Method, device and electronic device for determining internet service information

Abstract

The present invention discloses a method, device, and electronic device for determining internet service information, belonging to the field of internet technology. The method comprises: receiving a first input; in response to the first input, obtaining behavioral characteristic data of a foreground user on a target internet page, the behavioral characteristic data including consultation request data; determining the value of each indicator in N-dimensional indicators based on the behavioral characteristic data, where N is an integer greater than 1, and the N-dimensional indicators are used to quantify the behavioral characteristics from N dimensions; converting the values ​​of all indicators in the N-dimensional indicators into a two-dimensional single-channel image; and calculating the two-dimensional single-channel image based on a preset network model to predict a first service person matching the consultation request data. Embodiments of the present invention can improve the efficiency of consulting services.

Description

Technical Field

[0001] The present invention belongs to the field of Internet technology, and in particular relates to a method, device and electronic device for determining Internet service information. Background Art

[0002] In related technologies, when consulting users conduct Internet consultations, the consulting content they input will be sent to a unified service staff. However, as the fields involved in Internet services become more and more extensive, there may be a problem that the service staff's knowledge is limited, and thus they are unable to provide consulting services to a large number of consulting users, thereby reducing the efficiency of Internet services. Summary of the Invention

[0003] The purpose of the present invention is to provide a method, device and electronic device for determining Internet service information, which can solve the problem of low service efficiency of Internet services in related technologies where the consultation content input by the consulting user is sent to a unified service personnel.

[0004] In order to solve the above-mentioned technical problems, the present invention is achieved as follows:

[0005] In a first aspect, the present invention provides a method for determining Internet service information, comprising:

[0006] receiving a first input;

[0007] In response to the first input, obtaining behavioral characteristic data of a foreground user on a target Internet page, the behavioral characteristic data including consultation request data;

[0008] Determine, based on the behavior characteristic data, a value of each indicator in an N-dimensional indicator, where N is an integer greater than 1, and the N-dimensional indicator is used to quantify the behavior characteristic from N dimensions;

[0009] Convert the values ​​of all the indicators in the N-dimensional indicators into a two-dimensional single-channel image;

[0010] The two-dimensional single-channel image is calculated based on a preset network model to predict a first service personnel matching the consultation request data.

[0011] In a second aspect, the present invention further provides an Internet service information determination device, comprising:

[0012] A receiving module, configured to receive a first input;

[0013] a first acquisition module, configured to acquire, in response to the first input, behavioral characteristic data of a foreground user on a target Internet page, the behavioral characteristic data including consultation request data;

[0014] A first determining module is configured to determine, based on the behavior feature data, a value of each indicator in an N-dimensional indicator, where N is an integer greater than 1, and the N-dimensional indicator is configured to quantify the behavior feature from N dimensions;

[0015] A conversion module, configured to convert the values ​​of all the indicators in the N-dimensional indicators into a two-dimensional single-channel image;

[0016] A calculation module is used to calculate the two-dimensional single-channel image based on a preset network model to predict a first service personnel matching the consultation request data.

[0017] In a third aspect, the present invention further provides an electronic device comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the first aspect.

[0018] In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0019] In an embodiment of the present invention, behavioral feature data of a front-end user on a target Internet page can be obtained, and an N-dimensional indicator capable of quantifying the behavioral feature from N dimensions can be obtained. After the value of the N-dimensional indicator is converted into a two-dimensional single-channel image, a preset network model can be used to calculate the two-dimensional single-channel image to predict a first service personnel who matches the consultation request data. In this way, the front-end user can be provided with a first service personnel who is more matched with his or her behavioral features, the content to be consulted, etc., thereby improving the efficiency of the consultation service. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of a method for determining Internet service information provided by the present invention;

[0021] Figure 2 This is a schematic diagram of the framework of the Inception network model;

[0022] Figure 3 This is a schematic diagram of the interactive process of consulting a front desk in a method for determining Internet service information provided by the present invention;

