follows: Figure 3 As shown, for example, the first model may include an input layer, a preprocessing layer, a linear module, a neural network module, and an output layer.

[0110] In the input layer, discrete and continuous features are input into the first model. The preprocessing layer may include an embedding layer and a normalization layer. The input of the embedding layer includes discrete features, and the output of the embedding layer includes a first feature vector. For example, the embedding layer can map discrete features to a first feature vector to capture the potential correlation between discrete features. The input of the normalization layer includes continuous features, and the output of the normalization layer includes a second feature vector. The linear module includes a feature cross layer, which includes first feature vectors corresponding to at least two defined features. The output of the feature cross layer includes a cross feature vector. The neural network module includes a concatenation layer and a fully connected layer. The input of the concatenation layer and the fully connected layer includes a first feature vector and a second feature vector. The output of the concatenation layer and the fully connected layer includes a fused feature. The output layer includes an activation function (softmax) layer. The input of the activation function layer includes the fused feature and the cross feature vector. The output of the activation function layer includes a probability distribution. Thus, the first model can output first information indicating the caching strategy for session credential information.

[0111] Thus, the first model can be a wide-deep model, with the linear module in the wide part responsible for memorization and capturing frequently co-occurring patterns in historical data, and the neural network module in the deep part responsible for generalization and connecting the implicit relationships between feature vectors, together improving the inference performance of the first model.

[0112] The training process of the first model will be explained below.

[0113] In some cases, the first model can be trained as follows: multiple training data are obtained from multiple historical communication processes, and the initial model is trained using the training data to obtain the first model.

[0114] The initial model can be understood as the model before training. For example, the initial model can be a randomly initialized model, that is, the initial model can be a model whose weights are generated by random distribution and trained from zero; or the initial model can be a pre-trained model, that is, the initial model can be a model with pre-trained weights loaded.

[0115] Multiple training data can come from multiple historical communication processes. In each historical communication process, the first server sends session credential information to the first client. That is, the historical communication process can be understood as the communication process used to establish a session connection between the first client and the first server. For example, the historical communication process can be the first communication process (e.g., a 1-RTT process).

[0116] Each training data set may include historical features of the historical communication process and transmission result labels corresponding to the session credential information.

[0117] For example, multiple training data include training data A and training data B. Training data A comes from historical communication process A, and training data B comes from historical communication process B. Training data A may include historical features of historical communication process A and transmission result tags corresponding to session credential information in historical communication process A. Training data B may include historical features of historical communication process B and transmission result tags corresponding to session credential information in historical communication process B.

[0118] Historical characteristics of the historical communication process can be used to characterize the data transmission status (e.g., session link status) between the first client and the first server at the moment when the first server sends session credential information to the first client during the historical communication process. For example, historical characteristics of the historical communication process may include one or more of request characteristics, historical behavior characteristics, and communication quality characteristics. Request characteristics, historical behavior characteristics, and communication quality characteristics can be referred to in the previous introduction and will not be repeated here.

[0119] The transmission result label corresponding to the session credential information can characterize the transmission result of the first client based on the session credential information. That is, the transmission result label corresponding to the session credential information can be understood as: after the first client caches the session credential information sent to the first client by the first server in the historical communication process, the first client uses the session credential information to send application data to the first server.

[0120] For example, the transmission result label corresponding to the session credential information can be either transmission success or transmission failure. For instance, when the first client successfully sends application data to the first server using the session credential information (e.g., the 0-RTT process is successful), the transmission result label corresponding to the session credential information can be transmission success; or, for instance, when the first client fails to send application data to the first server using the session credential information (e.g., the 0-RTT process fails), the transmission result label corresponding to the session credential information can be transmission failure.

[0121] In other words, the historical features of the historical communication process in the training data can be understood as the transmission log information of the first client, and the transmission result label corresponding to the session credential information in the training data can represent the transmission status of application data sent with the session credential information under the historical features.

[0122] The initial model is trained in one or more rounds using multiple training data, such as gradient descent training, to obtain the first model. For example, during the training of the first model, the loss function can be the binary cross-entropy loss function, and a regularization term is added to prevent overfitting.

[0123] For example, the training process of the first model can be performed entirely in the cloud. That is, after the first model is trained in the cloud, it is deployed in the first client to realize the collaborative model training and inference process between the end and the cloud.

