Network data processing method and apparatus, device and storage medium

Abstract

Provided in the embodiments of the present disclosure are a method and apparatus used for network data processing, a device and a storage medium. The method comprises: acquiring respective reference data corresponding to a group of network scenarios, the reference data reflecting performances of the corresponding network scenarios; next, for a first scenario in the group of network scenarios, selecting from the reference data at least one piece of positive sample data associated with the first scenario and at least one piece of negative sample data associated with other scenarios in the group of network scenarios; and then, generating a comparison model on the basis of the at least one piece of positive sample data and the at least one piece of negative sample data, the comparison model describing an association relationship between network scenarios and comparison features of the network scenarios, and the comparison model enabling the similarity between a comparison feature and another comparison feature of another network scenario to indicate whether the network scenario and the other network scenario are the same network scenario.

Description

Network data processing method, apparatus, device, and storage medium

[0001] This application claims priority to the Chinese patent application No. 202411313544.3, filed on September 19, 2024, entitled “Network data processing method, apparatus, device, and storage medium”, the entire content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The example embodiments of the present disclosure generally relate to the field of computer technology, and in particular, to a method, apparatus, device, and computer readable storage medium for network data processing. BACKGROUND

[0003] Currently, a network provides various services, such as communication services, for users. The network is usually dynamic. Sometimes, the changes of the network will have an impact on the services provided by the network. Different network scenarios can have different change trends and different impacts on the provided services. It is desirable to be able to optimize the impact on the services caused by the change trends of various network types. SUMMARY

[0004] In a first aspect of the present disclosure, a network data processing method is provided. The method comprises: obtaining reference data corresponding to each of a set of network scenarios, the reference data reflecting performance of the corresponding network scenario; for a first scenario in the set of network scenarios, selecting at least one positive sample data associated with the first scenario and at least one negative sample data associated with other scenarios in the set of network scenarios from the reference data; and generating a contrast model based on the at least one positive sample data and the at least one negative sample data, the contrast model describing an association between a network scenario and a contrast feature of the network scenario, the contrast model making similarity between the contrast feature and another contrast feature of another network scenario indicative of whether the network scenario and the another network scenario belong to the same network scenario.

[0005] In a second aspect of the present disclosure, a network data processing apparatus is provided. The apparatus comprises: an obtaining module configured to obtain reference data corresponding to each of a set of network scenarios, the reference data reflecting performance of the corresponding network scenario; a data selection module configured to, for a first scenario in the set of network scenarios, select at least one positive sample data associated with the first scenario and at least one negative sample data associated with other scenarios in the set of network scenarios from the reference data; and a generation module configured to generate a contrast model based on the at least one positive sample data and the at least one negative sample data, the contrast model describing an association between a network scenario and a contrast feature of the network scenario, the contrast model making similarity between the contrast feature and another contrast feature of another network scenario indicative of whether the network scenario and the another network scenario belong to the same network scenario.

[0006] In a third aspect of the disclosure, an electronic device is provided. The device includes at least one processor; and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor. The instructions, when executed by the at least one processor, cause the device to perform the method of the first aspect.

[0007] In a fourth aspect of the disclosure, a computer-readable storage medium is provided. The medium has stored thereon computer-executable instructions that, when executed by a processor, implement the method of the first aspect.

[0008] In a fifth aspect of the disclosure, a computer program product is provided. The computer program product is tangibly stored in a computer storage medium and includes computer- executable instructions. The computer-executable instructions, when executed by a device, cause the device to perform the method of the first aspect.

[0009] It is to be understood that the details set forth in this section are not intended to limit key or critical features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0010] The above and other features, aspects, and advantages of embodiments of the present disclosure will become more apparent from the following detailed description in conjunction with the accompanying drawings, in which like reference numerals denote like elements, wherein:

[0011] FIG. 1 illustrates a schematic diagram of an example environment in which embodiments of the present disclosure can be implemented;

[0012] FIG. 2 illustrates a block diagram of a process for training a contrast model, according to some embodiments of the present disclosure;

[0013] FIG. 3 illustrates a block diagram of a process for training a classification model, according to some embodiments of the present disclosure;

[0014] FIG. 4 illustrates a block diagram of a process for processing network data, according to some embodiments of the present disclosure;

[0015] FIG. 5 illustrates a flowchart of a method for processing network data, according to some embodiments of the present disclosure;

[0016] FIG. 6 illustrates a block diagram of an apparatus for network data processing, according to some embodiments of the present disclosure; and

[0017] FIG. 7 illustrates a block diagram of a device capable of implementing embodiments of the present disclosure. DETAILED DESCRIPTION

[0018] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be implemented in various forms and should not be interpreted as being limited to the embodiments set forth herein, but rather, these embodiments are provided so as to more completely and thoroughly understand the present disclosure. It is understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0019] In the description of embodiments of the present disclosure, the term "comprising" and its conjugations should be understood as open-ended, i.e., "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "an embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit or implicit definitions can also be included below.

