Scene determination method and device based on joint learning platform, equipment and storage medium

Abstract

The present disclosure relates to the technical field of machine learning, and provides a scene determination method and device based on a joint learning platform, equipment and a storage medium. The method comprises: loading scene information corresponding to a preset application scene; obtaining scene demand information sent by a client, and determining whether there is scene information matching the scene demand information; if there is scene information matching the scene demand information, calling a plurality of function programs corresponding to the application scene; when receiving a combination application of the plurality of function programs by the client, establishing a joint learning application community and a corresponding scene strategy corresponding to the application scene; when receiving training data fed back by the client, verifying the training data by using the scene strategy to obtain a training data verification result; and determining a scene of the client in the joint learning application community according to the training data verification result. The present disclosure can help each participant to quickly and accurately find a joint learning scene that matches the business demand.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of machine learning, and particularly relates to a scene determination method and device based on a joint learning platform, equipment and a storage medium. BACKGROUND

[0002] Generally, different business requirements correspond to different joint learning scenes. For example, a business requirement is to predict annual gas consumption, which corresponds to a gas load prediction scene. For another example, a business requirement is to predict annual electricity consumption, which corresponds to an electricity load prediction scene. Usually, the joint learning algorithms, joint types, and required training data designed to solve different business requirements are also different.

[0003] Therefore, each participant who wants to participate in joint learning needs to find a joint learning scene that matches its business requirement and join the joint learning, so as to obtain a joint learning model that meets its expectations, solve its business requirement. The first preliminary work that needs to be done is to find a joint learning scene that matches its business requirement and join.

[0004] However, the prior art has not yet provided a method that can assist each participant to quickly and accurately find a joint learning scene that matches its business requirement. SUMMARY

[0005] Therefore, the embodiments of the present disclosure provide a scene determination method and device based on a joint learning platform, and a storage medium, to provide a method that can assist each participant to quickly and accurately find a joint learning scene that matches its business requirement.

[0006] In a first aspect, the embodiments of the present disclosure provide a scene determination method based on a joint learning platform, comprising:

[0007] loading scene information corresponding to a preset application scene;

[0008] obtaining scene requirement information sent by a client, and determining whether there is scene information matching the scene requirement information;

[0009] if there is scene information matching the scene requirement information, invoking a plurality of function programs corresponding to the application scene;

[0010] when receiving a combination application of the plurality of function programs by the client, establishing a joint learning application community corresponding to the application scene and a corresponding scene strategy;

[0011] when receiving training data fed back by the client, verifying the training data by using the scene strategy to obtain a training data verification result;

[0012] According to a training data verification result, a scenario of the client in the joint learning application community is determined.

[0013] In a second aspect, the embodiment of the present disclosure provides a scenario determination apparatus based on a joint learning platform, comprising:

[0014] A loading module is configured to load scenario information corresponding to a preset application scenario;

[0015] A judging module is configured to obtain scenario demand information sent by the client and judge whether there is scenario information matching the scenario demand information;

[0016] A calling module is configured to call a plurality of function programs corresponding to the application scenario if there is scenario information matching the scenario demand information;

[0017] An establishing module is configured to establish a joint learning application community and a corresponding scenario strategy corresponding to the application scenario when receiving a combination application of the plurality of function programs by the client;

[0018] A verifying module is configured to verify training data by using the scenario strategy to obtain a training data verification result when receiving the training data fed back by the client;

[0019] A scenario determination module is configured to determine the scenario of the client in the joint learning application community according to the training data verification result.

[0020] In a third aspect, the embodiment of the present disclosure provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above method.

[0021] In a fourth aspect, the embodiment of the present disclosure provides a computer readable storage medium, which stores a computer program, wherein the computer program is executed by a processor to implement the steps of the above method.

[0022] The beneficial effects of the embodiments of the present disclosure compared with the prior art at least include: loading scene information corresponding to a preset application scenario; obtaining scene demand information sent by a client, judging whether there is scene information matching the scene demand information; if there is scene information matching the scene demand information, calling a plurality of function programs corresponding to the application scenario; when receiving a combination application of the plurality of function programs by the client, establishing a joint learning application community and a corresponding scene strategy corresponding to the application scenario; when receiving training data fed back by the client, verifying the training data by using the scene strategy to obtain a training data verification result; and determining a scene of the client in the joint learning application community according to the training data verification result, which can well assist each participant to quickly and accurately find a joint learning scene matching the business demand. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0024] Figure 1 is a schematic diagram of an architecture of joint learning provided by the embodiments of the present disclosure;