[0023] Figure 4 This is a schematic diagram of the interactive process of the consultation background in the Internet service information determination method provided by the present invention;

[0024] Figure 5 is a flow chart of another method for determining Internet service information provided by the present invention;

[0025] Figure 6 It is a structural diagram of an Internet service information determination device provided by the present invention;

[0026] Figure 7 This is a structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0027] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0028] The terms "first," "second," and the like in the specification and claims of the present invention are used to distinguish similar objects, and are not used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that embodiments of the present invention can be implemented in orders other than those illustrated or described herein. Furthermore, the objects distinguished by "first," "second," and the like generally refer to a class of objects and do not limit the number of objects. For example, the first object may be one or more.

[0029] The Internet service information determination method provided by the present invention can be applied to a variety of Internets with consulting service functions. For the sake of convenience, in the following embodiments, the Internet service information determination method provided by the present invention is applied to the industrial Internet as an example to illustrate the Internet service information determination method provided by the present invention.

[0030] In practice, the Industrial Internet touches various industry segments, each with its own unique needs and expertise. Therefore, users require extensive professional consultation when purchasing industrial products or acquiring solutions. How to better answer customer inquiries and seize business opportunities is a key issue facing the Industrial Internet.

[0031] It should be noted that in the related art, the user's consultation questions are simply recorded, and then the consultation questions are classified and answered. However, there is no good solution for how to notify the appropriate questions to the solvers who can answer the questions in a timely and accurate manner.

[0032] In this regard, the Internet service information determination method provided by the embodiment of the present invention collects and organizes user behavior data into usable indicators, and then uses a preset network model to predict the user's possible preferences and other behavioral characteristics. When the user submits consultation request data at the front desk, the first service personnel will be recommended to the user based on the behavioral characteristics, so that the first service personnel can solve the user's consultation request more efficiently.

[0033] To facilitate understanding of the Internet service information determination method, Internet service information determination device, and electronic device provided by the present invention, the present invention is described below with reference to the accompanying drawings:

[0034] See also Figure 1 , a method for determining Internet service information provided by an embodiment of the present application may include the following steps:

[0035] Step 101: Receive a first input.

[0036] In practice, the first input may be any input by the front-end user on the target internet page. For example, the first input may be one or more of the following operations: opening the target internet page, browsing the target internet page, clicking on a product on the target internet page, entering consultation request data into the consultation service portal on the target internet page, adding a product to favorites, etc., and the examples are not exhaustive.

[0037] Step 102: In response to the first input, obtain behavioral characteristic data of the foreground user on the target Internet page, wherein the behavioral characteristic data includes consultation request data.

[0038] In implementation, the above-mentioned behavioral feature data can be any data and / or log and / or process behavioral data generated based on the above-mentioned first input. In implementation, the behavioral data of the front-end user can be collected by pre-setting a tracking point on the front end.

[0039] The aforementioned consultation request data may be input by a front-end user into a consultation service portal on a target internet page. Accordingly, through steps 103 through 105, the back-end of the target internet can match a first service person corresponding to the request information and recommend the first service person to provide consulting services to the user. For example, if a front-end user inputs their desired product request information into the consultation service portal on the front-end of the target internet, the back-end of the target internet can locate professional B based on the user's behavioral characteristic data and notify professional B of the request information, allowing professional B to recommend a suitable product based on the front-end user's desired product request information.

[0040] Among them, the above-mentioned consultation request data can be the consultation data entered by the front-end user in the consultation service entrance. In implementation, in order to prevent consultation requests submitted maliciously, the consultation content can be verified, and the corresponding consultation request data is generated only after verification.