[0020] In this document, unless explicitly stated, performing a step "in response to A" does not mean that the step is performed immediately after A, but can include one or more intermediate steps.

[0021] It can be understood that the data involved in the technical solutions of the present disclosure (including but not limited to the data itself, the obtaining or use of the data) should comply with the requirements of relevant laws and regulations and relevant provisions.

[0022] It can be understood that, before using the technical solutions disclosed in the embodiments of the present disclosure, the type of personal information involved in the present disclosure, the scope of use, the scenario of use, etc. should be informed to the user and the authorization of the user should be obtained through appropriate means according to relevant laws and regulations.

[0023] For example, in response to receiving the active request of the user, a prompt information is sent to the user to explicitly prompt the user that the operation requested to be performed will require obtaining and using the personal information of the user, so that the user can voluntarily choose whether to provide the personal information to the software or hardware such as electronic device, application program, server or storage medium, etc. performing the operation of the technical solutions of the present disclosure according to the prompt information.

[0024] As an optional but non-limiting implementation manner, in response to receiving the active request of the user, the manner of sending the prompt information to the user may, for example, be the manner of pop-up window, and the prompt information may, for example, be presented in the form of text in the pop-up window. In addition, the pop-up window may also carry selection controls for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0025] It can be understood that the above notification and user authorization process is only illustrative and does not limit the implementation manner of the present disclosure, and other manners meeting the relevant laws and regulations can also be applied to the implementation manner of the present disclosure.

[0026] As briefly described above, network dynamics can have an impact on network services, such as network real-time communication. Such network dynamics are a pain point problem in real-time communication. In order to optimize network services, it is necessary to classify networks. Some conventional solutions attempt to classify networks based on objective indicators of networks, based on manual experience, so as to perform targeted optimization according to network types. However, these conventional solutions usually manually define network classification rules based on manual experience. Due to various dynamic changes of networks, manually defined rules are difficult to cover all network types. Based on manually defined rules, a mapping relationship between network objective indicators and network types is established, and potential relationships within network indicators cannot be mined. For scenarios at the boundary of adjacent network types, it is difficult for such conventional solutions to distinguish.

[0027] To address the above and other potential problems, embodiments of the present disclosure propose a solution for network data processing. In the solution, reference data corresponding to a set of network scenarios is obtained, and the reference data reflects performance of the corresponding network scenarios. Next, for a first scenario in the set of network scenarios, at least one positive sample data associated with the first scenario and at least one negative sample data associated with other scenarios in the set of network scenarios are selected from the reference data. Then, a contrast model is generated based on the at least one positive sample data and the at least one negative sample data. The contrast model describes an association between a network scenario and a contrast feature of the network scenario. The contrast model makes similarity between the contrast feature and another contrast feature of another network scenario indicate whether the network scenario and the another network scenario belong to the same network scenario.

[0028] The solution of the present disclosure can use the contrast model to extract contrast features of data of a network scenario. Through such contrast learning of the contrast model, contrast feature representations of unified network scenario types can be made as close as possible, and contrast feature representations of different network scenario types can be made as far apart as possible. The contrast features thus derived can more easily distinguish features of different network types, so as to more accurately classify networks based on these features.

[0029] Some example embodiments of the present disclosure will be described below with reference to the accompanying drawings.

[0030] FIG. 1 shows a block diagram of an example environment 100 in which implementations of the present disclosure can be implemented. In the environment 100 of FIG. 1, it is desirable to train and use a model (i.e., a network data processing model 130) that is configured for classifying networks based on network data. For example, networks can be classified according to different network scenarios based on communication data of the networks. As shown in FIG. 1, the environment 100 includes a model training system 150 and a model application system 152. The upper part of FIG. 1 shows the process of the model training stage, and the lower part shows the process of the model application stage. Before training, the parameter values of the network data processing model 130 can have initial values, or can have pre-trained parameter values obtained through a pre-training process. After the training process, the parameter values of the network data processing model 130 can be updated and adjusted. After the training is completed, the network data processing model 130’ can be obtained. At this time, the parameter values of the network data processing model 130’ have been updated, and based on the updated parameter values, the network data processing model 130 can be used to implement a prediction task in the model application stage.