[0025] Figure 2 is a flowchart of a scene determination method based on a joint learning platform provided by the embodiments of the present disclosure;

[0026] Figure 3 is a structural schematic diagram of a scene determination device based on a joint learning platform provided by the embodiments of the present disclosure;

[0027] Figure 4 is a structural schematic diagram of an electronic device provided by the embodiments of the present disclosure. DETAILED DESCRIPTION

[0028] In the following description, specific details such as specific system structures, techniques, etc. are presented in order to thoroughly understand the embodiments of the present disclosure, but it should be clear to those skilled in the art that the present disclosure can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits and methods are omitted to avoid unnecessary details that hinder the description of the present disclosure.

[0029] Joint learning is to comprehensively use multiple AI (Artificial Intelligence) technologies, and jointly mine data value with multiple parties to give birth to new intelligent formats and modes based on joint modeling. Joint learning has at least the following characteristics:

[0030] (1) In different application scenarios, AI algorithms are screened and / or combined, privacy protection computing is used, and multiple model aggregation optimization strategies are established to obtain high-level and high-quality models.

[0031] (2) Based on multiple model aggregation optimization strategies, an efficiency method for improving the joint learning engine is obtained, wherein the efficiency method can be to improve the overall efficiency of the joint learning engine by solving problems including parallel computing architecture, information interaction under large-scale cross-domain network, intelligent perception, and exception handling mechanism.

[0032] (3) Obtain the needs of multiple users in each scenario, determine the real contribution of each joint participant through a mutual trust mechanism, and allocate incentives.

[0033] Based on the above method, an AI technology ecosystem based on joint learning can be established to fully realize the value of industry data and promote the landing of vertical field scenarios.

[0034] Next, a modular joint learning service platform and system according to an embodiment of the present disclosure will be described in detail with reference to the accompanying drawings.

[0035] Figure 1 is a schematic diagram of a joint learning architecture according to an embodiment of the present disclosure. As shown in Figure 1 , the joint learning architecture can include a service platform 100, a participant 102, a participant 103, and a participant 104. The service platform 100 can include a server (central node) 101.

[0036] In the joint learning process, the basic model can be established by the server 101, and the server 101 sends the model to the participants 102, 103 and 104 which establish communication connection with it. The basic model can also be established by any participant and uploaded to the server 101, and the server 101 sends the model to other participants which establish communication connection with it. The server 101 starts the model training of the participants 102, 103 and 104 using the training data, obtains the updated model parameters, aggregates the model parameters sent by the participants 102, 103 and 104, obtains the global model parameters, and transmits the global model parameters back to the participants 102, 103 and 104. The participants 102, 103 and 104 iterate the respective models according to the received global model parameters until the model converges finally, thereby realizing the training of the model. It should be noted that the number of participants is not limited to the above three, but can be set according to needs, and the embodiments of the present disclosure do not limit this.

[0037] Figure 2 is a flowchart of a scene determination method based on a joint learning platform provided by an embodiment of the present disclosure. Figure 2 The scene determination method based on the joint learning platform can be executed by the server 101. Figure 1 As shown in FIG. 1, the scene determination method based on the joint learning platform includes the following steps. Figure 2

[0038] Step S201, loading scene information corresponding to a preset application scene.

[0039] The preset application scene generally refers to a business scene classified according to a business type. Generally, one business scene corresponds to solving one type (or one) of business problem, for example, the civil building gas use business scene involves the business problem of gas load prediction. For another example, the business scene of commercial electricity involves the business problem of electricity load prediction. Therefore, according to the above classification standard, different businesses can be divided into different business scenes (application scenes).

[0040] It can be understood that the division of application scenes (or joint learning scenes) can also be divided according to other standards, for example, according to enterprise types, specifically, it can be divided into internet enterprise joint learning scene, financial enterprise joint learning scene, medical service enterprise joint learning scene, etc.

[0041] ​The scene information includes the scene name of each application scene (such as "XX load prediction", "soft measurement", etc.), the scene introduction content (such as "using joint learning technology, combining third-party data, expanding the dimensions of employee data, such as employee financial credit and consumption data, etc., to greatly improve the accuracy and credibility of credit assessment", etc.), the joint way used (such as "joint way: horizontal / vertical"), the algorithm used (such as "algorithm used: XGBOOST"), the number of participants (such as "participants: X people", that is, the number of active participants in this scene), the encryption method (such as no encryption, homomorphic encryption, etc.), and the like.

