Data processing method and apparatus, and electronic device

Abstract

The present disclosure relates to a data processing method, device and electronic equipment, and belongs to the technical field of live broadcast. The method comprises: obtaining to-be-processed data of a service subject; determining subject reference features of multiple dimensions corresponding to the service subject; processing the to-be-processed data by using the subject reference features of the multiple dimensions to obtain first processing results corresponding to the dimensions of each subject reference feature; and processing the first processing results according to service reference features to obtain second processing results corresponding to the service subject. Through this method, the overall data processing effect of the service subject can be effectively improved, the data processing result is more accurate, and the referenceability of the data processing result is effectively improved.

Description

Technical Field

[0001] This disclosure relates to the field of live streaming technology, and in particular to a data processing method, apparatus, and electronic device. Background Technology

[0002] With the development of live video streaming technology, especially in the live e-commerce sector, streamers provide services based on the public / private traffic of service platforms. Some service providers (merchants) hope to leverage influencer traffic to offer services through methods such as live chat, which can easily lead to non-compliant services provided by service platforms. Therefore, it is necessary to evaluate the service effectiveness of these providers.

[0003] In related technologies, when processing data related to service providers to evaluate the service effectiveness of these providers, the method is usually based on consumer subjective ratings (Detail Seller Rating, DSR).

[0004] In this approach, the reference factors are relatively subjective and the data processing dimensions are limited, which affects the data processing effect on the data related to the service subject, resulting in data processing results that are not objective and accurate enough and have weak reference value. Summary of the Invention

[0005] This disclosure provides a data processing method, apparatus, electronic device, storage medium, and computer program product to at least solve the technical problem in the related art that the reference elements are relatively subjective and the data processing dimensions are limited, thereby affecting the data processing effect on the data related to the service subject, resulting in data processing results that are not objective and accurate enough and have weak reference value.

[0006] According to a first aspect of the present disclosure, a data processing method is provided, the method comprising: acquiring data to be processed of a service subject; determining multiple dimensions of subject reference features corresponding to the service subject; processing the data to be processed using the multiple dimensions of subject reference features to obtain a first processing result corresponding to each dimension of the subject reference features; and processing the first processing result according to service reference features to obtain a second processing result corresponding to the service subject.

[0007] In some embodiments of this disclosure, the step of processing the first processing result according to the service reference features to obtain a second processing result corresponding to the service subject includes:

[0008] For each dimension of the subject reference feature, determine the reference description information corresponding to the subject reference feature;

[0009] The first processing result is processed according to the reference description information to obtain a second processing result corresponding to the service subject.

[0010] In some embodiments of this disclosure, determining the subject reference features of multiple dimensions corresponding to the service subject includes:

[0011] Determine the main reference features for multiple candidate dimensions;

[0012] Based on the service reference features, at least some of the candidate dimensions' subject reference features are selected from the multiple candidate dimensions' subject reference features as the corresponding dimension's subject reference features.

[0013] In some embodiments of this disclosure, selecting at least a portion of the candidate dimension's subject reference features as the corresponding dimension's subject reference features from the plurality of candidate dimension subject reference features based on the service reference features includes:

[0014] A correlation analysis is performed on the service reference features and the subject reference features of the candidate dimensions to obtain the correlation results;

[0015] From the subject reference features of the multiple candidate dimensions, the subject reference features of the candidate dimensions whose correlation results satisfy the reference conditions are selected as the subject reference features of the corresponding dimensions.

[0016] In some embodiments of this disclosure, selecting the subject reference features of the candidate dimensions whose relevance results satisfy the reference conditions from the subject reference features of the plurality of candidate dimensions as the subject reference features of the corresponding dimension includes:

[0017] From the subject reference features of the multiple candidate dimensions, the subject reference features of the candidate dimensions that satisfy the first reference condition with positive correlation are selected as the subject reference features of the corresponding dimensions, and the subject reference features of the candidate dimensions that satisfy the second reference condition with negative correlation are selected as the subject reference features of the corresponding dimensions.

[0018] In some embodiments of this disclosure, determining the reference description information corresponding to the subject reference feature for each dimension includes:

[0019] Construct a positive sample set and a negative sample set. The positive sample set includes: main reference features of candidate dimensions with positive correlation satisfying the first reference condition. The negative sample set includes: main reference features of candidate dimensions with negative correlation satisfying the second reference condition.

[0020] The positive sample set, the negative sample set, and the service reference features are input into a pre-trained network model to obtain reference description information output by the network model corresponding to the subject reference features, wherein the network model is used to obtain the reference description information.

[0021] In some embodiments of this disclosure, processing the first processing result based on the multiple reference description information to obtain a second processing result corresponding to the service subject includes:

[0022] The dimensions of the main reference features are classified to determine at least one processing dimension;

[0023] Determine the main reference features of the dimension corresponding to the processing dimension;

[0024] Based on the reference description information of the subject reference features of the dimension corresponding to the processing dimension, determine the dimension description information corresponding to the processing dimension;

[0025] The first processing result is optimized based on the processing dimension to obtain the dimension processing result;

[0026] Based on the dimension processing result and the dimension description information, a second processing result corresponding to the service subject is determined.

[0027] In some embodiments of this disclosure, determining the dimension description information corresponding to the processing dimension based on the reference description information of the subject reference feature of the dimension corresponding to the processing dimension includes:

[0028] A weighted average is applied to the reference description information of the subject reference features of the dimension corresponding to the processing dimension, and the result of the weighted average is used as the dimension description information.

[0029] In some embodiments of this disclosure, optimizing the first processing result based on the processing dimension to obtain a dimension processing result includes:

[0030] Determine the number of orders for the service provider and the average number of orders for the industry to which the service provider belongs;

[0031] Determine the fitting function corresponding to the processing dimension;

[0032] The fitting function is adjusted based on the comparison between the order quantity and the average order quantity to obtain the target fitting function;

[0033] The first processing result is optimized using the target fitting function to obtain the dimension processing result.

[0034] In some embodiments of this disclosure, optimizing the first processing result based on the processing dimension to obtain a dimension processing result includes:

[0035] The first processing result is optimized using a fitting function corresponding to the processing dimension to obtain the dimension processing result.

[0036] According to a second aspect of the present disclosure, a data processing apparatus is provided, comprising: an acquisition module configured to acquire data to be processed of a service subject; a first determination module configured to determine multiple dimensions of subject reference features corresponding to the service subject; a first processing module configured to process the data to be processed using the multiple dimensions of subject reference features to obtain a first processing result corresponding to each dimension of the subject reference features; a second determination module configured to determine the first processing result corresponding to the service data based on the subject reference features of each candidate dimension; and a second processing module configured to process the first processing result according to service reference features to obtain a second processing result corresponding to the service subject.

[0037] In some embodiments of this disclosure, the second processing module is configured to execute:

[0038] The determination submodule is configured to perform subject reference features for each dimension and determine reference description information corresponding to the subject reference features;

[0039] The processing submodule is configured to process the first processing result according to the reference description information to obtain a second processing result corresponding to the service subject.

[0040] In some embodiments of this disclosure, the first determining module is configured to perform:

[0041] Determine the main reference features for multiple candidate dimensions;

[0042] Based on the service reference features, at least some of the candidate dimension subject reference features are selected from the subject reference features of the multiple candidate dimensions as the subject reference features of the corresponding dimension.