[0041] For example: As shown in 3, when the backend administrator sets the target portal column to be consultable (i.e., the consultation service entrance is displayed in the web page corresponding to the target portal column), the front-end user can click on the consultation service entrance and enter the consultation content in the consultation service entrance. At this time, in response to the front-end user's click operation on the consultation service entrance, the service consultation platform can be requested to generate a verification code, and the verification code is returned to the target Internet page and stored, wherein the target Internet page includes the above-mentioned consultation service entrance. In this way, the front-end user also needs to enter the verification code in the verification box of the consultation service entrance, so that when the front-end user submits the consultation content, the service consultation platform can match the verification code entered by the front-end user with the stored verification code, and only generate the consultation request data corresponding to the consultation content if the match is successful.

[0042] Step 103: Determine the value of each indicator in N-dimensional indicators based on the behavior feature data, where N is an integer greater than 1, and the N-dimensional indicators are used to quantify the behavior feature from N dimensions.

[0043] In implementation, the above-mentioned N-dimensional indicators can be indicators that can quantify the behavioral characteristic data from N dimensions, such as: traffic time of a certain product or page, number of clicks on a certain product in a day, an hour and a week, etc.

[0044] As an optional implementation manner, determining the value of each indicator in the N-dimensional indicators based on the behavioral characteristic data includes:

[0045] Determining a preset M-dimensional index based on the behavioral characteristic data, where M is an integer greater than N;

[0046] Based on the random forest algorithm, the M-dimensional indicators are screened to obtain N-dimensional indicators;

[0047] Determine the value of each indicator in the N indicator.

[0048] In practice, the aforementioned M-dimensional indicators can be pre-set indicators of a large number of dimensions. However, too many indicators will, firstly, increase the complexity of the subsequent preset network model, thereby increasing computational overhead; secondly, some indicators may have multicollinearity, thereby interfering with the prediction accuracy of the preset network model. In the embodiment of the present application, the M-dimensional indicators are screened using a highly robust random forest algorithm, so that the robustness of the screened N-dimensional indicators is higher than that of the M-dimensional indicators, thereby reducing computational overhead while also improving the prediction accuracy of the preset network model.

[0049] For example, the aforementioned behavioral feature data can be stored in the Spark big data platform. This allows Spark's superior performance in big data processing to be leveraged to perform preliminary data cleansing. This data is then calculated and organized into multi-dimensional metrics based on operational statistical indicators, ultimately yielding over 1,500 metrics. These metrics include basic metrics (e.g., the number of times front-end users view or click on a product), time series metrics broken down by time (e.g., clicks within one day, clicks within two days), and cross-metrics derived by combining metrics from various dimensions (e.g., the percentage of products added to favorites after being clicked within one day, the percentage of products purchased after being viewed within three days, and other cross-metrics).

[0050] Optionally, the random forest algorithm is used to screen the M-dimensional indicators to obtain N-dimensional indicators, including:

[0051] Determine the feature importance coefficient of each indicator in the M-dimensional indicators based on the random forest algorithm, wherein the feature importance coefficient is positively correlated with the Gini coefficient of the corresponding indicator;

[0052] An N-dimensional indicator whose feature importance coefficient is greater than or equal to a preset value is selected from the M-dimensional indicators.

[0053] In practice, the above feature importance coefficient can be calculated by the following formula:

[0054]

[0055] Among them, H j Represents the characteristic importance coefficient of indicator j, VIM j represents the Gini coefficient of indicator j, Represents the sum of gains of all indicators.

[0056] The aforementioned selection of N-dimensional indicators whose feature importance coefficients are greater than or equal to a preset value from the M-dimensional indicators may be performed by arranging the M-dimensional indicators in descending order of their feature importance coefficients and selecting the N-dimensional indicators ranked in the top N positions. A larger feature importance coefficient indicates a higher degree of importance for the corresponding indicator. In this way, selecting N-dimensional indicators with higher importance from the M-dimensional indicators can reduce computational overhead while also improving the prediction accuracy of the preset network model.

[0057] Step 104: Convert the values ​​of all the indicators in the N-dimensional indicators into a two-dimensional single-channel image.