[0042] The name of the application scene can be named according to the corresponding business requirement. For example, for the business requirement of predicting annual electricity load, the name of the joint learning scene can be named as "electricity load prediction". For example, for the business requirement of predicting annual gas load, the name of the joint learning scene can be named as "gas load prediction".

[0043] As an example, a corresponding relationship table of application scene and scene information can be established in advance, as shown in Table 1 below.

[0044] Table 1: Corresponding relationship table of application scene and scene information

[0045]

[0046] In step S202, the scene requirement information sent by the client is obtained, and it is judged whether there is scene information matched with the scene requirement information.

[0047] The scene requirement information usually refers to the application information of a certain application scene that the participant wants to apply to join, for example, the application information of participant A who wants to apply to join the XX application scene.

[0048] As an example, when participant A wants to apply to join application scene 1, the scene requirement information can be sent to the service platform by clicking / touching the preset application scene selection icon displayed on the client interface, etc. For example, assuming that participant A clicks the selection icon of "application scene 1" displayed on the client interface, the requirement information of "application scene 1" can be sent to the service platform via the client. At this time, the service platform can obtain the scene requirement information sent by the client.

[0049] After receiving the scene requirement information sent by the client, the service platform can query and judge whether there is scene information corresponding to "application scene 1" according to Table 1 above.

[0050] In step S203, if there is scene information matched with the scene requirement information, a plurality of function programs corresponding to the application scene are called.

[0051] In combination with the above example, if there is scenario information corresponding to "application scenario 1", the multiple function programs corresponding to "application scenario 1" are invoked.

[0052] The multiple function programs include, but are not limited to, a scenario name display function program, a scenario introduction display function program, a joint mode display function program, an algorithm display function program, and a scenario activity display function program.

[0053] At step S204, when receiving a combination application of the multiple function programs by the client, a joint learning application community corresponding to the application scenario and a corresponding scenario strategy are established.

[0054] In combination with the above example, assuming that a combination application of the scenario name display function program, the scenario introduction display function program, the joint mode display function program, the algorithm display function program, and the scenario activity display function program by the participating party through the client is received, a joint learning application community corresponding to "application scenario 1" is established. In the joint learning application community, the scenario name, the scenario introduction, the joint mode, the algorithm used, and the scenario activity (i.e., the number of participants in the application community) of "application scenario 1" are integrated and displayed, so that the user can understand the business scenario to which the scenario is adapted, thereby better assisting the user to find a joint learning scenario that adapts to the user's business demand and join the joint learning scenario to participate in joint learning, thereby obtaining a joint learning model that meets the user's expectation and further solving the user's business demand.

[0055] As an example, when a combination application of the multiple scenario name display function programs, the scenario introduction display function programs, the joint mode display function programs, the algorithm display function programs, and the scenario activity display function programs by the participating party through the client is received, a joint learning application community corresponding to each application scenario can be established, and all or part of the joint learning application communities can be displayed on the interface, so that the participating party can understand and select a joint learning application community that the participating party wants to join.

[0056] Generally, one joint learning application community (corresponding to one application scenario) can correspond to a data verification rule (i.e., a scenario strategy) for verifying whether the training data fed back by the client matches the application scenario of the joint learning application community.

[0057] At step S205, when receiving the training data fed back by the client, the scenario strategy is used to verify the training data to obtain a training data verification result.

[0058] The training data refers to data used by the participating party to participate in a joint learning task of a joint model corresponding to the joint learning application community.

[0059] The scene strategy (i.e., a preset data verification rule) generally refers to a strategy (rule) for judging whether the training data provided by a participant matches an application scene applied for by the participant.

[0060] In step S206, the scene of the client in the joint learning application community is determined according to the training data verification result.

[0061] As an example, when the training data provided by the participant passes the preset data verification rule, i.e., the training data provided by the participant matches the joint learning application community applied for by the participant, the client of the participant is allowed to join the corresponding application scene in the joint learning application community applied for by the participant (i.e., the application scene of the client in the joint learning application community is determined), so as to jointly train a joint learning task with other participants (clients) in the joint learning application community, thereby obtaining an application model capable of solving the business requirement.

[0062] For example, the training data a provided by the participant A passes the preset scene strategy of gas load prediction, i.e., the training data a provided by the participant A matches the application scene of gas load prediction in the joint learning application community, at this time, A is allowed to join the scene of gas load prediction in the joint learning application community, i.e., the scene of A in the joint learning application community is determined to be the scene of gas load prediction.