[0124] In this way, by collecting a large amount of transmission log information from the first client, efficient model training can be achieved, improving the decision-making ability of the first model in terms of caching strategy for session credentials information. This enables the first model to output a reasonable and effective caching strategy for session credentials information based on the current actual data transmission situation.

[0125] Furthermore, in the case where the first information indicates the first caching strategy and the caching time indicated by the first information has not exceeded, the transmission result tag of the first message can also be obtained.

[0126] The transmission result tag of the first message can be used to characterize the transmission result of the first message.

[0127] For example, obtaining the transmission result tag of the first message includes: obtaining information indicating whether the first message was successfully transmitted from the response data packet sent by the first server, and obtaining the transmission result tag of the first message based on the information indicating whether the first message was successfully transmitted.

[0128] For example, the transmission result label of the first message can be either transmission successful or transmission failed. For instance, when the first message is successfully transmitted to the first server (e.g., the 0-RTT process is successful), the transmission result label of the first message can be transmission successful; or, for instance, when the first message is not successfully transmitted to the first server (e.g., the 0-RTT process fails), the transmission result label of the first message can be transmission failed.

[0129] The transmission result labels of the first feature, the first information, and the first message can be used to optimize the first model. For example, the transmission result labels of the first feature, the first information, and the first message can be used as training parameters to update the model parameters of the first model.

[0130] Thus, by combining the actual communication process of the first client and the first server transmitting data without first establishing a session connection, the first model is continuously optimized to continuously improve the model reasoning ability of the first model, making the caching strategy of the session credential information output by the first model more accurate, thereby improving the transmission efficiency of the first client sending application data to the first server.

[0131] Figure 4 The schematic diagram illustrates a flowchart of an information processing method provided in at least one of the scenarios described herein.

[0132] like Figure 4 As shown, the first client includes a local log library and a feature library. After obtaining the first feature, the first feature can be stored as a feature source in the local log library and the feature library. After processing the first feature using the model to obtain the first information, the first information can be stored as a decision record in the local log library and the feature library. After processing the session credential information based on the caching strategy indicated by the first information, the transmission result tag of the first message can be obtained and stored in the local log library and the feature library.

[0133] In this way, the first feature, the first information, and the transmission result tag of the first message are read from the local log library and feature library, the first model is optimized, and the optimized first model is then sent to the first client. This allows the optimized first model to be used to make decisions on the caching strategy of the session credential information after the session credential information is received, thus forming a data processing closed loop.

[0134] Based on the information processing method provided in at least one aspect of this paper, an information processing apparatus is also provided in at least one aspect of this paper. The following will be combined with... Figure 5 A detailed description of the information processing device is provided.

[0135] Figure 5 The schematic diagram illustrates the structure of an information processing device provided in at least one of the present invention.

[0136] like Figure 5 As shown, the information processing apparatus 500 in this scenario includes a communication module 501, an acquisition module 502, and a processing module 503. For example, the communication module 501, acquisition module 502, and processing module 503 can be implemented using hardware (e.g., circuit) modules or software modules, as in the following cases, and will not be described further. For example, the communication module 501, acquisition module 502, and processing module 503 can be implemented using a central processing unit (CPU), a general-purpose computing on graphics processing unit (GPGPU), a graphics processing unit (GPU), a tensor processing unit (TPU), a field-programmable gate array (FPGA), or other forms of processing units with data processing capabilities and / or instruction execution capabilities, along with corresponding computer instructions.

[0137] The communication module 501 is configured to receive session credential information, wherein the session credential information is sent by the first server and is used for data transmission between the first client and the first server. For example, the communication module 501 can be configured to execute step S201 described above; its specific implementation principle can be found in the relevant description of step S201, and will not be repeated here.

[0138] The acquisition module 502 is configured to acquire a first feature, wherein the first feature represents the feature information of the first client, and the first feature is related to the data transmission between the first client and the first server. For example, the acquisition module 502 can be configured to execute step S202 described above; its specific implementation principle can be found in the relevant description of step S202, and will not be repeated here.

[0139] The communication module 501 is further configured to: process the first feature using a first model to obtain first information, wherein the first information indicates the caching strategy of the first client for the session credential information. For example, the communication module 501 can be configured to execute step S203 described above; its specific implementation principle can be found in the relevant description of step S203, and will not be repeated here.