[0043] In some embodiments of this disclosure, the first determining module is configured to perform:

[0044] A correlation analysis is performed on the service reference features and the subject reference features of the candidate dimensions to obtain the correlation results;

[0045] From the subject reference features of the multiple candidate dimensions, the subject reference features of the candidate dimensions whose correlation results satisfy the reference conditions are selected as the subject reference features of the corresponding dimensions.

[0046] In some embodiments of this disclosure, the first determining module is configured to perform:

[0047] From the subject reference features of the multiple candidate dimensions, the subject reference features of the candidate dimensions that satisfy the first reference condition with positive correlation are selected as the subject reference features of the corresponding dimensions, and the subject reference features of the candidate dimensions that satisfy the second reference condition with negative correlation are selected as the subject reference features of the corresponding dimensions.

[0048] In some embodiments of this disclosure, the determining submodule is configured to perform:

[0049] Construct a positive sample set and a negative sample set. The positive sample set includes: main reference features of candidate dimensions with positive correlation satisfying the first reference condition. The negative sample set includes: main reference features of candidate dimensions with negative correlation satisfying the second reference condition.

[0050] The positive sample set, the negative sample set, and the service reference features are input into a pre-trained network model to obtain reference description information output by the network model corresponding to the subject reference features, wherein the network model is used to obtain the reference description information.

[0051] In some embodiments of this disclosure, the second processing module is configured to execute:

[0052] The dimensions of the main reference features are classified to determine at least one processing dimension;

[0053] Determine the main reference features of the dimension corresponding to the processing dimension;

[0054] Based on the reference description information of the subject reference features of the dimension corresponding to the processing dimension, determine the dimension description information corresponding to the processing dimension;

[0055] The first processing result is optimized based on the processing dimension to obtain the dimension processing result;

[0056] Based on the dimension processing result and the dimension description information, a second processing result corresponding to the service subject is determined.

[0057] In some embodiments of this disclosure, the second processing module is configured to execute:

[0058] A weighted average is applied to the reference description information of the subject reference features of the dimension corresponding to the processing dimension, and the result of the weighted average is used as the dimension description information.

[0059] In some embodiments of this disclosure, the second processing module is configured to execute:

[0060] Determine the number of orders for the service provider and the average number of orders for the industry to which the service provider belongs;

[0061] Determine the fitting function corresponding to the processing dimension;

[0062] The fitting function is adjusted based on the comparison between the order quantity and the average order quantity to obtain the target fitting function;

[0063] The first processing result is optimized using the target fitting function to obtain the dimension processing result.

[0064] In some embodiments of this disclosure, the second processing module is configured to execute:

[0065] The first processing result is optimized using a fitting function corresponding to the processing dimension to obtain the dimension processing result.

[0066] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the data processing method provided as in the first aspect of the present disclosure.

[0067] According to a fourth aspect of the present disclosure, a storage medium is provided that, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to perform a data processing method as provided in the first aspect of the present disclosure.

[0068] According to a fifth aspect of the present disclosure, a computer program product is provided that, when executed by a processor of an electronic device, enables the electronic device to perform the data processing method provided in the first aspect of the present disclosure.

[0069] The technical solutions provided by the embodiments of this disclosure have at least the following beneficial effects:

[0070] By acquiring the data to be processed from the service subject, determining multiple dimensions of subject reference features corresponding to the service subject, and processing the data to be processed using these multiple dimensions of subject reference features, a first processing result corresponding to each dimension of the subject reference features is obtained. Furthermore, by processing the first processing result based on the service reference features, a second processing result corresponding to the service subject is obtained. Because the data to be processed from the service subject is processed by referencing multiple dimensions of subject reference features and service reference features related to the service subject, the overall data processing effect on the service subject can be effectively improved, making the data processing results more accurate and effectively enhancing the referenceability of the data processing results.

[0071] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0072] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0073] Figure 1 This is a flowchart illustrating a data processing method according to an exemplary embodiment.

[0074] Figure 2 This is a flowchart illustrating a data processing method according to another exemplary embodiment.

[0075] Figure 3 This is a flowchart illustrating a data processing method according to another exemplary embodiment.

[0076] Figure 4 This is a block diagram illustrating a data processing apparatus according to an exemplary embodiment.

[0077] Figure 5 This is a block diagram illustrating a data processing apparatus according to another exemplary embodiment.

[0078] Figure 6 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0079] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0080] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this disclosure.

[0081] Figure 1 This is a flowchart illustrating a data processing method according to an exemplary embodiment.

[0082] This embodiment illustrates the example of a data processing method configured in a data processing device.

[0083] In this embodiment, the data processing method can be configured in a data processing device, which can be located in a server or an electronic device. This disclosure does not limit this.

[0084] This embodiment uses the example of a data processing method configured in an electronic device. The electronic device can be a hardware device with various operating systems and imaging devices, such as a mobile phone, tablet computer, personal digital assistant, or wearable device.

[0085] It should be noted that the execution entity of the embodiments disclosed herein may be, in hardware, a central processing unit (CPU) in a server or electronic device, and in software, a related background service in a server or electronic device, without limitation.

[0086] The implementing entity disclosed herein can be, for example, a live-streaming e-commerce service platform running on electronic devices. This service platform can accommodate multiple service entities and assist each service entity in providing services in the live-streaming e-commerce field, such as live-streaming services, order placement services, and product sales services. This service platform can be presented in the form of a client, such as the client of a live-streaming application. A client, or user terminal, refers to a program that corresponds to a server and provides local services to users, and there is no limitation on this.

[0087] The execution entity in this embodiment can be, for example, a client on the service platform management side. This client is used to manage applications, such as live streaming applications. The application scenario in this embodiment can be described as follows: the service platform can pre-connect multiple service entities (such as merchant stores). Each service entity can rely on the service platform to provide corresponding live streaming services, order placement services, product sales services, and other services in the live streaming e-commerce field. The client on the service platform management side can process the pending data related to the multiple service entities that have been connected. See the following embodiments for details.

[0088] like Figure 1 As shown, the data processing method includes the following steps.

[0089] In step S101, the data to be processed for the service subject is obtained.

[0090] The data to be processed can be, for example, business data related to the service provider, or business performance data related to the services provided by the service provider, such as logistics services, product quality, merchant services, order quantity, user rating scores, after-sales service, and product unit price, etc. There are no restrictions on this.

[0091] It is understandable that live streaming applications typically connect to big data platforms, which record and store all business data involved in providing services to the service provider in real time, so that there is a basis for subsequent traceability. Therefore, in this embodiment of the disclosure, when processing data for a service provider, business data within a certain period of time can be retrieved from the big data platform as data to be processed based on the identifier of the service provider, without any restrictions.

[0092] In step S102, multiple dimensions of subject reference features corresponding to the service subject are determined.

[0093] It is understandable that, since different service providers offer different types of services, the content, type, and other characteristics of the data to be processed may vary depending on the type of service offered by the service provider. Therefore, when processing data from different service providers, the dimensions of the subject reference features used may be the same or different. Specifically, it can be determined based on the type of service offered by the service provider. For example, the subject reference features of a financial service provider may be reflected in real-time performance, while product services may emphasize product quality and after-sales experience. Thus, in this embodiment of the disclosure, multiple dimensions of subject reference features corresponding to the service provider can be determined.