[0063] The technical solution provided by the embodiments of the present disclosure can load the scene information corresponding to the preset application scene, obtain the scene requirement information sent by the client, judge whether there is scene information matching the scene requirement information, if there is scene information matching the scene requirement information, call the plurality of function programs corresponding to the application scene, when receiving the combination application of the plurality of function programs by the client, establish the joint learning application community and the scene strategy corresponding to the application scene, when receiving the training data fed back by the client, verify the training data by using the scene strategy to obtain the training data verification result, and determine the scene of the client in the joint learning application community according to the training data verification result, which can help the participants to quickly and accurately find the joint learning scene matching the business requirement.

[0064] In some embodiments, the above step S201 specifically includes:

[0065] Obtaining the scene access request sent by the client, the scene access request including a scene information loading instruction;

[0066] Executing the scene information loading instruction to load the scene information of the application scene corresponding to the scene information loading instruction.

[0067] As an example, the participant can send a scenario access request to the service platform by clicking / touching a preset "application scenario access" icon (which can be a static icon or a dynamic icon (such as a floating icon), etc.) displayed on the client interface of the participant, and the like. At this time, the service platform can receive the scenario access request sent by the participant via the client when it monitors the above-mentioned triggering operation of the participant on the client. For example, the preset "application scenario access" icon can be "scenario access", or "scenario access" with graphic design. The shape, text description, etc. of the specific icon can be flexibly designed according to the actual situation, which is not limited in the present disclosure.

[0068] After receiving the scenario access request sent by the client, the service platform can load the scenario information of the application scenario corresponding to the scenario information loading instruction carried by the request by executing the scenario information loading instruction, so as to obtain the scenario information of the application scenario corresponding to the scenario information loading instruction. For example, the application scenario corresponding to the scenario information loading instruction is application scenario 1, then according to the corresponding relationship between the application scenario and the scenario information in Table 1 above, the scenario information corresponding to the application scenario 1 is obtained, and the scenario information of the application scenario 1 is displayed on the current page of the client, so as to help the participant (user) to view and understand the related scenario information, thereby helping the participant to quickly and accurately find the joint learning scenario that matches the business requirement.

[0069] In some embodiments, the above step S202 specifically comprises:

[0070] extracting at least one key information of the scenario information;

[0071] determining whether the at least one key information contains the same or similar information as the scenario requirement information.

[0072] As an example, assuming that the scenario information of the application scenario 1 includes: the scenario name is "gas load prediction"; the scenario introduction information is "use joint learning technology, combine third-party data, expand the dimension of employee data, such as: employee financial credit and consumption data, etc., greatly improve the accuracy and credibility of credit evaluation..."; the joint mode is "horizontal joint"; the algorithm used is "XGBOOST"; and the number of participants is "97". The key information in the above scenario information can be extracted by text analysis, for example, the existing OCR technology can be used to perform text recognition and analysis on the above scenario information, and the key information can be extracted.

[0073] For example, it is assumed that by performing text analysis on the scenario information of the above-mentioned application scenario 1, the key information extracted is: gas load prediction, employee data expansion, financial credit, consumption data, horizontal joint, and XGBOOST. It is assumed that the participant A clicks the selection icon of "application scenario 1" displayed on the client interface, that is, the client sends the scenario demand information of "application scenario 1" (gas load prediction) to the service platform. At this time, the service platform further determines whether the above-mentioned "gas load prediction, employee data expansion, financial credit, consumption data, horizontal joint, and XGBOOST" key information contains the same or similar information as "application scenario 1" (gas load prediction).

[0074] Among them, the information similar to "application scenario 1" (gas load prediction) can be the application scenario (for example, load prediction) belonging to the same category as "application scenario 1" (gas load prediction) or the description information of the same scenario (for example, gas load prediction) described by different description languages.

[0075] In some embodiments, the above-mentioned step S203 specifically comprises:

[0076] If at least one key information contains the same or similar information as the scenario demand information, at least three of the application scenario name display function program, the scenario introduction display function program, the joint method display function program, the used joint algorithm display function program, or the scenario activity display function program corresponding to the same or similar information of the scenario demand information are called.

[0077] As an example, a corresponding relationship between each application scenario and its corresponding function program can be established in advance, and each application scenario and its function program can be associated and stored according to the corresponding relationship. For example, the corresponding relationship between the application scenario and the function program is shown in Table 2.