[0031] In the model training stage, the network data processing model 130 can be trained based on a reference data set 110 including a plurality of reference data 112, and by using the model training system 150. In this article, “reference data” is also referred to as “training data”, and “reference data set” is also referred to as “training data set”. Here, each reference data 112 can reflect the performance of a corresponding network scenario. That is, different reference data 112 can correspond to different network categories (also referred to as network scenario categories).

[0032] Specifically, the training process can be iteratively performed by using a large amount of reference data. After the training is completed, the network data processing model 130 can include knowledge about identifying the network scenario category of the network data. In the model application stage, the network data processing model 130’ (at this time, the prediction model, i.e., the network data processing model 130’ has the trained parameter values) can be invoked by using the model application system 152. For example, input data 140 (including network data to be processed or network data to be classified) can be received, and the network scenario category 144 corresponding to the input data 140 is output.

[0033] In FIG. 1, the model training system 150 and the model application system 152 can include any computing system with computing capability, such as various computing devices / systems, terminal devices, servers, etc. The terminal devices can involve any type of mobile terminal, fixed terminal, or portable terminal including a mobile handset, a desktop computer, a laptop computer, a notebook computer, a netbook computer, a tablet computer, a media player, a multimedia tablet, or any combination thereof, including accessories and peripherals of same or any combination thereof. The servers include, but are not limited to, mainframes, edge computing nodes, computing devices in a cloud environment, etc.

[0034] It should be understood that the structure and function of the environment 100 are described for illustrative purposes only and not to imply any limitation on the scope of the present disclosure. It should be understood that the components and arrangements of the environment 100 shown in FIG. 1 are merely examples and a computing system suitable for implementing the example implementations described in the present disclosure can include one or more different components, other components, and / or different arrangements. For example, while shown as separate, the model training system 150 and the model application system 152 can be integrated in the same system or device. Implementations of the present disclosure are not limited in this regard. Example implementations of model training and model application will be described below with continued reference to the drawings.

[0035] According to the scheme of the present disclosure, in order to better determine the network scene category to which the network data corresponds, a contrast model can be utilized to determine the contrast features of the network data. In this way, the contrast features of the same network scene and different network scenes can be better learned, and the contrast features of the same network scene are made as close as possible in the feature space, and the contrast features of different network scenes are made as far away as possible in the feature space. Thus, each network scene can be better distinguished.

[0036] FIG. 2 shows a block diagram 200 of a process for training a contrast model according to some embodiments of the present disclosure. The process can be implemented at the model training system 150. The contrast model training process of FIG. 2 will be described in conjunction with FIG. 1.

[0037] In the embodiment of FIG. 2, the network data processing model 130 includes a contrast model 220. The contrast model 220 can employ a contrast learning network structure that is currently known and / or will be developed in the future, such as any structure employing a Transformer stack or other appropriate contrast learning network structure. As described in FIG. 2, the model training system 150 obtains reference data corresponding to each of a set of network scenes. The reference data reflects the performance of the corresponding network scene.

[0038] In some embodiments, the set of network scenarios can include a packet loss scenario, a jitter scenario, a delay scenario, a buffer bloat scenario, an out-of-order scenario, and the like. The buffer bloat scenario can include a large buffer bloat scenario and / or a small buffer bloat scenario. In some embodiments, the network scenario can be a combined scenario of any of the above network scenarios or other suitable network scenarios.

[0039] In some embodiments, for each network scenario in the set of network scenarios, a network impairment instrument can be utilized to construct or simulate the corresponding network scenario, and in turn collect communication data of the corresponding network scenario as reference data. The network impairment instrument can be a software network impairment instrument, a hardware network impairment instrument, or a combination thereof. The network impairment instrument can employ any currently known and / or to be developed in the future manner to simulate network data or network communication data under different network scenarios. Embodiments of the present disclosure are not limited herein. Alternatively, in some embodiments, the reference data can be collected from networks actually operated under different scenarios.