[0094] For example, based on the type of service provided by the service provider, multiple dimensions of subject reference features corresponding to the service provider can be determined. Alternatively, based on user feedback data (such as service evaluations) of the service provider, n dimensions of subject reference features f1, f2, ..., fn corresponding to the service provider can be determined, where n is a positive integer greater than 1, and there are no restrictions on this.

[0095] In this way, subject reference features corresponding to candidate dimensions for each service type can be set in advance on the service platform, and the subject reference features of each candidate dimension can be associated with its corresponding service type. Thus, subject reference features of the corresponding candidate dimensions can be determined according to the type of service provided by the service subject, so as to serve as subject reference features of multiple dimensions corresponding to the service subject.

[0096] In this embodiment of the disclosure, another method for determining subject reference features corresponding to the service subject in a dimension may also be provided. For example, it may be to determine subject reference features of multiple candidate dimensions, and select at least some of the subject reference features of the candidate dimensions as subject reference features of the corresponding dimension based on the service reference features.

[0097] The subject reference features of the corresponding dimension determined based on the type of service provided by the service subject can be called subject reference features of the candidate dimension. Thus, after determining the service reference features of the service platform, at least some of the subject reference features of the candidate dimensions can be selected as subject reference features of the corresponding dimension by combining the service reference features of the service platform.

[0098] For example, the main reference features of the candidate dimensions can be main reference feature 1 of dimension 1, main reference feature 2 of dimension 2, main reference feature 3 of dimension 3, and main reference feature 4 of dimension 4, while the service reference feature A of the service platform can be used as a reference to select main reference feature 2 of dimension 2 and main reference feature 3 of dimension 3 as the main reference features of the corresponding dimensions mentioned above. The selection rules can be adaptively configured according to the actual data processing scenario. For example, the selection can be based on the correlation between service reference feature A and the main reference features of each dimension, without any restrictions.

[0099] This effectively reduces the amount of data. When there are many data dimensions on the supply side, it can automatically select the best candidate reference features for the main body of the candidate dimensions, selecting the candidate reference features with high relevance. This not only ensures the accuracy of data processing but also improves data processing efficiency, thereby optimizing the overall evaluation mechanism of the service platform.

[0100] When selecting at least some of the candidate dimension's subject reference features from multiple candidate dimension's subject reference features based on service reference features, specifically, a correlation analysis can be performed on the service reference features and the candidate dimension's subject reference features to obtain the correlation results. Then, from multiple candidate dimension's subject reference features, the candidate dimension's subject reference features whose correlation results meet the reference conditions can be selected as the corresponding dimension's subject reference features. This approach is simple to implement and ensures the overall evaluation efficiency of the service platform.

[0101] The reference conditions can be preset or adaptively adjusted according to actual evaluation needs, without any restrictions.

[0102] For example, a correlation threshold can be used as a reference condition. If the correlation result between the subject reference feature and the service reference feature of a candidate dimension is greater than the correlation threshold, it indicates that the correlation result of the subject reference feature of the candidate dimension meets the reference condition. Conversely, if the correlation result is less than the threshold, it indicates that the correlation result of the subject reference feature of the candidate dimension does not meet the reference condition.

[0103] For example, Pearson correlation analysis can be performed on the subject reference features and service reference features of candidate dimensions. The correlations between f1, f2, f3, ..., fn (where n is a positive integer less than or equal to N) and the service reference feature label can be calculated. For instance, the data to be processed for the service subject can be input into the Pearson correlation analysis function along with the values ​​of the subject reference features and the service reference feature labels for each candidate dimension. That is, the value of the subject reference feature and the service reference feature label for each candidate dimension are used as a pair of input data, and this pair of input data is fed into the Pearson correlation analysis function. The correlation analysis function determines the Pearson correlation coefficient based on its output. A Pearson correlation coefficient closer to 1 indicates a stronger positive correlation between the service reference feature and the main reference feature of the candidate dimension, while a coefficient closer to -1 indicates a stronger negative correlation. This allows for the selection of indicators (f1, f2, ..., fz) whose Pearson correlation coefficients meet expectations. Here, z is a positive integer less than or equal to n, and n is a positive integer less than or equal to N. z represents the number of indicators whose Pearson correlation coefficients meet expectations.

[0104] In other embodiments, when selecting the subject reference features of a candidate dimension whose correlation results satisfy the reference conditions from multiple candidate subject reference features, the subject reference features of the candidate dimension that satisfy the first reference condition with positive correlation can be selected as the subject reference features of the corresponding dimension, and the subject reference features of the candidate dimension that satisfy the second reference condition with negative correlation can be selected as the subject reference features of the corresponding dimension. This expands the differences between the subject reference features of the candidate dimensions and improves the objectivity of the data processing results.

[0105] The first and second reference conditions can be pre-configured or adaptively adjusted according to actual evaluation needs, without any restrictions.

[0106] For example, the above describes how the data to be processed by the service subject is mapped to the values ​​of the subject reference features and service reference feature labels of each candidate dimension. That is, the values ​​of the subject reference features and service reference feature labels of each candidate dimension are used as a pair of input data. This pair of input data is input into the Pearson correlation analysis function. After determining the Pearson correlation coefficient based on the output of the Pearson correlation analysis function, the subject reference features of the candidate dimensions that meet the positive correlation and the subject reference features of the candidate dimensions that meet the negative correlation can also be determined based on the Pearson correlation coefficient. Then, the subject reference features of the candidate dimensions that meet the first reference condition (e.g., sorting the positive correlations and selecting the top 20%) are selected as the subject reference features of the corresponding dimensions, and the subject reference features of the candidate dimensions that meet the second reference condition (e.g., sorting the negative correlations and selecting the top 20%) are selected as the subject reference features of the corresponding dimensions. There are no restrictions on this.

[0107] In step S103, the data to be processed is processed using subject reference features of multiple dimensions to obtain the first processing result corresponding to the dimensions of each subject reference feature.

[0108] After obtaining the data to be processed for the service subject and determining the subject reference features of various dimensions corresponding to the service subject, the first processing result of the data to be processed based on the subject reference features of each dimension can be determined.

[0109] In other words, the data to be processed is processed using multiple dimensions of subject reference features to obtain the first processing result corresponding to the dimensions of each subject reference feature.

[0110] For example, a corresponding fitting function can be set for the subject reference feature of each dimension. This fitting function fits the correspondence between different independent and dependent variables. The independent variable can represent the data to be processed, and the dependent variable can represent the first processing result. Thus, the data to be processed is used as the independent variable of the fitting function, and the dependent variable corresponding to the fitting function is calculated to obtain the first processing result corresponding to the subject reference feature of each dimension. There are no restrictions on this.

[0111] In step S104, the first processing result is processed according to the service reference features to obtain the second processing result corresponding to the service subject.

[0112] After determining the first processing result corresponding to the subject reference features of the data to be processed based on each dimension, the first processing result can be processed according to the service reference features to obtain the second processing result corresponding to the service subject.

[0113] Among them, the service reference feature can be associated with the service platform to which the service subject belongs. This service reference feature can be used as a reference for the service platform dimension when processing the data to be processed. For example, the service reference feature is the platform's transaction rate, or it can be other service reference features associated with the service platform (e.g., platform transaction volume). There are no restrictions on this.