[0078] Table 2: Corresponding relationship table of application scenario and function program

[0079]

[0080] In combination with the above example, it is found that the above-mentioned key information contains the same key information "gas load prediction" as "application scenario 1" (gas load prediction), and then at least three of the application scenario name display function program, the scenario introduction display function program, the joint method display function program, the used joint algorithm display function program, or the scenario activity display function program corresponding to the "application scenario 1" (gas load prediction) can be further called according to the above Table 2.

[0081] In some embodiments, the above step, when receiving a combined application of the client to the plurality of function programs, establishes a joint learning application community corresponding to the application scenario and a corresponding scenario strategy, including:

[0082] When receiving a combined application of the client to at least two function programs of the application scenario name display function program, the scene introduction display function program, the joint method display function program, the joint algorithm used display function program or the scene activity display function program, a joint learning application community containing at least two function programs of the combined application is established and a corresponding scenario strategy.

[0083] In combination with the above example, when receiving a combined application of the participating party through its client to the application scenario name display function program, the scene introduction display function program, the joint method display function program, the joint algorithm used display function program and the scene activity display function program corresponding to the application scenario 1, a joint learning application community containing the function programs of the combined application is established, and the scenario strategy of the joint learning application community can be set up at the same time (i.e. the data verification strategy / data verification rule used to verify whether the training data provided by the participating party matches the joint learning application community). In the joint learning application community, the integrated page of the scene name, scene introduction, joint method, joint algorithm used and scene activity corresponding to the application scenario 1 can be displayed to the client, so as to facilitate the participating party (user) to view and understand the relevant scene information, thereby helping the participating party to quickly and accurately find the joint learning scenario that matches the business requirements.

[0084] In some embodiments, the above step S205 specifically includes:

[0085] Providing a data entry template corresponding to the application scenario to the client, the data entry template including data entry requirement instructions and a reference data entry table;

[0086] Receiving training data fed back by the client based on the data entry requirement instructions and the reference data entry table;

[0087] Verifying the training data using the scenario strategy to obtain a training data verification result;

[0088] The above step, according to the training data verification result, determines the scenario of the client in the joint learning application community, including:

[0089] When the verification result is passed, the scenario of the client in the joint learning application community is determined, and the training data is associated with the application scenario and stored.

[0090] As an example, assuming that the scenario requirement information sent by the participant to the service platform via the client is gas load prediction, the service platform establishes a joint learning application community corresponding to the gas load prediction according to the above steps, and the client can be provided with a data entry template corresponding to the gas load prediction. The data entry template has data entry requirement instructions of the training data of the application scenario, including data format, numerical range and other requirements, and a reference data entry table. For example, participant A wants to predict the annual gas load in 20xx, A applies to join the joint learning scenario of gas load prediction, at this time, the data entry template can be provided to participant A. Among them, the data entry template includes the following data entry requirement instructions: ① At least 8 years of consecutive 12 months of monthly gas load is required, and there is no limit to the maximum, as long as the whole year of consecutive monthly gas load is acceptable, and it is best not to appear monthly missing; ② If you want to predict the annual gas load in 20XX, at least the gas load of the whole year before 20XX is required; ③ The monthly gas load is the total gas consumption of the month (note: cumulative value, not instantaneous value), unit: ten thousand square meters, which needs to be provided according to the requirements; ④ The data is mainly used for joint training (training data); ⑤ After preparing the relevant data according to the requirements, upload the data in CSV (comma separated value file format) file format through the "select data" channel.

[0091] Among them, the CSV file can be named according to the naming method of "data.csv", that is, "data name + file format suffix". For example, it can be named as "training data1.csv". As an example, the CSV file includes a list type data entry requirement instruction of "column name", "type", "description" three columns (as shown in Table 3 below), and the data entry table can be as shown in Table 4 below.

[0092] Table 3 Data entry requirement instruction table

[0093]

[0094]

[0095] Table 4 Data entry table

[0096]

[0097] As an example, participant A can prepare data according to the data entry requirement instructions in Table 3 above, and enter the corresponding gas load data in Table 4, and save it as a "training data1.csv" file, and then upload it to the service platform through the "select data" channel. At this time, the service platform can obtain the training data1 provided by participant A.

[0098] Then, the data verification rule (i.e., the scene strategy corresponding to the application scene in the joint learning application community) can be generated according to the above description, and the training data 1 fed back by the participant A is verified. The following will be described in detail according to the data verification rule generated according to the above description ①. First, the key information of "8 years", "12 consecutive months", and "monthly gas load" can be extracted from the above description ①, and the data verification rule "the training data is monthly gas load, and the number of training data is greater than or equal to 96" can be generated according to the key information.