[0058] In implementation, the above-mentioned two-dimensional single-channel image can also be called a two-dimensional grayscale image. The storage form of the two-dimensional single-channel image in the computer is actually a two-dimensional array, which can be converted into the above-mentioned two-dimensional single-channel image through the Python Image Library (PIL). For example: assuming that N is equal to 784, the two-dimensional single-channel image can be equivalent to a 28×28 two-dimensional array.

[0059] Step 105: Calculate the two-dimensional single-channel image based on a preset network model to predict a first service personnel matching the consultation request data.

[0060] The preset network model can be a pre-trained Inception network model. This allows the user's likely preferences to be predicted using the Inception network model, which has proven effective in image recognition. The Inception network model was first proposed by Christian Szegedy et al. in their paper "Going Deeper with Convolutions" and is primarily used for two-dimensional image processing.

[0061] Of course, in implementation, the above-mentioned preset network model can also be other network models, such as a neural network model, etc., which is not specifically limited here.

[0062] In implementation, converting the N-dimensional index into a two-dimensional single-channel image can simplify the data structure of the N-dimensional index, thereby simplifying the model complexity of the above-mentioned preset network model.

[0063] In implementation, the two-dimensional single-channel image is calculated through a preset network model to predict the first service personnel, which can be a service personnel that matches the behavioral habits, preferences, etc. of the front-end user. In this way, the service personnel can provide more efficient consulting services to the front-end user.

[0064] As an optional implementation, the preset network model includes a preset Inception network model, and the calculating of the two-dimensional single-channel image based on the preset network model to predict a first service personnel matching the consultation request data includes:

[0065] Input the two-dimensional single-channel image into the preset Inception network model;

[0066] According to the output result of the preset Inception network model, a first service personnel matching the consultation request data is predicted.

[0067] In the implementation, the preset Inception network model includes a convolutional layer, a maximum pooling layer, a first Inception model layer, a second Inception model layer, a global average pooling layer and an output layer;

[0068] The convolution layer is used to extract the first feature of the N-dimensional indicator, the maximum pooling layer is used to perform maximum pooling processing on the first feature, and input the first feature after the maximum pooling processing into the first Inception model layer;

[0069] The first Inception model layer is used to extract the second feature from the first feature after the maximum pooling process;

[0070] The second Inception model layer is used to extract a third feature from the second feature;

[0071] The global average pooling layer is used to perform global average pooling processing on the second feature to obtain a first prediction result; the global average pooling layer is also used to perform global average pooling processing on the third feature to obtain a second prediction result;

[0072] The output layer is used to determine a first service person matching the consultation request data according to the first prediction result and the second prediction result, and output identification information of the first service person.

[0073] As Figure 2 In the Inception network model shown, the convolutional layer can include two layers of 32 3*3 convolution kernels and one layer of 64 3*3 convolution kernels, so as to use multi-dimensional convolution kernels to extract features simultaneously. The first feature extracted by the maximum pooling convolution layer is then subjected to maximum pooling.

[0074] In addition, the first Inception model layer may include four columns of convolution groups, whereby the first features after maximum pooling are input into the four columns of convolution groups in the first Inception model layer for joint feature extraction. The first column of convolution groups uses 64 1*1 convolution kernels for feature extraction; the second column of convolution groups uses 48 1*1 convolution kernels followed by 64 5*5 convolution kernels for feature extraction; the third column of convolution groups uses 64 1*1 convolution kernels followed by 96 3*3 convolution kernels for feature extraction; and the fourth column of convolution groups performs 3*3 average pooling followed by 32 1*1 convolution kernels for feature extraction. Finally, the second features extracted from the four columns are combined and input into a global average pooling layer, which is then input into a second Inception model layer. The structure of the second Inception model layer is identical to that of the first Inception model layer and is not further described here. After the second Inception model layer extracts the third feature from the second feature, the third feature is input into a global average pooling layer. The global average pooling layer is followed by an output layer (i.e., a softmax layer), and the second feature after global average pooling (i.e., the first prediction result) and the third feature after global average pooling (i.e., the second prediction result) are merged in the softmax layer to obtain a prediction result (i.e., the first service personnel).