[0114] Assuming the service's reference characteristic is the platform conversion rate, the platform conversion rate can be determined as follows:

[0115] Understandably, based on the demands of service platforms, they typically aim to increase LTV (Long-Time GMV, which is the transaction volume of a long-term website, mainly including completed and ongoing transactions). LTV is related to factors such as platform conversion rate, merchant conversion rate, and unit price of goods. Since unit price of goods is related to platform characteristics and consumer habits, it has uncontrollable factors for service platforms. Data processing results are usually more clearly correlated with merchant conversion rate and platform conversion rate. Therefore, in this embodiment of the disclosure, in order to balance the interests of the service subject (merchant) and the service platform, the platform conversion rate can be used as a reference for evaluating service effectiveness.

[0116] It should be noted that the platform conversion rate of the aforementioned service platform can be the platform conversion rate over a certain period of time, or it can be the average of the platform conversion rates over a longer period of time. This platform conversion rate can be used to describe the probability of users making a transaction on the service platform within a certain period of time, and there are no restrictions on it.

[0117] In this embodiment, by acquiring the data to be processed of the service subject, determining multiple dimensions of subject reference features corresponding to the service subject, and processing the data to be processed using these multiple dimensions of subject reference features, a first processing result corresponding to each dimension of the subject reference features is obtained. Furthermore, by processing the first processing result based on the service reference features, a second processing result corresponding to the service subject is obtained. Since the data to be processed of the service subject is processed by referencing multiple dimensions of the subject reference features of the service subject, as well as service reference features related to the service subject, the overall data processing effect of the service subject can be effectively improved, making the data processing results more accurate and effectively enhancing the referenceability of the data processing results.

[0118] Figure 2 This is a flowchart illustrating a data processing method according to another exemplary embodiment.

[0119] like Figure 2 As shown, this data processing method includes at least the following steps.

[0120] In step S201, the data to be processed for the service subject is obtained.

[0121] In step S202, multiple dimensions of subject reference features corresponding to the service subject are determined.

[0122] In step S203, the data to be processed is processed using subject reference features of multiple dimensions to obtain the first processing result corresponding to the dimensions of each subject reference feature.

[0123] For specific examples of steps S201-S203, please refer to the above embodiments, which will not be repeated here.

[0124] In step S204, for each dimension of the subject reference feature, the reference description information corresponding to the subject reference feature is determined.

[0125] After determining the subject reference features of multiple dimensions corresponding to the service subject, the reference description information corresponding to the subject reference features can be determined. For example, the reference description information corresponding to the subject reference features can be determined by referring to the service reference features mentioned above. The reference description information can be used to describe the importance of the corresponding subject reference features, and this importance can be a reference weight.

[0126] In other words, this embodiment not only determines the personalized multi-dimensional subject reference features of the service subject, but also determines the reference description information corresponding to the subject reference features of each dimension based on the service reference features of the service platform. This enables data processing not only to refer to the personalized subject reference features of the service subject, but also to refer to the service reference features of the service platform on which it relies, thereby improving the overall data processing effect of the service platform on the service subject.

[0127] When the performance of service reference features of service platforms is different, it may affect the reference description information of the subject reference features to a certain extent. Therefore, in this embodiment of the disclosure, the step of determining the reference description information of the subject reference features based on the service reference features is performed.

[0128] In some embodiments, when determining the reference description information of the subject reference features of the candidate dimension, a positive sample set and a negative sample set may be constructed. The positive sample set includes the subject reference features of the candidate dimension whose positive correlation satisfies the first reference condition, and the negative sample set includes the subject reference features of the candidate dimension whose negative correlation satisfies the second reference condition. The positive sample set, the negative sample set, and the service reference features are then input into a pre-trained network model to obtain the reference description information corresponding to the subject reference features output by the network model. The network model is used to obtain the reference description information. Since the network model is pre-trained, the accuracy and efficiency of obtaining the reference description information can be effectively guaranteed.

[0129] When using subject reference features of various dimensions to process data of a service subject, the subject reference features of each dimension may have different or the same weight values. The weight values ​​corresponding to the subject reference features of each dimension can be called reference description information.

[0130] In this embodiment of the disclosure, model training can be used to determine the reference description information corresponding to the subject reference features in each dimension.

[0131] For example, the network model described above can be trained based on the subject reference features of candidate dimensions that satisfy the first reference condition for positive correlation and the subject reference features of candidate dimensions that satisfy the second reference condition for negative correlation. The network model can be any neural network model in artificial intelligence. For example, reference setting weight values ​​corresponding to the subject reference features of each candidate dimension can be preset, and then a neural network model in artificial intelligence can be trained. When the difference between the predicted weight value output by the neural network model and the reference setting weight value meets the requirements, it is determined that the neural network model training is complete. Thus, the reference description information corresponding to the subject reference features of each dimension can be obtained based on the trained neural network model.

[0132] For example, training a network model and using that network model (which is used to obtain reference description information) to obtain reference description information for the subject's reference features in various dimensions can be illustrated as follows:

[0133] 1. Normalize the main reference features.

[0134] Normalization of the main reference features can be achieved by using the maximum-min standardization method (fz-min) / (max-min) to dedimensionalize all main reference features, thereby effectively avoiding the influence of the selection of main reference features with different dimensions on the model calculation. Here, min represents the minimum value of the main reference feature, and max represents the maximum value of the main reference feature.

[0135] 2. Construct a positive sample set and a negative sample set. The positive sample set includes the main reference features of the candidate dimensions that have positive correlation and satisfy the first reference condition. The negative sample set includes the main reference features of the candidate dimensions that have negative correlation and satisfy the second reference condition.

[0136] Determine positive and negative samples: Sort the subject reference features of each dimension according to the correlation results calculated with the service reference feature label, use the first 20% of the data to form a positive sample set, and use the last 20% of the data to form a negative sample set, thereby increasing the difference between samples.

[0137] 3. Input the positive sample set, negative sample set, and service reference features into the pre-trained network model to obtain the reference description information output by the network model corresponding to the main reference features.

[0138] The pre-trained network model can be a binary classification logistic regression model:

[0139]

[0140] Where: w is the reference description information of the subject reference feature for each dimension, b is the bias, h(x) is the prediction function, utilizing the characteristics of the sigmoid function, the final output value of h(x) is between (0, 1), e is the natural constant, a mathematical constant, an infinite non-repeating decimal, and a transcendental number, its value is approximately 2.718281828459045, t x This represents the relevance corresponding to the x-th subject reference feature in dimension w.

[0141] 4. This pre-trained network model also corresponds to a cost function J(θ):

[0142]

[0143] Where: m is the number of dimensions of the main reference feature, y is the true value, i.e., the service reference feature label, h(x) is the formula (1) above, where θ is the sampled value used to assist in calculating the model gradient, for example, it can be a partial value sampled from the values ​​corresponding to multiple main reference features, i is used to index each sampled value, i is less than or equal to the number of sampled values, J(θ) is the cost function J(θ) above, used to calculate the cost loss of the model, N represents a positive integer, n represents the number of sampled values, that is to say, h θ (x (i) ) can be obtained by sampling the above h(x).