[0099] In combination with the above example, assuming that the training data a uploaded by the user A is 100 monthly gas loads, according to the above data verification rule "the training data is monthly gas load, and the number of training data is greater than or equal to 96", the training data uploaded by the user A is verified, and the verification result of "passing the verification" (or "failing the verification") can be obtained. When the verification result is passing, the client of the participant A is allowed to join the joint learning application community, and the training data a of the user A is saved to the preset storage space of the gas load prediction application scene. At the same time, the user A, the gas load prediction, and the training data a can be recorded in the corresponding column in Table 5, so as to facilitate the subsequent joint learning to find and call the corresponding training data.

[0100] Table 5: Association table of training data and application scene

[0101]

[0102] In subsequent joint learning, the participant can select a training data in a certain joint learning scene to train a certain joint learning task. For example, the participant A applies to join the gas load prediction application scene, and uploads the training data 1 and 2 in the scene. When the A wants to join the joint learning task a in the scene, the training data 1 can be selected as the training data, or the training data 2 can be selected as the training data.

[0103] The technical solution provided by the embodiments of the present disclosure can ensure the association between each training data and the application scene in the joint learning application community by verifying the training data uploaded by the participant via the client, and avoid the problems that the subsequent joint learning cannot be performed or the performance of the joint model obtained by training is poor due to the inadaptation between the training data and the application scene.

[0104] In some embodiments, after determining whether there is scene information matching the scene requirement information, the method further includes:

[0105] When there is no scene information matching the scene requirement information, the preset extended application function program is loaded.

[0106] send scene data collection instructions to the client, and receive scene data fed back by the client according to the data collection instructions;

[0107] process the scene data by using the extended application function program, and generate an extended application scene.

[0108] As an example, it is assumed that the participant A clicks the selection icon of "more scenes to be developed" displayed on the client interface, and sends the demand information of "more scenes to be developed" to the service platform via the client. At this time, the service platform can obtain the scene demand information sent by the client. After detection, it is found that there is no scene information matching the scene demand information of "more scenes to be developed" (i.e., no corresponding scene information can be queried according to Table 1), a preset extended application function program is loaded, and scene data collection instructions are sent to the client. The scene data collection instructions usually refer to instructions including the collection of data such as "scene name, scene introduction, joint mode, training model, and training data". When the client receives the scene data collection instructions, it can collect relevant data according to the instructions and upload them to the service platform. After receiving the scene data fed back by the client, the service platform analyzes and integrates these scene data by using the preset extended application function program, trains the training model by using the training data, obtains an application model, and further determines whether the service platform supports the extended application scene by testing the performance of the application model. If the effect of the application model meets the standard after the above steps, an extended application scene can be further generated.

[0109] All the optional technical solutions described above can be combined to form optional embodiments of the present application, which will not be described one by one here.

[0110] The following is an apparatus embodiment of the present disclosure, which can be used to execute the method embodiments of the present disclosure. For details not disclosed in the apparatus embodiments of the present disclosure, please refer to the method embodiments of the present disclosure.

[0111] Figure 3 is a structural schematic diagram of a scene determination apparatus based on a joint learning platform provided by an embodiment of the present disclosure. As shown in Figure 3 the scene determination apparatus based on the joint learning platform includes:

[0112] The loading module 301 is configured to load scene information corresponding to a preset application scene.

[0113] The judgment module 302 is configured to obtain scene demand information sent by a client, and judge whether there is scene information matching the scene demand information.

[0114] The calling module 303 is configured to call the plurality of function programs corresponding to the application scenario if there is the scene information matching the scene demand information.

[0115] The establishing module 304 is configured to establish the joint learning application community corresponding to the application scenario and the corresponding scene strategy when receiving the combination application of the plurality of function programs by the client.

[0116] The checking module 305 is configured to check the training data by using the scene strategy to obtain the training data checking result when receiving the training data fed back by the client.

[0117] The scene determining module 306 is configured to determine the scene of the client in the joint learning application community according to the training data checking result.