[0040] As an example, the communication data of the network scenario constructed by the network impairment instrument and / or the communication data under the actual network scenario can be periodically collected as reference data in the reference data set in a predetermined time length (e.g., 100 milliseconds or other suitable time length) as a period. Alternatively, the reference data can be collected at predetermined time points in a non-fixed time length. These reference data can include feature data (also referred to as index data) of the corresponding network scenario. Examples of the feature data include, but are not limited to, a packet loss rate, a packet loss rate after recovery, a sending code rate (e.g., an actual sending code rate), a receiving code rate (e.g., an actual receiving code rate), jitter data, an out-of-order amplitude, a network round-trip delay (RTT), a delay gradient, a data packet arrival interval (e.g., a minimum value, an average value, or a maximum value of the data packet arrival interval), and the like. A piece of reference data d i may be represented as (1):

[0041] wherein ATT i represents a set of attribute values of the reference data d i , S i represents a corresponding network scenario category of the reference data d i , and i represents an index of the reference data d i in the reference data set. The entire reference data set can be represented as D: {d0, d1,..., di,..., d N N, where N is an integer greater than or equal to 1.

[0042] For a certain scenario (e.g., a first scenario) in the set of network scenarios, the model training system 150 selects at least one positive sample data 212 associated with the first scenario and at least one negative sample data 214 associated with other scenarios in the set of network scenarios from the reference dataset 110. That is, data belonging to the same network type can be labeled as positive samples, and data not belonging to the same network type can be labeled as negative samples. Assuming that the first scenario is a packet loss scenario, the at least one positive sample data 212 can include communication data under the packet loss scenario. The at least one negative sample data 214 can include communication data under other network scenarios (e.g., a jitter scenario, a delay scenario, a buffer bloat scenario, an out-of-order scenario, etc.).

[0043] In some embodiments, the positive and negative samples can be selected in batches. Assuming that the size of each batch is k (k is an integer greater than or equal to 1), k / 2 samples can be sampled each time, and one sample from the corresponding category is selected for each sample.

[0044] The model training system 150 generates a contrastive model 220 based on the at least one positive sample data 212 and the at least one negative sample data 214. The contrastive model 220 describes the association between a network scenario and a contrastive feature of the network scenario. The contrastive model 220 makes the similarity between the contrastive feature of the network scenario and another contrastive feature of another network scenario indicative of whether the network scenario and the other network scenario belong to the same network scenario.

[0045] In some embodiments, a loss function of the contrastive model 220 can be generated based on the positive sample data 212 and the negative sample data 214, and the contrastive model 220 can be trained based on the loss function. Specifically, with the contrastive model 220, a contrastive feature of the first scenario, at least one positive contrastive feature of the at least one positive sample data 212, and at least one negative contrastive feature of the at least one negative sample data 214 can be determined, respectively. Based on the contrastive feature, the at least one positive contrastive feature, and the at least one negative contrastive feature, the loss function of the contrastive model 220 can be generated.

[0046] As an example, the loss function of the contrastive model 220 can be determined using the following equation (2):

[0047] wherein l represents the loss, q represents the output of the reference data after being encoded by f, k + represents the output of the positive sample data after being encoded by f, k i represents the output of the positive sample data after being encoded by f, k + represents the set of positive samples, |q + represents the number of positive samples, and f represents the function of the contrastive model 220.

[0048] It should be appreciated that the loss function shown in Equation (2) is merely exemplary and not limiting. The loss function of Equation (2) can be adjusted or modified. Other suitable loss functions for contrastive learning can be employed to train the contrastive model 220. Embodiments of the present disclosure are not limited in this regard.

[0049] In some embodiments, the loss function can be minimized by training to obtain a trained contrastive model. In one example embodiment, a threshold value of the loss function can be set. If the loss function value of the contrastive model 220 is less than the threshold value, the training of the contrastive model 220 can be ended. In other example embodiments, a number of training times can be set. In the case where a predetermined number of iterative training of the contrastive model 220 is completed, the training of the contrastive model 220 can be ended.

[0050] In some embodiments, the feature extraction granularity of the contrastive model can be adjusted by setting different loss function threshold values or setting different training times. For example, setting a larger loss function or fewer training times can cause the contrastive model 220 to generate contrastive features with a larger granularity. Conversely, setting a smaller loss function or more training times can cause the contrastive model 220 to generate contrastive features with a smaller granularity. Specifically, for two network scene types at the boundary of adjacent network scene categories, the contrastive model 220 can distinguish the contrastive features of the two similar network scenes by a smaller loss function or more training times.

[0051] In this document, the trained contrastive model 220 is also denoted as a contrastive model 220’. With the contrastive model 220’, contrastive features (also referred to as contrastive feature representations or feature representations) of network data can be generated. For example, by utilizing the trained contrastive model 220’, the contrastive feature representation of the network data d i , i.e., f(Att i ) can be obtained. The contrastive feature representations of data belonging to the same network scene type are closer in the feature space. The smaller distance represents greater similarity. Conversely, a larger distance in the feature space represents smaller similarity. If the contrastive feature representations of data of two network scenes are farther apart in the feature space, the two network scenes belong to different network scene categories.