[0075] From the above, it can be seen that the preset Inception network model provided by the embodiment of the present invention uses multi-dimensional convolution kernels to simultaneously extract features, thereby improving the diversity of the extracted features, thereby making the preset Inception network model trained according to the diverse features and the prediction results calculated based on the preset Inception network model more reliable; the hidden layer in the preset Inception network model (i.e., the first Inception model layer) can also output the result (i.e., the first prediction result) through the global average pooling layer, and by combining the first prediction result with the second prediction result, the effect of model fusion can be achieved; in addition, using the global average pooling layer instead of the fully connected layer can reduce the amount of parameter calculation.

[0076] It should be noted that the preset Inception network model can also predict products that the front-end user may be interested in based on the value of each indicator in the above-mentioned N-dimensional indicators. The process of using the preset Inception network model to predict products that the front-end user may be interested in based on the value of each indicator in the above-mentioned N-dimensional indicators is similar to the process of using the preset Inception network model to predict the first service personnel based on the value of each indicator in the above-mentioned N-dimensional indicators, and will not be repeated here. In this way, after storing products that a front-end user may be interested in, product recommendation information can be displayed on subsequent Internet pages that the front-end user visits to improve the product recommendation effect.

[0077] As an optional implementation, after calculating the two-dimensional single-channel image based on a preset network model to predict a first service personnel matching the consultation request data, the method further includes:

[0078] Obtaining a pre-stored first communication address corresponding to the first service personnel;

[0079] A first prompt message is sent to the first communication address, where the first prompt message is used to prompt the first service personnel to respond to the consultation request data.

[0080] In implementation, the database of the industrial Internet that applies the Internet service information determination method provided by the present invention can store the contact address of each service personnel, such as mobile phone number, email address, account name of social application, etc. In this way, after determining the first service personnel, the first service personnel can be reminded to provide consulting services to the front desk user by sending text messages, making calls, sending emails, sending reminder messages through social applications, etc., so that the front desk user's consulting requests can be responded to in a timely manner.

[0081] As an optional implementation manner, the Internet service information determination method further includes:

[0082] Obtaining a pre-configured first association relationship, wherein the first association relationship includes an association relationship between each pre-configured role and a corresponding consultation column;

[0083] In a case where the consultation request data is data input in a target consultation column, determining a target role corresponding to the target consultation column according to the first association relationship;

[0084] Determining a second service personnel pre-configured with attributes of the target role;

[0085] Obtaining a pre-stored second communication address corresponding to the second service personnel;

[0086] A second prompt message is sent to the second communication address, where the second prompt message is used to prompt the second service personnel to respond to the consultation request data.

[0087] In implementation, the first association relationship may be an association relationship between each portal column configured by a backend user and its corresponding role.

[0088] Among them, the above-mentioned portal columns can be product themes, consulting themes, etc., and the above-mentioned roles can be parameters for distinguishing permissions configured by the backend user through the backend management system, and each service personnel can have their own role attributes. For example: assuming that the backend user configures role A, role B and role C, and configures role A to be associated with consulting column A, configures role B to be associated with consulting column B, and configures role C to be associated with consulting column C, then when the front-end user submits consulting request data through consulting column A, a second service personnel with the attributes of role A can be recommended to provide consulting services to the front-end user. It should be noted that in actual applications, a role can be associated with one or at least two consulting columns, and / or, a consulting column can be associated with one or at least two roles, and / or a service personnel can have one or at least two role attributes, which will not be elaborated here.