[0118] The technical scheme provided by the embodiments of the present disclosure can load the scene information corresponding to the preset application scenario by configuring the loading module 301, acquire the scene demand information sent by the client by configuring the judging module 302, judge whether there is the scene information matching the scene demand information, configure the calling module 303 to call the plurality of function programs corresponding to the application scenario if there is the scene information matching the scene demand information, configure the establishing module 304 to establish the joint learning application community corresponding to the application scenario when receiving the combination application of the plurality of function programs by the client, configure the checking module 305 to check the training data by using the scene strategy to obtain the training data checking result when receiving the training data fed back by the client, and configure the scene determining module 306 to determine the scene of the client in the joint learning application community according to the training data checking result, so that the joint learning scene adapted to the business demand of each participant can be quickly and accurately found.

[0119] In some embodiments, the loading module 301 described above includes:

[0120] The obtaining unit is configured to obtain the scene access request sent by the client, and the scene access request includes the scene information loading instruction.

[0121] The execution unit is configured to execute the scene information loading instruction to load the scene information of the application scenario corresponding to the scene information loading instruction.

[0122] In some embodiments, the judging module 302 described above includes:

[0123] The extraction unit is configured to extract at least one key information of the scene information.

[0124] The judging unit is configured to judge whether the at least one key information contains the same or similar information as the scene demand information.

[0125] In some embodiments, the calling module 303 comprises:

[0126] The calling unit is configured to call at least three of the application scenario name display function program, the scene introduction display function program, the joint mode display function program, the used joint algorithm display function program, or the scene activity display function program corresponding to the information same or similar to the scene demand information.

[0127] In some embodiments, the establishing module 304 comprises:

[0128] The establishing unit is configured to establish a joint learning application community and a corresponding scene strategy containing at least two function programs of the combined application when receiving a combined application of at least two function programs of the application scenario name display function program, the scene introduction display function program, the joint mode display function program, the used joint algorithm display function program, or the scene activity display function program.

[0129] In some embodiments, the verification module 305 comprises:

[0130] The providing unit is configured to provide the client with a data entry template corresponding to the application scenario, and the data entry template comprises data entry requirement instructions and a reference data entry table.

[0131] The receiving unit is configured to receive training data fed back by the client based on the data entry requirement instructions and the reference data entry table.

[0132] The verification unit is configured to verify the training data using the scene strategy to obtain a training data verification result.

[0133] The scene determination module 306 comprises:

[0134] The scene determination unit is configured to determine the scene of the client in the joint learning application community when the verification result is passed, and store the training data in association with the application scenario.

[0135] In some embodiments, the device further comprises:

[0136] The extension loading module is configured to load a preset extension application function program when there is no scene information matching the scene demand information.

[0137] The acquisition module is configured to send a scene data acquisition instruction to the client and receive scene data fed back by the client according to the data acquisition instruction.

[0138] The scene extension module is configured to process the scene data by using an extension application function program to generate an extended application scene.

[0139] It should be understood that the size of the serial number of each step in the above embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure.

[0140] Figure 4 is a schematic diagram of the electronic device 400 provided by the embodiments of the present disclosure. As shown in the figure, the electronic device 400 of the embodiments includes a processor 401, a memory 402, and a computer program 403 stored in the memory 402 and executable on the processor 401. The processor 401 implements the steps in each of the above method embodiments when executing the computer program 403. Alternatively, the processor 401 implements the functions of each module / unit in each of the above device embodiments when executing the computer program 403. Figure 4

[0141] By way of example, the computer program 403 can be divided into one or more modules / units, which are stored in the memory 402 and executed by the processor 401 to complete the present disclosure. One or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program 403 in the electronic device 400.

[0142] The electronic device 400 can be a desktop computer, a notebook computer, a palm computer, a cloud server, and the like. The electronic device 400 can include but is not limited to the processor 401 and the memory 402. Those skilled in the art can understand that the electronic device 400 can include more or fewer components, or combine certain components, or different components, for example, the electronic device can also include an input / output device, a network access device, a bus, and the like. Figure 4 The electronic device 400 is only an example and does not constitute a limitation on the electronic device 400, and can include more or fewer components than the figure, or combine certain components, or different components, for example, the electronic device can also include an input / output device, a network access device, a bus, and the like.

[0143] ​The processor 401 can be a central processing unit (CPU), or other general purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or the like. The general purpose processor can be a microprocessor, or the processor can be any conventional processor, etc.

[0144] The memory 402 can be an internal storage unit of the electronic device 400, for example, a hard disk or a memory of the electronic device 400. The memory 402 can also be an external storage device of the electronic device 400, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, or the like equipped on the electronic device 400. Further, the memory 402 can include both the internal storage unit and the external storage device of the electronic device 400. The memory 402 is used to store computer programs and other programs and data required by the electronic device. The memory 402 can also be used to temporarily store data that has been output or will be output.