[0144] 5. Then, using gradient descent, i.e., taking the derivative of the cost function above, we obtain the point of fastest gradient descent:

[0145]

[0146] Using the above formula (3), the sampled value that makes the gradient descent of the network model the fastest is obtained, and J(θ) is determined at this time. The value of J(θ) is used until the value of J(θ) is the minimum loss threshold (for example, after sampling some values ​​from the values ​​corresponding to multiple subject reference features, the value corresponding to each sampled subject reference feature, the reference description information and correlation corresponding to the subject reference feature can be input into the above formula (1), and formulas (2) and (3) are constructed according to the formula (1) with variables substituted, and the value of the cost function J(θ) output corresponding to the fastest gradient descent of the network model is calculated. Then, the value of the cost function J(θ) output is compared with the preset minimum loss threshold. If it is equal to the minimum loss threshold, the iteration process can be ended. If the minimum loss threshold is not reached, it can be resampled, and iterative training can be carried out based on the resampled partial values ​​until the value of J(θ) is the minimum loss threshold). Finally, w1, w2, w3, ... are obtained, which are the reference description information of the subject reference features (f1, f2, f3...) in each dimension.

[0147] In step S205, the first processing result is processed according to the reference description information to obtain the second processing result corresponding to the service subject.

[0148] For example, the reference description information and the first processing result can be weighted and fused, and the fused result can be used as the second processing result. Alternatively, any other possible method can be used to process the first processing result based on the reference description information to obtain the second processing result corresponding to the service subject. There are no restrictions on this.

[0149] In this embodiment, by acquiring the data to be processed of the service subject, determining multiple dimensions of subject reference features corresponding to the service subject, and processing the data using these multiple dimensions of subject reference features, a first processing result corresponding to each dimension of the subject reference features is obtained. Then, the first processing result is processed based on the service reference features to obtain a second processing result corresponding to the service subject. Since the data to be processed of the service subject is processed by referencing multiple dimensions of the service subject's subject reference features and service reference features related to the service subject, the overall data processing effect on the service subject can be effectively improved, making the data processing results more accurate and effectively enhancing the referability of the data processing results. Not only are the personalized multiple dimensions of the service subject's subject reference features determined, but also the reference description information corresponding to each dimension of the subject reference features is determined based on the service platform's service reference features. This achieves data processing not only by referring to the personalized subject reference features of the service subject but also by referring to the service reference features of the service platform on which it relies, thereby improving the overall data processing effect of the service platform on the service subject.

[0150] Figure 3 This is a flowchart illustrating a data processing method according to another exemplary embodiment.

[0151] This embodiment illustrates the specific implementation steps for obtaining data processing information of the service subject based on the first processing result and reference description information. For example... Figure 3 As shown, the data processing method includes the following steps.

[0152] In step S301, the dimensions of the subject reference features are classified to determine at least one processing dimension.

[0153] When there are multiple dimensions of the subject reference features, subject reference features with similar dimensions can be classified into the same processing dimension. This allows data processing to be performed based on the processing dimension, which can be used to describe the evaluation dimension. The evaluation dimension can be specifically determined based on the business scenario requirements of data processing. As a result, the data processing results based on the processing dimension are more intuitive and improve the referenceability of the data processing results.

[0154] For example, the processing dimensions can be user evaluation, customer service, dispute experience, or after-sales service. Of course, they can also be adaptively set and expanded according to the business scenario requirements of data processing, thereby ensuring the flexibility and applicability of the data processing methods.

[0155] In step S302, the subject reference feature of the dimension corresponding to the processing dimension is determined.

[0156] For example, assuming the dimensions of the subject's reference features are logistics services, product quality, merchant services, order quantity, user rating scores, after-sales services, product unit price, and number of complaints, then logistics services and merchant services can be classified under the customer service dimension, while product quality and user rating scores can be classified under the user rating dimension, after-sales services can be classified under the after-sales service dimension, and the number of complaints can be classified under the dispute experience dimension, and so on.

[0157] In other words, when there are multiple dimensions of the subject reference features, the subject reference features of multiple candidate dimensions can be processed and the dimensions can be consolidated to improve the intuitiveness of the data processing results.

[0158] In step S303, the dimension description information corresponding to the processing dimension is determined based on the reference description information of the subject reference feature of the dimension corresponding to the processing dimension.

[0159] After mapping the multiple dimensions of the main reference features to their corresponding processing dimensions, the weight value of each processing dimension can be determined based on the reference description information of the main reference features obtained under each processing dimension (the weight value of the processing dimension can be called the dimension description information). This not only broadens the scope of reference elements from the perspective of main reference features, but also realizes the processing dimension aggregation of main reference features of multiple dimensions, thereby improving the intuitiveness of the data processing results. Furthermore, the reference description information of the main reference features is also mapped to the dimension description information of the processing dimensions, thereby improving the objectivity of data processing in each processing dimension.

[0160] In some embodiments, a weighted average is performed on the reference description information of the subject reference features of the dimension corresponding to the processing dimension, and the result of the weighted average is used as the dimension description information.

[0161] For example, the data processing result: comprehensive score. ω can be the weighted average of the reference description information of the main reference features within the user evaluation dimension, while the user evaluation score is the rating value corresponding to the user evaluation dimension; τ is the weighted average of the reference description information of the main reference features within the dispute experience dimension, while the product quality score is the rating value corresponding to the dispute experience dimension; γ is the weighted average of the reference description information of the main reference features within the customer service dimension, while the customer service score is the rating value corresponding to the customer service dimension. The after-sales service score is the weighted average of the reference description information of the main reference features within the after-sales service dimension, and the score is the rating value corresponding to the after-sales service dimension.

[0162] In practical applications, the service platform can adaptively adjust ω, τ, and γ based on the business needs of different data processing scenarios. The values ​​or characteristics of these values ​​indicate that the data processing process in this embodiment is controllable by the service platform, enabling the achievement of business objectives based on data processing and completing the aforementioned ω, τ, γ values ​​during the data processing process. Adaptive adjustment of the value or feature meaning.

[0163] Of course, any other possible methods can be used to determine the dimensional description information of the processing dimension based on the reference description information of the subject reference feature. For example, it can be a modeling method. Specifically, for example, a matching mapping relationship between the reference description information and the dimensional description information can be obtained by modeling, and the dimensional description information of the processing dimension can be determined based on the matching mapping relationship. There are no restrictions on this.

[0164] In other words, this embodiment first classifies the multiple dimensions of the subject reference features to obtain the subject reference features corresponding to the processing dimensions. Specifically, the subject reference features corresponding to dimensions A, B, and C under processing dimension 1, the subject reference features corresponding to dimensions D, E, and F under processing dimension 2, and the subject reference features corresponding to dimensions G, H, and I under processing dimension 3. Furthermore, processing dimension 1 corresponds to dimension description information 1, processing dimension 2 corresponds to dimension description information 2, and processing dimension 3 corresponds to dimension description information 3.

[0165] In step S304, the first processing result is optimized based on the processing dimension to obtain the dimension processing result.

[0166] After determining the subject reference features of the dimension corresponding to the processing dimension and determining the dimension description information corresponding to the processing dimension based on the reference description information of the subject reference features of the dimension corresponding to the processing dimension, the first processing result can be optimized based on the processing dimension to obtain the dimension processing result. This can further improve the objectivity of the data processing result, avoid the influence of subjective judgment, and make the data processing result more reasonable overall.

[0167] Among them, optimization processing includes adjusting the first processing result based on the processing dimension, such as summarizing multiple first processing results within the processing dimension, or calculating the weighted average, or calculating the variance, until all multiple first processing results in each processing dimension have been adjusted (adjustment can be, for example, summarizing, or calculating the weighted average, or calculating the variance). The result obtained based on the adjustment of each processing dimension can be used as the dimension processing result, and there are no restrictions on this.