[0089] For example: Figure 4 As shown in the figure, the role configuration process includes:

[0090] Step 1: The administrator logs in to the backend of the consulting service, adds a new role A, and associates the role with the relevant service personnel A, for example, by binding through ID. The user center stores the relationship between each role and its associated service personnel, as well as the communication information of each service personnel;

[0091] Step 2: The administrator configures the consulting column that the role A will receive the consulting request data from (i.e., the first association relationship), and saves the configuration;

[0092] Step 3: The user submits a consultation request data, and the backend saves the consultation request data;

[0093] Step 4: According to the consultation column consulted by the consultation request data, obtain the corresponding configured role (assuming that the role is role A);

[0094] Step 5: Go to the user center based on role A to check if service person A is configured for role A;

[0095] Step 6: Send a reminder message to service staff A, reminding service staff A to go online to handle the consultation request.

[0096] It should be noted that in the embodiment of the present invention, service personnel can only see the consulting content of the configured associated columns (configured associated columns indicate columns with the same role as the service personnel), and the consulting content of other columns is not visible, which can improve data security.

[0097] In this implementation, consulting services are integrated with the Industrial Internet portal. By configuring specific roles for portal columns and assigning service personnel to these roles, when a front-end user submits a consultation request based on a portal column, the service personnel with the role associated with that portal column will immediately receive a prompt and respond accordingly. This allows both the primary and secondary service personnel to be selected to respond to the front-end user's consultation request, increasing the probability that the service personnel recommended to the front-end user can effectively answer their consultation request, thereby improving the reliability of the consulting service.

[0098] As an optional implementation, before obtaining the behavior characteristic data of the foreground user on the target Internet page, wherein the behavior characteristic data includes consultation request data, the method further includes:

[0099] Receiving a Hypertext Markup Language (HTML) template of the target Internet page;

[0100] The target Internet page is updated according to the HTML template of the target Internet page, wherein the updated target Internet page includes a consulting service portal, and the consulting request data is data obtained from the consulting service portal.

[0101] In practice, the HTML template can be uploaded by a backend administrator to the target internet server. Uploading the HTML template can then update (add / delete) the consulting service entry on the target internet page. For example, a backend administrator can set up a target internet page at any time where they want to display the consulting service entry and upload the HTML template. Then, by refreshing the data structure (redis) cache server, the target internet page on the front end will display the consulting service entry.

[0102] In this embodiment, the backend administrator can add or delete consulting service entrances in any page of the industrial Internet portal in real time through backend configuration, realizing the free combination of consulting service entrances and any pages, and facilitating users in various sub-industries in the industrial Internet to consult related topics.

[0103] In order to facilitate understanding of the Internet service information determination method provided by the embodiment of the present invention, Figure 5The flowchart shown in FIG. 1 is used as an example to illustrate the method for determining Internet service information provided by an embodiment of the present invention. Figure 5 As shown, the method for determining Internet service information may include the following steps:

[0104] Step 1: The user enters the consulting service portal on the target internet page, and the consulting module transmits the verification code to the front desk;

[0105] Step 2: Provide consultable columns to the front desk through the portal module;

[0106] Step 3: The front-end user submits the consultation request data;

[0107] Step 4: Save the user's behavioral characteristics data to the Spark big data platform through front-end tracking.

[0108] Step 5: The Spark big data platform calculates the behavioral characteristic data of front-end users in real time to obtain usable indicators (before this step, the management staff has completed the association between service personnel, roles and columns);

[0109] Step 6: The Inception network model algorithm engine calculates data from the Spark big data platform and predicts the products of interest and the matching first service personnel;

[0110] Step 7: Synchronize the calculation results of the Inception network model algorithm engine to the feature database;

[0111] Step 8: After verifying that the verification code entered by the front-end user is correct, the back-end saves the consultation request data;

[0112] Step 9: Query the corresponding role according to the column corresponding to the consultation request data;

[0113] Step 10: Based on the role, query the contact address of the service personnel with the role in the user center;

[0114] Step 10: Notify the second service personnel to go online and handle the consultation request based on the communication address;

[0115] Step 11: Query the matching first service personnel from the feature database and notify them to go online to handle the consultation request.