[0140] The processing module 503 is configured to process the session credential information based on the caching strategy indicated by the first information. For example, the processing module 503 can be configured to execute step S204 as described above; its specific implementation principle can be found in the relevant description of step S204, and will not be repeated here.

[0141] In at least one of the embodiments described herein, the first feature includes at least one of the following: a request feature, wherein the request feature includes feature information of request information, the request information being initiated by the first client to the first server; a historical behavior feature, wherein the historical behavior feature includes historical communication behavior information between the first client and the first server; and a communication quality feature, wherein the communication quality feature includes communication quality information between the first client and the first server.

[0142] In at least one of the embodiments described herein, the processing module 503 is further configured to: in response to the first information indicating a first caching strategy, cache the session credential information in the first client based on the caching time indicated by the first information; or in response to the first information indicating a second caching strategy, discard the session credential information.

[0143] In at least one of the embodiments described herein, the communication module 501 is further configured to: in response to the first information indicating a first caching policy and in response to the caching time not exceeding the first information indicating a first caching time, generate a first message based on the session credential information and the first application data, and send the first message to the first server.

[0144] In at least one embodiment of this document, the information processing apparatus 500 further includes an optimization module, the optimization module being configured to: obtain a transmission result tag of the first message, wherein the transmission result tag of the first message is used to characterize the transmission result of the first message; wherein the first feature, the first information, and the transmission result tag of the first message are used to fine-tune the first model.

[0145] In at least one of the embodiments described herein, the communication module 501 is further configured to: receive the session credential information during a first communication process, wherein the first communication process is used for the first client to establish a session connection with the first server, and the session credential information is used for the first client to transmit data to the first server during a second communication process.

[0146] In at least one scenario described herein, the first feature includes discrete features and continuous features. The communication module 501 is further configured to: perform embedding processing on the discrete features to obtain a first feature vector, and perform normalization processing on the continuous features to obtain a second feature vector; fuse the first feature vector and the second feature vector to obtain a fused feature; perform classification processing on the fused feature to obtain a probability distribution, wherein the probability distribution is used to describe the probability of multiple candidate caching strategies; and output the first information based on the probability distribution.

[0147] In at least one of the embodiments described herein, the discrete features include at least two types of discrete features, and the communication module 501 is further configured to: perform feature cross processing on the at least two types of discrete features based on the first feature vector to obtain a cross feature vector, and concatenate the first feature vector and the second feature vector to obtain a concatenated vector; and fuse the cross feature vector and the concatenated vector to obtain the fused feature.

[0148] In at least one scenario described herein, the first model is trained as follows: multiple training data are acquired from multiple historical communication processes, wherein in each of the historical communication processes, the first server sends session credential information to the first client, and each training data includes historical features of the historical communication process and a transmission result label corresponding to the session credential information, wherein the transmission result label characterizes the transmission result of the first client based on the session credential information; using the training data, an initial model is trained to obtain the first model.

[0149] It should be noted that, for clarity and brevity, not all components of the information processing device 500 are shown in at least one of the embodiments herein. To achieve the necessary functions of the information processing device 500, those skilled in the art may provide or configure other components not shown, according to specific needs, and the embodiments herein do not impose any limitations on this.

[0150] The information processing apparatus 500 provided in at least one aspect of this document and the information processing method provided in at least one aspect of this document are based on the same technical concept and can achieve the same technical effect and the same technical purpose as the information processing method provided in at least one aspect of this document. For details, please refer to the relevant descriptions above, which will not be repeated here.

[0151] This document also provides, in at least one embodiment, an electronic device including a processing device and a storage device, the storage device including one or more computer program modules; wherein the one or more computer program modules are stored in the storage device and configured to be executed by the processing device, the one or more computer program modules being used to implement the information processing method provided in any embodiment of this document.

[0152] For example, the processing device may be a processor, such as a central processing unit (CPU), a digital signal processor (DSP), a graphics processing unit (GPU), a general-purpose computing on graphics processing unit (GPGPU), or other forms of processing unit with data processing capabilities and / or instruction execution capabilities. It may be a general-purpose processor or a dedicated processor, and may control other components in the electronic device to perform the desired functions.

[0153] For example, the storage device may be a memory, which may include one or more computer program products. These computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and a processing device may execute these program instructions to implement the functions described in at least one of the embodiments herein (implemented by the processing device) and / or other desired functions. Various application programs and various data may also be stored on the computer-readable storage medium, which is not limited in the embodiments described herein.