[0052] The technical solution can determine the contrast features of different network scene data by using the contrast model. In this way, the contrast features of the same network scene category and different network scene categories can be better learned, and the contrast features of the same network scene are made as close as possible in the feature space, and the contrast features of different network scenes are made as far away as possible in the feature space. Thus, each network scene category can be better distinguished. By using the exemplary implementation of the present disclosure, more discriminative and consistent contrast features can be provided, thereby ensuring accurate identification of each network scene category.

[0053] In some embodiments, the network data processing model 130 can further include a classification model for determining the category of the network scene based on the contrast features of the network data determined by the contrast model 220. FIG. 3 shows a block diagram 300 of a process for training a classification model 330 according to some embodiments of the present disclosure. The training process of the classification model 330 can be implemented at the model training system 150 in FIG. 1. The classification model training process of FIG. 3 will be described in conjunction with FIG. 1.

[0054] In some embodiments, a set of reference contrast features 322 of a set of network scenes can be determined based on the respective reference data 112 of the set of network scenes by using the trained contrast model 220’. Next, a classification model 330 for network scene classification can be generated based on the set of reference contrast features 322 of the set of network scenes. The ith piece of data used to generate the classification model 330 can be represented as {f(Att l ),S l}, where l is an integer greater than or equal to 1, f(Att l ) represents the contrast feature representation of the ith piece of reference data derived by the trained contrast model 220’, and S l represents the network scene category of the ith piece of reference data. The entire data set used to generate the classification model 330 can be represented as where N is an integer greater than or equal to 1.

[0055] In the context of the present disclosure, the classification model 330 can be implemented based on various models that are currently known and / or will be developed in the future. As an example, the classification model 330 can be created using any modeling method such as a support vector machine (svm), a deep neural network (DNN), etc. The parameter values of the classification model 330 can be initialized. The process of generating the classification model 330 based on the set of reference contrast features 322 is also referred to as the training process of the classification model 330. That is, the parameter values of the classification model 330 are determined through the training process.

[0056] In particular, with the classification model 330, a set of predicted network scenes corresponding to the set of reference contrast features 322 can be determined respectively. Based on a comparison between the network scene categories 332 of the set of network scenes and the set of predicted network scenes, a value of a loss function of the classification model 330 can be determined. Based on the value of the loss function, the classification model 330 can be trained.

[0057] In some embodiments, a corresponding loss function can be selected according to a type or modeling method of the selected classification model 330. As an example, the loss function can be mean square error, cross-entropy, etc. It should be understood that any appropriate loss function can be employed to train the classification model 330. Embodiments of the present disclosure are not limited in this regard.

[0058] Through training, the loss function of the classification model 330 can be minimized, thereby obtaining a relationship between the contrast feature representation (e.g., the feature vector f(Att l )) and the network scene type. The trained classification model 330 is also denoted as classification model 330’ herein. By utilizing the contrast model 220’ and the classification model 330’, the network scene type corresponding to different network data can be determined.

[0059] FIG. 4 illustrates a block diagram 400 of a process for processing network data according to some embodiments of the present disclosure. The network data processing process of FIG. 4 can be implemented at the model application system 152 in FIG. 1. The network data processing model 130’ employed in the model application system 152 includes the trained contrast model 220’ and the trained classification model 330’ described above.

[0060] In particular, for the network data to be processed or the network data to be classified, e.g., the input data 140, the trained contrast model 220’ can be utilized to determine the target contrast feature 422 of the network data. Further, based on the target contrast feature 422, the trained classification model 330’ can be utilized to determine the network scene, e.g., the network scene category 144, corresponding to the network data to be processed, e.g., the input data 140.

[0061] In this way, by utilizing the trained contrast model, the contrast features of the network data under different network scenarios can be determined. Then, based on the contrast features, the network category corresponding to the network data can be determined by utilizing the classification model. This network scenario classification method based on contrast learning is suitable for network services such as real-time communication. In this way, network types that cannot be covered by artificially defined network classification rules can be classified, and in addition, this contrast learning method can learn the potential relationship in the network data, such as the network indicator data, well, thereby improving the network classification accuracy. For the scenario of adjacent network type boundaries that are difficult to classify by artificial classification, the present disclosure can also classify them well.