[0168] This disclosure provides two methods for optimizing a first processing result based on a processing dimension to obtain a dimension processing result. In practical applications, the method can be adaptively selected and used according to the type of processing dimension, without any limitation.

[0169] In some embodiments, the number of orders of the service subject and the average number of orders corresponding to the industry to which the service subject belongs can be determined, a fitting function corresponding to the processing dimension can be determined, and the fitting function can be adjusted according to the comparison result of the number of orders and the average number of orders to obtain a target fitting function. The target fitting function is then used to optimize the first processing result to obtain the dimension processing result. For example, the first processing result can be used as the independent variable of the target fitting function, and the output of the target fitting function based on the independent variable can be used as the dimension processing result to optimize the first processing result.

[0170] For example, when the processing dimension is user rating, the fitting function corresponding to the user rating dimension can be, for example, as follows:

[0171] Score = α*(DSR+A)

[0172] This fitting function indicates that when the order quantity is less than a specified value, such as 400, the influence of the industry average order size can be disregarded. In other words, when the order quantity is determined to be less than the average order quantity, Score = α*(DSR + A) can be directly used as the fitting function. However, when the order quantity is greater than the average order quantity, the fitting function can be adjusted. For example, the adjusted fitting function (which can be called the target fitting function) can be in the following form:

[0173] Score=α*(DSR+A)+β*B*ΔDSR,

[0174] DSR stands for Detail Seller Rating, which is the DSR of the service provider, while ΔDSR is the industry average coefficient.

[0175] The meanings of the parameters in the above objective fitting function can be seen in Table 1 below:

[0176] Table 1

[0177]

[0178]

[0179] In other embodiments, the first processing result is optimized based on the processing dimension to obtain the dimension processing result. Alternatively, the first processing result can be optimized using a fitting function corresponding to the processing dimension to obtain the dimension processing result.

[0180] For example, if the processing dimension is customer service, dispute experience, or after-sales service, then the fitting function corresponding to the processing dimension can be directly used to optimize the first processing result to obtain the dimension processing result.

[0181] For example, when the processing dimension is customer service, the main reference feature under this processing dimension is: 5-minute response rate (for example, the dialogue interaction data between users and customer service can be obtained, and the 5-minute response rate can be obtained by statistical analysis of the dialogue interaction data. This 5-minute response rate can refer to the effective response rate of customer service within 5 minutes in the last 3 days. The calculation method of the effective response rate of customer service is as follows: Total number of consumers with effective customer service response / Total number of consumers consulting the merchant × 100% = (Total number of consumers consulting the merchant - Total number of consumers with invalid response) / Total number of consumers consulting the merchant × 100%). Then, assuming that the distribution quantile of the main reference feature of the service platform is as shown in Table 2 below:

[0182] Table 2

[0183]

[0184]

[0185] Therefore, the fitting function corresponding to the customer service dimension can be determined as follows:

[0186]

[0187] Where x′ represents the main reference feature of the service platform (5-minute response rate * 100).

[0188] For example, when the processing dimension is the dispute experience dimension, the main reference feature under this processing dimension is: self-built Youzan dispute rate (the self-built Youzan dispute rate can refer to a mini-program loaded by the service platform, which is a sales statistical analysis mini-program in related technologies. The self-built Youzan dispute rate can be the dispute rate obtained from the statistical analysis of the mini-program loaded by the service platform. The dispute rate can be determined by identifying the types of all sales events, identifying a portion of the sales events that are dispute types from all sales events, and then determining the ratio of the number of dispute-type sales events to the total number of all sales events as the dispute rate). Then, assuming that the distribution quantiles of the main reference features of the service platform are as shown in Table 3 below:

[0189] Table 3

[0190] Ideal value S+ 5 points maximum Self-built Youzan dispute rate: 0% Excellent S 4.5-5 points Self-built Youzan dispute rate: 0.008% - 0% excellent A 3.5-4.5 points Self-built Youzan dispute rate: 0.045%-0.008% good B 2.5-3.5 points The dispute rate for self-built Youzan systems is 0.22%-0.045%. The Doctrine of the Mean C 1.5-2.5 points Self-built Youzan dispute rate: 1.3% - 0.22% Poor D 0-1.5 points Self-built Youzan dispute rate: 15%-1.3%

[0191] Therefore, the fitting function corresponding to the dispute experience dimension can be determined as follows:

[0192]

[0193] Where x″ represents the subject reference characteristics of the service provider (dispute experience score * 100).

[0194] For example, when the processing dimension is after-sales service, the main reference feature under this processing dimension is: merchant responsibility return rate (for example, if the total number of merchant transactions is 50 and the number of returns is 1, then the return rate is 2%). Then, assuming the quantiles of the main reference feature distribution of the service platform are as shown in Table 4 below:

[0195] Table 4

[0196] Ideal value S+ 5 points maximum Merchant liability return rate: within 0.4% Excellent S 4.5-5 points Merchant liability return rate: 0.55%-0.4% excellent A 3.5-4.5 points Merchant liability return rate: 1%-0.55% good B 2.5-3.5 points Merchant liability return rate: 2.5%-1% The Doctrine of the Mean C 1.5-2.5 points Merchant liability return rate: 5%-2.5% Poor D 0-1.5 points Merchant liability return rate: 10%-5%

[0197] Therefore, the fitting function corresponding to the after-sales service dimension can be determined as follows:

[0198]

[0199] Where x″′ is the main reference feature of the service provider (commercial responsibility cancellation rate * 100).

[0200] In step S305, a second processing result corresponding to the service subject is determined based on the dimension processing result and the dimension description information.

[0201] The data processing described above, which determines the service subject based on the dimension processing results and dimension description information, can be achieved by weighting the dimension processing results according to the dimension description information of each processing dimension, and then using the weighted average as the second processing result corresponding to the service subject.

[0202] After determining the processing results for each processing dimension, the results can be made more interpretable. For example, based on the dimensional description information of each processing dimension, the processing results can be converted into corresponding score ranges, so that the service provider can obtain more intuitive data processing results.

[0203] As shown in Table 5 below:

[0204] Table 5

[0205] Serial Number Dimension score range 1 User Reviews 0-30 2 Customer service 0-25 3 Dispute Experience 0-20 4 After-sales service 0-25

[0206] In this embodiment, after mapping the subject reference features of multiple dimensions to their corresponding processing dimensions, the dimension description information of each processing dimension can be determined based on the reference description information of the subject reference features obtained under each processing dimension. This allows the dimension processing results based on the processing dimensions to also reference the service reference features. This not only broadens the scope of reference elements from the perspective of subject reference features but also achieves index dimension aggregation processing of multiple subject reference features, thereby improving the intuitiveness of the data processing results. Furthermore, the reference description information of the subject reference features is similarly mapped to the dimension description information of the processing dimensions, thus improving the objectivity of data processing in each processing dimension. This makes the data processing process in this embodiment controllable by the service platform, achieving business objectives based on data processing and completing the self-training and weighting of the data processing mechanism. The first processing result can also be optimized based on the processing dimensions to obtain dimension processing results, further improving the objectivity of the data processing results and preventing them from being influenced by subjective judgments, making the overall data processing results more reasonable.

[0207] Figure 4 This is a block diagram illustrating a data processing apparatus according to an exemplary embodiment.