[0154] The following is for reference. Figure 6 The diagram illustrates a structural schematic of an electronic device (e.g., a terminal device or a server) 600 suitable for implementing at least one of the embodiments described herein. The terminal device in at least one embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, personal digital assistants (PDAs), portable access devices (PADs), portable media players (PMPs), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital televisions and desktop computers. Figure 6 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of at least one of the situations described herein.

[0155] like Figure 6As shown, the electronic device 600 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory 602 or a program loaded from a storage device 608 into a random access memory 603. The random access memory 603 also stores various programs and data required for the operation of the electronic device 600. The processing unit 601, the read-only memory 602, and the random access memory 603 are interconnected via a bus 604. An input / output interface 605 is also connected to the bus 604.

[0156] Typically, the following devices can be connected to the input / output interface 605: input devices 606 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 607 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 608 including, for example, magnetic tape, hard disk, etc.; and communication devices 609. Communication device 609 allows electronic device 600 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6 An electronic device 600 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0157] In particular, according to one or more embodiments herein, the processes described in the above-referenced flowcharts can be implemented as computer software programs. For example, one or more embodiments herein include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network via communication device 609, or installed from storage device 608, or installed from read-only memory 602. When the computer program is executed by processing device 601, it performs the functions defined in the methods of at least one embodiment herein.

[0158] The electronic device 600 provided in at least one aspect of this document and the information processing method provided in at least one aspect of this document are based on the same technical concept and can achieve the same technical effect and the same technical purpose as the information processing method provided in at least one aspect of this document. For details, please refer to the relevant descriptions above, which will not be repeated here.

[0159] It should be noted that the computer-readable medium mentioned above can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this document, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such transmitted data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, radio frequency (RF), etc., or any suitable combination thereof.

[0160] The computer-readable storage medium provided in at least one aspect of this article and the information processing method provided in at least one aspect of this article are based on the same technical concept and can achieve the same technical effect and the same technical purpose as the information processing method provided in at least one aspect of this article. For details, please refer to the relevant descriptions above, which will not be repeated here.

[0161] In some scenarios, clients and servers can communicate using any currently known or future-developed network protocol, such as the Hypertext Transfer Protocol (HTTP), and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet (e.g., the Internet), and peer-to-peer networks, as well as any currently known or future-developed networks.

[0162] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0163] The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the aforementioned information processing method.

[0164] Computer program code for performing the operations described herein can be written in one or more programming languages, or a combination thereof, including but not limited to object-oriented programming languages, as well as conventional procedural programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0165] One or more embodiments of this document also provide a computer program product comprising one or more computer instructions. When these computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in any of these embodiments are generated.

[0166] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0167] When the computer program product is executed by a computer, the computer performs any of the aforementioned information processing methods. The computer program product can be a software installation package; when any of the aforementioned information processing methods needs to be used, the computer program product can be downloaded and executed on the computer.

[0168] The computer program product provided in at least one of the embodiments described herein and the information processing method provided in at least one of the embodiments described herein are based on the same technical concept and can achieve the same technical effect and the same technical purpose as the information processing method provided in at least one of the embodiments described herein. For details, please refer to the relevant descriptions above, which will not be repeated here.

[0169] The flowcharts and block diagrams in the accompanying figures illustrate the architecture, functionality, and operation of possible implementations of the systems, methods, and computer program products according to the various scenarios described herein. In this respect, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the figures. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0170] The units or modules described in at least one of the scenarios herein can be implemented in software or hardware. The names of the units or modules do not, in some cases, constitute a limitation on the unit or module itself.

[0171] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard parts (ASSPs), systems-on-chips (SoCs), complex programmable logic devices (CPLDs), and so on.

[0172] In the context of this document, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0173] Based on one or more scenarios described herein, Example 1 provides an information processing method, including: Receive session credential information, wherein the session credential information is sent by the first server and is used for data transmission between the first client and the first server; Obtain a first feature, wherein the first feature represents the feature information of the first client, and the first feature is related to the data transmission between the first client and the first server; The first feature is processed using a first model to obtain first information, wherein the first information indicates the caching strategy of the first client for the session credential information; The session credential information is processed based on the caching strategy indicated by the first information.