[0062] FIG. 5 shows a flowchart of a method 500 for network data processing, according to some embodiments of the present disclosure. The method 500 can be performed, for example, by the network data processing device 110 in FIG. 1 or other appropriate device.

[0063] At block 510, the network data processing device 110 obtains reference data corresponding to each of a set of network scenarios. The reference data reflects the performance of the corresponding network scenario.

[0064] At block 520, the network data processing device 110 selects, for a first scenario in the set of network scenarios, at least one positive sample data associated with the first scenario and at least one negative sample data associated with other scenarios in the set of network scenarios from the reference data.

[0065] At block 530, the network data processing device 110 generates a contrast model based on the at least one positive sample data and the at least one negative sample data. The contrast model describes the association between the network scenario and the contrast feature of the network scenario. The contrast model makes the similarity between the contrast feature and another contrast feature of another network scenario indicate whether the network scenario and the another network scenario belong to the same network scenario.

[0066] In some example embodiments, generating the contrast model includes: determining, by utilizing the contrast model, a contrast feature of the first scenario, at least one positive contrast feature of the at least one positive sample data, and at least one negative contrast feature of the at least one negative sample data, respectively; generating a loss function of the contrast model based on the contrast feature, the at least one positive contrast feature, and the at least one negative contrast feature; and training the contrast model based on the loss function.

[0067] In some example embodiments, the method 500 further includes: determining feature data associated with the first scenario, the feature data including at least one of: a packet loss rate, a sending code rate, a receiving code rate, jitter data, a disorder amplitude, a network round-trip delay, a delay gradient, and a data packet arrival interval; and updating the loss function based on the feature data.

[0068] In some example embodiments, the method 500 further includes: determining, based on the reference data corresponding to each of the set of network scenarios, a set of reference contrast features of the set of network scenarios by using the trained contrast model; and generating a classification model for network scenario classification based on the set of reference contrast features of the set of network scenarios.

[0069] In some example embodiments, generating the classification model includes: determining, by using the classification model, a set of predicted network scenarios corresponding to the set of reference contrast features, respectively; generating a loss function of the classification model based on a comparison between the set of network scenarios and the set of predicted network scenarios; and training the classification model based on the loss function of the classification model.

[0070] In some example embodiments, the method 500 further includes: determining, by using the trained contrast model, a target contrast feature of the network data to be processed; and determining, by using the trained classification model, a network scenario corresponding to the network data to be processed based on the target contrast feature.

[0071] In some example embodiments, the set of network scenarios includes at least one of: a packet loss scenario, a jitter scenario, a delay scenario, a buffer bloat scenario, and an out-of-order scenario.

[0072] In some example embodiments, obtaining the reference data corresponding to each of the set of network scenarios includes: constructing, by using a network impairment instrument, a respective network scenario for each network scenario in the set of network scenarios; and collecting communication data of the respective network scenario as the reference data.

[0073] In some example embodiments, FIG. 6 illustrates a block diagram of an apparatus 600 for network data processing, according to some embodiments of the present disclosure. The apparatus 600 can be implemented as or included in the network data processing device 110 of FIG. 1. Various modules / components in the apparatus 600 can be implemented by hardware, software, firmware, or any combination thereof.

[0074] As shown, the apparatus 600 includes an obtaining module 610 configured to obtain reference data corresponding to respective network scenarios of a set of network scenarios. The reference data reflects performance of the respective network scenarios. The apparatus 600 further includes a data selecting module 620 configured to, for a first scenario of the set of network scenarios, select, from the reference data, at least one positive sample data associated with the first scenario and at least one negative sample data associated with other scenarios of the set of network scenarios. The apparatus 600 further includes a generating module 630 configured to generate, based on the at least one positive sample data and the at least one negative sample data, a contrast model. The contrast model describes an association between a network scenario and a contrast feature of the network scenario. The contrast model makes similarity between the contrast feature and another contrast feature of another network scenario indicative of whether the network scenario and the another network scenario belong to a same network scenario.

[0075] In some example embodiments, the generating module 630 can be configured to: determine, using the contrast model, respectively, a contrast feature of the first scenario, at least one positive contrast feature of the at least one positive sample data, and at least one negative contrast feature of the at least one negative sample data; generate, based on the contrast feature, the at least one positive contrast feature, and the at least one negative contrast feature, a loss function of the contrast model; and train the contrast model based on the loss function.