[0145] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for easy distinction, and do not limit the protection scope of the present application. The specific working process of the unit and module in the above system can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0146] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in detail in a certain embodiment can be referred to the related description of other embodiments.

[0147] Those skilled in the art can understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present disclosure.

[0148] In the embodiments provided by the present disclosure, it should be understood that the disclosed apparatus / equipment and method can be implemented in other ways. For example, the apparatus / equipment embodiments described above are merely schematic, for example, the division of the modules or units is merely a logical function division, and there can be another division manner in actual implementation, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0149] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment.

[0150] In addition, each functional unit in each embodiment of the present disclosure can be integrated into one processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0151] The integrated modules / units, if implemented in the form of software functional units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by instructing related hardware through a computer program, and the computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be implemented. The computer program can include computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electric carrier signal and telecommunication signal.

[0152] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than limit them; although the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure.

Claims

1. A method for determining a scene based on a joint learning platform, characterized in that, The method comprises the following steps: loading scene information corresponding to a preset application scenario; obtaining scene demand information sent by a client, and judging whether there is scene information matched with the scene demand information; if there is scene information matched with the scene demand information, calling a plurality of function programs corresponding to the application scenario; when receiving a combination application of the plurality of function programs by the client, establishing a joint learning application community corresponding to the application scenario and a corresponding scene strategy; when receiving training data fed back by the client, verifying the training data by using the scene strategy to obtain a training data verification result; determining a scene of the client in the joint learning application community according to the training data verification result; when receiving a combination application of at least two function programs of the application scenario name display function program, the scene introduction display function program, the joint mode display function program, the used joint algorithm display function program or the scene activity display function program by the client, establishing a joint learning application community comprising the at least two function programs of the combination application and a corresponding scene strategy; when receiving training data fed back by the client, verifying the training data by using the scene strategy to obtain a training data verification result, comprising: providing a data entry template corresponding to the application scenario to the client, wherein the data entry template comprises data entry requirement instructions and a reference data entry table; receiving training data fed back by the client based on the data entry requirement instructions and the reference data entry table; verifying the training data by using the scene strategy to obtain a training data verification result; when the verification result is passed, determining the scene of the client in the joint learning application community, and storing the training data in association with the application scenario; after obtaining the scene demand information sent by the client and judging whether there is scene information matched with the scene demand information, further comprising: when there is no scene information matched with the scene demand information, loading a preset extended application function program; sending a scene data collection instruction to the client, and receiving scene data fed back by the client according to the data collection instruction; processing the scene data by using the extended application function program to generate an extended application scenario. The method comprises the following steps: obtaining a scene access request sent by a client, wherein the scene access request comprises a scene information loading instruction; 2. The method of claim 1, wherein, executing the scene information loading instruction to load scene information of an application scenario corresponding to the scene information loading instruction. The method comprises the following steps: ​ 3. The method of claim 1, wherein, ​ extracting at least one key information of the scene information; determining whether the at least one key information contains the same or similar information as the scene requirement information.

4. The method of claim 3, wherein, If the scene information matches the scene requirement information, the application scene corresponding function programs are invoked, including: If the at least one key information contains the same or similar information as the scene requirement information, at least three function programs of the application scene name display function program, the scene introduction display function program, the joint mode display function program, the used joint algorithm display function program or the scene activity display function program corresponding to the same or similar information as the scene requirement information are invoked. 5.A scenario determination apparatus based on a joint learning platform, the apparatus employing the method of any one of claims 1-4, characterized in that, The device comprises: a loading module configured to load scene information corresponding to a preset application scene; a determining module configured to obtain scene requirement information sent by a client and determine whether there is scene information matching the scene requirement information; an invoking module configured to invoke a plurality of function programs corresponding to the application scene if there is scene information matching the scene requirement information; an establishing module configured to establish a joint learning application community and a corresponding scene strategy corresponding to the application scene when receiving a combination application of the plurality of function programs by the client; a verifying module configured to verify training data fed back by the client using the scene strategy to obtain a training data verification result; a scene determining module configured to determine a scene of the client in the joint learning application community according to the training data verification result.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the steps of the method in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: receiving a request for a resource from a client; determining whether the client is authorized to access the resource; and if the client is authorized to access the resource, providing the resource to the client. The computer program is executed by the processor to realize the steps of the method in any one of claims 1 to 4.

Images