[0174] According to one or more scenarios in this article, Example 2 provides that the first feature in Example 1 includes at least one of the following: The request features include feature information of the request information, which is initiated by the first client to the first server. Historical behavior characteristics, wherein the historical behavior characteristics include historical communication behavior information between the first client and the first server; Communication quality characteristics, wherein the communication quality characteristics include communication quality information between the first client and the first server.

[0175] Based on one or more scenarios described herein, Example 3 provides the caching strategy based on the first information indication from Example 1 to process the session credential information, including: In response to the first information indicating a first caching strategy, the session credential information is cached in the first client based on the caching time indicated by the first information; or In response to the first information indicating the second caching policy, the session credential information is discarded.

[0176] Depending on one or more scenarios described in this article, Example 4 provides the method from Example 3, and also includes: In response to the first information indicating the first caching policy, and in response to the caching time not exceeding the first information indicating the first caching time, a first message is generated based on the session credential information and the first application data, and the first message is sent to the first server.

[0177] Depending on one or more scenarios described in this article, Example 5 provides the method from Example 4, and also includes: Obtain the transmission result tag of the first message, wherein the transmission result tag of the first message is used to characterize the transmission result of the first message; The first feature, the first information, and the transmission result label of the first message are used to optimize the first model.

[0178] Depending on one or more scenarios described herein, Example 6 provides the receiving session credential information from Example 1, including: During the first communication process, the session credential information is received, wherein the first communication process is used for the first client to establish a session connection with the first server, and the session credential information is used for the first client to transmit data to the first server during the second communication process.

[0179] According to one or more scenarios in this paper, Example 7 provides that the first feature in any of Examples 1 to 6 includes discrete features and continuous features, and processes the first feature using a first model to obtain first information, including: The discrete features are embedded to obtain a first feature vector, and the continuous features are normalized to obtain a second feature vector. The first feature vector and the second feature vector are fused to obtain the fused feature; The fused features are classified to obtain a probability distribution, wherein the probability distribution is used to describe the probability of multiple candidate caching strategies; Based on the probability distribution, the first information is output.

[0180] According to one or more scenarios in this paper, Example 8 provides that the discrete features in Example 7 include at least two types of discrete features, and the fusion of the first feature vector and the second feature vector to obtain the fused feature includes: Based on the first feature vector, feature cross processing is performed on the at least two types of discrete features to obtain a cross feature vector, and the first feature vector and the second feature vector are concatenated to obtain a concatenated vector; The fused feature is obtained by fusing the cross feature vector and the concatenated vector.

[0181] Based on one or more scenarios in this paper, Example 9 provides an example of how the first model from any of Examples 1 through 6 was trained: Multiple training data are obtained from multiple historical communication processes. In each of the historical communication processes, the first server sends session credential information to the first client. Each training data includes historical features of the historical communication process and a transmission result label corresponding to the session credential information. The transmission result label represents the transmission result of the first client based on the session credential information. Using the training data, an initial model is trained to obtain the first model.

[0182] According to one or more of the provisions of this article, Example 10 provides an information processing apparatus, comprising: The communication module is configured to receive session credential information, wherein the session credential information is sent by a first server and is used for data transmission between a first client and the first server. The acquisition module is configured to: acquire a first feature, wherein the first feature represents the feature information of the first client, and the first feature is related to the data transmission between the first client and the first server; The communication module is further configured to: process the first feature using a first model to obtain first information, wherein the first information indicates the caching strategy of the first client for the session credential information; The processing module is configured to process the session credential information based on the caching strategy indicated by the first information.

[0183] According to one or more of the provisions of this document, Example 11 provides an electronic device comprising: At least one processor; and At least one memory, including one or more computer program instructions; The one or more computer program instructions are provided by an information processing method that is executed by a processor at runtime in at least one of the following situations.

[0184] According to one or more of the embodiments described herein, Example Twelve provides a computer-readable storage medium that non-transitory stores computer-readable instructions, wherein the information processing method provided by at least one of the embodiments described herein is implemented when the computer-readable instructions are executed by a processor.

[0185] According to one or more of the embodiments herein, Example Thirteen provides a computer program product including a computer program that, when executed by a processor, implements the information processing method provided by at least one of the embodiments herein.

[0186] The above description is merely a preferred embodiment and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure herein is not limited to technical solutions formed by specific combinations of the above-described technical features, but also includes other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed herein that have similar functions.