[0076] In some example embodiments, the apparatus 600 further includes: a first determining module configured to determine feature data associated with the first scenario, the feature data comprising at least one of: a packet loss rate, a sending code rate, a receiving code rate, jitter data, an out-of-order amplitude, a network round-trip delay, a delay gradient, a packet inter-arrival interval; and an updating module configured to update the loss function based on the feature data.

[0077] In some example embodiments, the apparatus 600 further includes: a second determining module configured to determine, based on the respective reference data of the set of network scenarios, using the trained contrast model, a set of reference contrast features of the set of network scenarios; and a second generating module configured to generate, based on the set of reference contrast features of the set of network scenarios, a classification model for network scenario classification.

[0078] In some example embodiments, the second generating module is configured to: determine, using the classification model, respectively, a set of predicted network scenarios corresponding to the set of reference contrast features; generate, based on a comparison between the set of network scenarios and the set of predicted network scenarios, a loss function of the classification model; and train the classification model based on the loss function of the classification model.

[0079] In some example embodiments, the apparatus 600 further includes a third determining module configured to determine, for the network data to be processed, a target contrast feature of the network data to be processed by utilizing the trained contrast model; and a fourth determining module configured to determine, based on the target contrast feature, a network scenario corresponding to the network data to be processed by utilizing the trained classification model.

[0080] In some example embodiments, the set of network scenarios includes at least one of the following: a packet loss scenario, a jitter scenario, a delay scenario, a bufferbloat scenario, an out-of-order scenario.

[0081] In some example embodiments, the obtaining module 610 can be configured to, for each network scenario in the set of network scenarios, construct the corresponding network scenario by utilizing a network impairments instrument; and collect communication data of the corresponding network scenario as the reference data.

[0082] FIG. 7 illustrates a block diagram of an electronic device 700 in which one or more embodiments of the disclosure can be implemented. It should be understood that the electronic device 700 illustrated in FIG. 7 is merely exemplary and should not be construed as limiting the functionality and scope of the embodiments described herein. The electronic device 700 illustrated in FIG. 7 can be used to implement the network data processing device 110 of FIG. 1 or the apparatus 600 for network simulation of FIG. 6.

[0083] As shown in FIG. 7, the electronic device 700 is in the form of a general computing device. Components of the electronic device 700 can include, but are not limited to, one or more processors or processing units 710, a memory 720, a storage device 730, one or more communication units 740, one or more input devices 750, and one or more output devices 760. The processing unit 710 can be a real or virtual processor and is capable of executing various processing in accordance with programs stored in the memory 720. In a multi-processor system, multiple processing units execute computer-executable instructions in parallel to improve the parallel processing capability of the electronic device 700.

[0084] The electronic device 700 typically includes a number of computer storage media. Such media can be any available media that is accessible by the electronic device 700 and includes both volatile and non-volatile media, removable and non-removable media. The memory 720 can be a volatile memory (e.g., registers, cache, random access memory (RAM)), a non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. The storage device 730 can be a removable or non-removable media and can include a machine-readable medium, such as a flash drive, a disk drive, or any other medium that can be used to store information and / or data and that can be accessed by the electronic device 700.

[0085] The electronic device 700 can further include additional detachable / non-detachable, volatile / non-volatile storage media. Although not shown in FIG. 7, a disk drive for reading from or writing to a detachable, non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a detachable, non-volatile optical disk (e.g., a CD-ROM) can be provided. In these cases, each drive can be connected to the bus (not shown) by one or more data media interfaces. The memory 720 can include a computer program product 725 having one or more program modules configured to carry out the various methods or acts of the various embodiments of the present disclosure.

[0086] The communication unit 740 enables communication with other electronic devices through communication media. Additionally, the functionality of the components of the electronic device 700 can be implemented in a single computing cluster or a plurality of computer machines capable of communicating over a communication connection. As such, the electronic device 700 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or another network nodes in the networking environment.

[0087] The input device 750 can be one or more input devices, such as a mouse, a keyboard, a trackball, etc. The output device 760 can be one or more output devices, such as a display, a speaker, a printer, etc. The electronic device 700 can also communicate with one or more external devices (not shown) such as a storage device, a display device, etc., one or more devices that enable a user to interact with the electronic device 700, or any devices (e.g., a network card, a modem, etc.) that enable the electronic device 700 to communicate with one or more other electronic devices, through the communication unit 740, as desired. Such communication can be carried out via an input / output (I / O) interface (not shown).