[0208] like Figure 4 As shown, the data processing device 40 includes:

[0209] The acquisition module 401 is configured to acquire the pending data of the service subject;

[0210] The first determining module 402 is configured to determine the subject reference features of multiple dimensions corresponding to the service subject;

[0211] The first processing module 403 is configured to perform processing of the data to be processed using subject reference features of multiple dimensions, and to obtain a first processing result corresponding to the dimensions of each subject reference feature;

[0212] The second determining module 404 is configured to execute the first processing result corresponding to the subject reference features of each candidate dimension for determining the business data.

[0213] The second processing module 405 is configured to process the first processing result based on the service reference features to obtain a second processing result corresponding to the service subject.

[0214] In some embodiments of this disclosure, such as Figure 5 As shown, Figure 5 This is a block diagram of a data processing apparatus according to another exemplary embodiment, wherein the second processing module 405 includes:

[0215] The determination submodule 4051 is configured to perform the determination of reference description information corresponding to the subject reference feature for each dimension;

[0216] The processing submodule 4052 is configured to process the first processing result based on the reference description information to obtain a second processing result corresponding to the service subject.

[0217] In some embodiments of this disclosure, the first determining module 402 is configured to perform:

[0218] Determine the main reference features for multiple candidate dimensions;

[0219] Based on the service reference features, at least some of the candidate dimensions' subject reference features are selected from the subject reference features of multiple candidate dimensions as the subject reference features of the corresponding dimensions.

[0220] In some embodiments of this disclosure, the first determining module 402 is configured to perform:

[0221] Correlation analysis is performed on the subject reference features of service reference features and candidate dimensions to obtain correlation results;

[0222] From the subject reference features of multiple candidate dimensions, the subject reference features of the candidate dimensions whose correlation results meet the reference conditions are selected as the subject reference features of the corresponding dimensions.

[0223] In some embodiments of this disclosure, the first determining module 402 is configured to perform:

[0224] From the subject reference features of multiple candidate dimensions, the subject reference features of the candidate dimensions that satisfy the first reference condition with positive correlation are selected as the subject reference features of the corresponding dimensions, and the subject reference features of the candidate dimensions that satisfy the second reference condition with negative correlation are selected as the subject reference features of the corresponding dimensions.

[0225] In some embodiments of this disclosure, submodule 4051 is configured to execute:

[0226] Construct a positive sample set and a negative sample set. The positive sample set includes the main reference features of the candidate dimensions whose positive correlation satisfies the first reference condition, and the negative sample set includes the main reference features of the candidate dimensions whose negative correlation satisfies the second reference condition.

[0227] The positive sample set, the negative sample set, and the service reference features are input into the pre-trained network model to obtain the reference description information corresponding to the subject reference features output by the network model. The network model is used to obtain the reference description information.

[0228] In some embodiments of this disclosure, the second processing module 405 is configured to perform:

[0229] Classify the dimensions of the main reference features and determine at least one processing dimension;

[0230] Determine the main reference features of the dimensions corresponding to the processing dimensions;

[0231] Based on the reference description information of the subject reference features of the dimension corresponding to the processing dimension, determine the dimension description information corresponding to the processing dimension;

[0232] The first processing result is optimized based on the processing dimension to obtain the dimension processing result;

[0233] Based on the dimension processing results and dimension description information, determine the second processing result corresponding to the service subject.

[0234] In some embodiments of this disclosure, the second processing module 405 is configured to perform:

[0235] The reference description information of the subject reference features of the dimension corresponding to the processing dimension is subjected to weighted average processing, and the result of the weighted average processing is used as the dimension description information.

[0236] In some embodiments of this disclosure, the second processing module 405 is configured to perform:

[0237] Determine the number of orders for the service provider and the average number of orders for the industry to which the service provider belongs;

[0238] Determine the fitting function corresponding to the processing dimension;

[0239] The fitting function is adjusted based on the comparison between the number of orders and the average number of orders to obtain the target fitting function;

[0240] The first processing result is optimized using the target fitting function to obtain the dimension processing result.

[0241] In some embodiments of this disclosure, the second processing module 405 is configured to perform:

[0242] The first processing result is optimized using a fitting function corresponding to the processing dimension to obtain the dimension processing result.

[0243] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0244] In this embodiment, by acquiring the data to be processed of the service subject, determining multiple dimensions of subject reference features corresponding to the service subject, and processing the data to be processed using these multiple dimensions of subject reference features, a first processing result corresponding to each dimension of the subject reference features is obtained. Furthermore, by processing the first processing result based on the service reference features, a second processing result corresponding to the service subject is obtained. Since the data to be processed of the service subject is processed by referencing multiple dimensions of the subject reference features of the service subject, as well as service reference features related to the service subject, the overall data processing effect of the service subject can be effectively improved, making the data processing results more accurate and effectively enhancing the referenceability of the data processing results.

[0245] This disclosure also provides an electronic device. Figure 6 This is a block diagram illustrating an electronic device according to an exemplary embodiment. For example, the electronic device 600 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0246] Reference Figure 6 The electronic device 600 may include one or more of the following components: a processing component 602, a memory 604, a power supply component 606, a multimedia component 608, an audio component 610, an input / output (I / O) interface 612, a sensor component 614, and a communication component 616.

[0247] Processing component 602 typically controls the overall operation of electronic device 600, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 602 may include one or more processors 620 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 602 may include one or more modules to facilitate interaction between processing component 602 and other components. For example, processing component 602 may include a multimedia module to facilitate interaction between multimedia component 608 and processing component 602.

[0248] Memory 604 is configured to store various types of data to support the operation of electronic device 600. Examples of this data include instructions for any application or method operating on electronic device 600, contact data, phonebook data, messages, pictures, videos, etc. Memory 604 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0249] Power supply component 606 provides power to various components of electronic device 600. Power supply component 606 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 600.

[0250] Multimedia component 608 includes a touch display screen that provides an output interface between electronic device 600 and user. In some embodiments, the touch display screen may include a liquid crystal display (LCD) and a touch panel (TP). The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of touch or swipe actions but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 608 includes a front-facing camera and / or a rear-facing camera. When electronic device 600 is in an operating mode, such as a shooting mode or video mode, the front-facing camera and / or rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0251] Audio component 610 is configured to output and / or input audio signals. For example, audio component 610 includes a microphone (MIC) configured to receive external audio signals when electronic device 600 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 604 or transmitted via communication component 616.

[0252] In some embodiments, the audio component 610 further includes a speaker for outputting audio signals.

[0253] I / O interface 612 provides an interface between processing component 602 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0254] Sensor assembly 614 includes one or more sensors for providing state assessments of various aspects of electronic device 600. For example, sensor assembly 614 may detect the on / off state of electronic device 600, the relative positioning of components such as the display and keypad of electronic device 600, changes in position of electronic device 600 or a component of electronic device 600, the presence or absence of user contact with electronic device 600, orientation or acceleration / deceleration of electronic device 600, and temperature changes of electronic device 600. Sensor assembly 614 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 614 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 614 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0255] Communication component 616 is configured to facilitate wired or wireless communication between electronic device 600 and other devices. Electronic device 600 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 616 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 616 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0256] In an exemplary embodiment, the electronic device 600 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the data processing method described above.

[0257] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 604 including instructions, which can be executed by a processor 620 of an electronic device 600 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0258] A non-transitory computer-readable storage medium that, when instructions in the storage medium are executed by a processor of an electronic device 600, enables the electronic device 600 to perform a data processing method.