[0187] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain contexts, multitasking and parallel processing may be advantageous. Similarly, while some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of this paper. Certain features described in the context of a single case can also be implemented in combination within that single case. Conversely, various features described in the context of a single case can also be implemented individually or in any suitable sub-combination in multiple cases.

[0188] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. An information processing method, comprising: Receive session credential information, wherein the session credential information is sent by the first server and is used for data transmission between the first client and the first server; Obtain a first feature, wherein the first feature represents the feature information of the first client, and the first feature is related to the data transmission between the first client and the first server; The first feature is processed using a first model to obtain first information, wherein the first information indicates the caching strategy of the first client for the session credential information; The session credential information is processed based on the caching strategy indicated by the first information.

2. The method according to claim 1, wherein, The first feature includes at least one of the following: The request features include feature information of the request information, which is initiated by the first client to the first server. Historical behavior characteristics, wherein the historical behavior characteristics include historical communication behavior information between the first client and the first server; Communication quality characteristics, wherein the communication quality characteristics include communication quality information between the first client and the first server.

3. The method according to claim 1, wherein, The caching strategy based on the first information, which processes the session credential information, includes: In response to the first information indicating a first caching strategy, the session credential information is cached in the first client based on the caching time indicated by the first information; or In response to the first information indicating the second caching policy, the session credential information is discarded.

4. The method according to claim 3, further comprising: In response to the first information indicating the first caching policy, and in response to the caching time not exceeding the first information indicating the first caching time, a first message is generated based on the session credential information and the first application data, and the first message is sent to the first server.

5. The method according to claim 4, further comprising: Obtain the transmission result tag of the first message, wherein the transmission result tag of the first message is used to characterize the transmission result of the first message; The first feature, the first information, and the transmission result label of the first message are used to optimize the first model.

6. The method according to claim 1, wherein, The received session credential information includes: During the first communication process, the session credential information is received, wherein the first communication process is used for the first client to establish a session connection with the first server, and the session credential information is used for the first client to transmit data to the first server during the second communication process.

7. The method according to any one of claims 1 to 6, wherein, The first feature includes discrete features and continuous features. The step of processing the first feature using a first model to obtain first information includes: The discrete features are embedded to obtain a first feature vector, and the continuous features are normalized to obtain a second feature vector. The first feature vector and the second feature vector are fused to obtain the fused feature; The fused features are classified to obtain a probability distribution, wherein the probability distribution is used to describe the probability of multiple candidate caching strategies; Based on the probability distribution, the first information is output.

8. The method according to claim 7, wherein, The discrete features include at least two types of discrete features, and the fusion of the first feature vector and the second feature vector to obtain the fused features includes: Based on the first feature vector, feature cross processing is performed on the at least two types of discrete features to obtain a cross feature vector, and the first feature vector and the second feature vector are concatenated to obtain a concatenated vector; The fused feature is obtained by fusing the cross feature vector and the concatenated vector.

9. The method according to any one of claims 1 to 6, wherein, The first model was trained in the following way: Multiple training data are obtained from multiple historical communication processes. In each of the historical communication processes, the first server sends session credential information to the first client. Each training data includes historical features of the historical communication process and a transmission result label corresponding to the session credential information. The transmission result label represents the transmission result of the first client based on the session credential information. Using the training data, an initial model is trained to obtain the first model.

10. An information processing apparatus, comprising: The communication module is configured to receive session credential information, wherein the session credential information is sent by a first server and is used for data transmission between a first client and the first server. The acquisition module is configured to: acquire a first feature, wherein the first feature represents the feature information of the first client, and the first feature is related to the data transmission between the first client and the first server; The communication module is further configured to: process the first feature using a first model to obtain first information, wherein the first information indicates the caching strategy of the first client for the session credential information; The processing module is configured to process the session credential information based on the caching strategy indicated by the first information.

11. An electronic device, comprising: At least one processor; as well as At least one memory, including one or more computer program instructions; The one or more computer program instructions are executed by the processor at any one of claims 1 to 9 as described in the method.

12. A computer-readable storage medium for non-transitory storage of computer-readable instructions, wherein, The method of any one of claims 1 to 9 is implemented when the computer-readable instructions are executed by a processor.

13. A computer program product comprising a computer program that, when executed by a processor, implements the method of any one of claims 1 to 9.

Images