[0088] According to an example implementation of the present disclosure, a computer readable storage medium having computer executable instructions stored thereon is provided, where the computer executable instructions are executed by a processor to implement the method described above. According to an example implementation of the present disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer readable medium and includes computer executable instructions, where the computer executable instructions are executed by a processor to implement the method described above.

[0089] Various aspects of the disclosure are now described with reference to the drawings. In general, the drawings described herein relate to a method, an apparatus, a device, and a storage medium implemented according to the present disclosure. It will be understood that each block of the flowchart diagrams and / or block diagrams, and combinations of blocks in the flowchart diagrams and / or block diagrams, can be implemented by computer readable program instructions.

[0090] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processing unit of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0091] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0092] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutive 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 action, or using a combination of dedicated hardware and computer instructions.

[0093] Various implementations of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is chosen to best explain the principles, practical applications, or improvements to technology in the market, or to enable others skilled in the art to understand the various implementations disclosed herein.

Claims

1. A method for network data processing, comprising: obtaining reference data corresponding to each of a set of network scenarios, the reference data reflecting performance of the corresponding network scenario; for a first scenario in the set of network scenarios, selecting at least one positive sample data associated with the first scenario and at least one negative sample data associated with other scenarios in the set of network scenarios from the reference data; and generating a contrast model based on the at least one positive sample data and the at least one negative sample data, the contrast model describing a relationship between a network scenario and a contrast feature of the network scenario, the contrast model making similarity between the contrast feature and another contrast feature of another network scenario indicative of whether the network scenario and the another network scenario belong to a same network scenario.

2. The method of claim 1, wherein generating the contrast model comprises: determining, respectively, a contrast feature of the first scenario, at least one positive contrast feature of the at least one positive sample data, and at least one negative contrast feature of the at least one negative sample data using the contrast model; generating a loss function of the contrast model based on the contrast feature, the at least one positive contrast feature, and the at least one negative contrast feature; and training the contrast model based on the loss function.

3. The method of claim 2, further comprising: determining feature data associated with the first scenario, the feature data comprising at least one of: packet loss rate, sending code rate, receiving code rate, jitter data, out-of-order amplitude, network round-trip delay, delay gradient, packet inter-arrival interval; and updating the loss function based on the feature data.

4. The method of any one of claims 1 to 3, further comprising: determining a set of reference contrast features of the set of network scenarios using the trained contrast model based on the reference data of each of the set of network scenarios; and generating a classification model for network scenario classification based on the set of reference contrast features of the set of network scenarios.

5. The method of claim 4, wherein generating the classification model comprises: determining, respectively, a set of predicted network scenarios corresponding to the set of reference contrast features using the classification model; generating a loss function of the classification model based on comparison between the set of network scenarios and the set of predicted network scenarios; and training the classification model based on the loss function of the classification model.

6. The method of claim 4, further comprising: determining a target contrast feature of network data to be processed using the trained contrast model; and determining a network scenario corresponding to the network data to be processed using the trained classification model based on the target contrast feature.

7. The method of any one of claims 1 to 6, wherein the set of network scenarios comprises at least one of: a packet loss scenario, a jitter scenario, a delay scenario, a buffer bloat scenario, an out-of-order scenario. ​ ​ ​ 8. The method of any one of claims 1-7, wherein obtaining reference data corresponding to each of a set of network scenarios comprises: for each network scenario in the set of network scenarios, constructing the respective network scenario with a network loss instrument; and collecting communication data of the respective network scenario as the reference data.

9. A network data processing apparatus comprising: an obtaining module configured to obtain reference data corresponding to each of a set of network scenarios, the reference data reflecting performance of the respective network scenario; a data selection module configured to, for a first scenario in the set of network scenarios, select from the reference data at least one positive sample data associated with the first scenario and at least one negative sample data associated with other scenarios in the set of network scenarios; and a generating module configured to generate, based on the at least one positive sample data and the at least one negative sample data, a contrast model that describes an association between a network scenario and a contrast feature of the network scenario, the contrast model such that a similarity between the contrast feature and another contrast feature of another network scenario indicates whether the network scenario and the another network scenario belong to a same network scenario.

10. An electronic device comprising: at least one processor; and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, the instructions, when executed by the at least one processor, cause the electronic device to perform the method of any one of claims 1-8.

11. A computer-readable storage medium having computer-executable instructions stored thereon that are executable by a processor to implement the method of any one of claims 1-8.

12. A computer program product tangibly stored in a computer storage medium and comprising computer-executable instructions that, when executed by a device, cause the device to perform the method of any one of claims 1-8. ​

Images