[0259] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0260] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A data processing method, characterized in that, include: Obtain the pending data of the service provider; Multiple candidate dimensions' subject reference features are determined, and a correlation analysis is performed on the service reference features and the subject reference features of the candidate dimensions to obtain correlation results. From the multiple candidate dimensions' subject reference features, the subject reference features of the candidate dimensions whose correlation results satisfy the reference conditions are selected as the subject reference features of the corresponding dimensions. The service reference features are associated with the service platform to which the service subject belongs, and the service reference features are used as a reference for the service platform dimension when processing the data to be processed. The data to be processed is processed using subject reference features of multiple dimensions to obtain a first processing result corresponding to the dimensions of each subject reference feature; The first processing result is processed according to the service reference features to obtain the second processing result corresponding to the service subject.

2. The method according to claim 1, characterized in that, The step of processing the first processing result according to the service reference features to obtain the second processing result corresponding to the service subject includes: For each dimension of the subject reference feature, reference description information corresponding to the subject reference feature is determined, wherein the reference description information is used to describe the importance of the corresponding subject reference feature; The first processing result is processed according to the reference description information to obtain a second processing result corresponding to the service subject.

3. The method according to claim 1, characterized in that, The step of selecting the subject reference features of the candidate dimensions whose relevance results satisfy the reference conditions from the subject reference features of the multiple candidate dimensions as the subject reference features of the corresponding dimension includes: From the subject reference features of the multiple candidate dimensions, the subject reference features of the candidate dimensions that satisfy the first reference condition with positive correlation are selected as the subject reference features of the corresponding dimensions, and the subject reference features of the candidate dimensions that satisfy the second reference condition with negative correlation are selected as the subject reference features of the corresponding dimensions.

4. The method according to claim 2, characterized in that, For each dimension of the subject reference feature, determining the reference description information corresponding to the subject reference feature includes: Construct a positive sample set and a negative sample set. The positive sample set includes: main reference features of candidate dimensions with positive correlation satisfying the first reference condition. The negative sample set includes: main reference features of candidate dimensions with negative correlation satisfying the second reference condition. The positive sample set, the negative sample set, and the service reference features are input into a pre-trained network model to obtain reference description information output by the network model corresponding to the subject reference features, wherein the network model is used to obtain the reference description information.

5. The method according to claim 2, characterized in that, The step of processing the first processing result according to the reference description information to obtain a second processing result corresponding to the service subject includes: The dimensions of the main reference features are classified to determine at least one processing dimension; Determine the main reference features of the dimension corresponding to the processing dimension; Based on the reference description information of the subject reference features of the dimension corresponding to the processing dimension, determine the dimension description information corresponding to the processing dimension; The first processing result is optimized based on the processing dimension to obtain the dimension processing result; Based on the dimension processing result and the dimension description information, a second processing result corresponding to the service subject is determined.

6. The method according to claim 5, characterized in that, The step of determining the dimension description information corresponding to the processing dimension based on the reference description information of the subject reference features of the dimension corresponding to the processing dimension includes: A weighted average is applied to the reference description information of the subject reference features of the dimension corresponding to the processing dimension, and the result of the weighted average is used as the dimension description information.

7. The method according to claim 5, characterized in that, The optimization of the first processing result based on the processing dimension to obtain the dimension processing result includes: Determine the number of orders for the service provider and the average number of orders for the industry to which the service provider belongs; Determine the fitting function corresponding to the processing dimension; The fitting function is adjusted based on the comparison between the order quantity and the average order quantity to obtain the target fitting function; The first processing result is optimized using the target fitting function to obtain the dimension processing result.

8. The method according to claim 5, characterized in that, The optimization of the first processing result based on the processing dimension to obtain the dimension processing result includes: The first processing result is optimized using a fitting function corresponding to the processing dimension to obtain the dimension processing result.

9. A data processing apparatus, characterized in that, include: The acquisition module is configured to acquire pending data from the service provider. The first determining module is configured to determine the subject reference features of multiple candidate dimensions, perform correlation analysis on the service reference features and the subject reference features of the candidate dimensions to obtain correlation results, and select the subject reference features of the candidate dimensions whose correlation results satisfy the reference conditions from the multiple candidate dimensions as the subject reference features of the corresponding dimensions. The service reference features are associated with the service platform to which the service subject belongs, and the service reference features are used as a reference for the service platform dimension when processing the data to be processed. The first processing module is configured to process the data to be processed using subject reference features of multiple dimensions to obtain a first processing result corresponding to the dimensions of each subject reference feature; The second determination module is configured to execute the first processing result corresponding to the subject reference features of each candidate dimension for determining the business data. The second processing module is configured to process the first processing result according to the service reference features to obtain a second processing result corresponding to the service subject.

10. The apparatus according to claim 9, characterized in that, The second processing module is configured to execute: The determination submodule is configured to perform subject reference features for each dimension and determine reference description information corresponding to the subject reference features, wherein the reference description information is used to describe the importance of the corresponding subject reference features; The processing submodule is configured to process the first processing result according to the reference description information to obtain a second processing result corresponding to the service subject.

11. The apparatus according to claim 9, characterized in that, The first determining module is configured to execute: From the subject reference features of the multiple candidate dimensions, the subject reference features of the candidate dimensions that satisfy the first reference condition with positive correlation are selected as the subject reference features of the corresponding dimensions, and the subject reference features of the candidate dimensions that satisfy the second reference condition with negative correlation are selected as the subject reference features of the corresponding dimensions.

12. The apparatus according to claim 10, characterized in that, The determination submodule is configured to execute: Construct a positive sample set and a negative sample set. The positive sample set includes: main reference features of candidate dimensions with positive correlation satisfying the first reference condition. The negative sample set includes: main reference features of candidate dimensions with negative correlation satisfying the second reference condition. The positive sample set, the negative sample set, and the service reference features are input into a pre-trained network model to obtain reference description information output by the network model corresponding to the subject reference features, wherein the network model is used to obtain the reference description information.

13. The apparatus according to claim 10, characterized in that, The second processing module is configured to execute: The dimensions of the main reference features are classified to determine at least one processing dimension; Determine the main reference features of the dimension corresponding to the processing dimension; Based on the reference description information of the subject reference features of the dimension corresponding to the processing dimension, determine the dimension description information corresponding to the processing dimension; The first processing result is optimized based on the processing dimension to obtain the dimension processing result; Based on the dimension processing result and the dimension description information, a second processing result corresponding to the service subject is determined.

14. The apparatus according to claim 13, characterized in that, The second processing module is configured to execute: A weighted average is applied to the reference description information of the subject reference features of the dimension corresponding to the processing dimension, and the result of the weighted average is used as the dimension description information.

15. The apparatus according to claim 13, characterized in that, The second processing module is configured to execute: Determine the number of orders for the service provider and the average number of orders for the industry to which the service provider belongs; Determine the fitting function corresponding to the processing dimension; The fitting function is adjusted based on the comparison between the order quantity and the average order quantity to obtain the target fitting function; The first processing result is optimized using the target fitting function to obtain the dimension processing result.

16. The apparatus according to claim 13, characterized in that, The second processing module is configured to execute: The first processing result is optimized using a fitting function corresponding to the processing dimension to obtain the dimension processing result.

17. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the method as described in any one of claims 1 to 8.

18. A storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the method as described in any one of claims 1 to 8.

19. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method of any one of claims 1 to 8.

Images