[0116] From the above, it can be seen that the binding between service personnel, roles, and consultation columns mentioned in the present invention can ensure that consultations in a specific sub-industry are only notified to relevant service personnel, and relevant service personnel can see such consultation content, thereby improving consultation processing efficiency and consultation processing results; in addition, the present invention uses the Spark big data platform and an algorithm engine based on the Inception network model to predict products that users are interested in and related matching service personnel, which can better play the role of the service consultation system in promoting the industrial Internet.

[0117] See also Figure 6 , is a structural diagram of an Internet service information determination device provided by an embodiment of the present invention, such as Figure 6 As shown, the Internet service information determination device 600 includes:

[0118] Receiving module 601, configured to receive a first input;

[0119] A first acquisition module 602 is configured to acquire, in response to the first input, behavioral characteristic data of a foreground user on a target Internet page, the behavioral characteristic data including consultation request data;

[0120] A first determining module 603 is configured to determine, based on the behavior feature data, a value of each of N-dimensional indicators, where N is an integer greater than 1, and the N-dimensional indicators are used to quantify the behavior feature from N dimensions;

[0121] A conversion module 604 is configured to convert the values ​​of all the indicators in the N-dimensional indicators into a two-dimensional single-channel image;

[0122] The calculation module 605 is configured to calculate the two-dimensional single-channel image based on a preset network model to predict a first service person matching the consultation request data.

[0123] Optionally, the Internet service information determining device 600 further includes:

[0124] A second acquisition module is used to acquire a pre-stored first communication address corresponding to the first service personnel;

[0125] The first sending module is used to send a first prompt message to the first communication address, where the first prompt message is used to prompt the first service personnel to respond to the consultation request data.

[0126] Optionally, the Internet service information determining device 600 further includes:

[0127] A third acquisition module is used to acquire a pre-configured first association relationship, wherein the first association relationship includes a pre-configured association relationship between each role and a corresponding consultation column;

[0128] A second determining module is configured to determine, when the consultation request data is data input in a target consultation column, a target role corresponding to the target consultation column according to the first association relationship;

[0129] A third determining module is used to determine a second service personnel pre-configured with attributes of the target role;

[0130] a fourth acquiring module, configured to acquire a pre-stored second communication address corresponding to the second service personnel;

[0131] The second sending module is used to send a second prompt message to the second communication address, where the second prompt message is used to prompt the second service personnel to respond to the consultation request data.

[0132] Optionally, the first determining module 603 includes:

[0133] A first determining unit is configured to determine a preset M-dimensional index based on the behavior characteristic data, where M is an integer greater than N;

[0134] A screening unit, configured to screen the M-dimensional indicators based on a random forest algorithm to obtain an N-dimensional indicator;

[0135] The second determining unit is used to determine the value of each indicator in the N indicators.

[0136] Optionally, the screening unit includes:

[0137] a determination subunit, configured to determine a feature importance coefficient of each indicator in the M-dimensional indicators based on a random forest algorithm, wherein the feature importance coefficient is positively correlated with the Gini coefficient of the corresponding indicator;

[0138] The screening subunit is used to select N-dimensional indicators whose feature importance coefficients are greater than or equal to a preset value from the M-dimensional indicators.

[0139] Optionally, the preset network model includes a preset Inception network model, and the calculation module 605 includes:

[0140] An input unit, configured to input the two-dimensional single-channel image into a preset Inception network model;

[0141] A prediction unit is used to predict a first service personnel matching the consultation request data based on the output result of the preset Inception network model.

[0142] Optionally, the preset Inception network model includes a convolutional layer, a maximum pooling layer, a first Inception model layer, a second Inception model layer, a global average pooling layer and an output layer;

[0143] The convolution layer is used to extract the first feature of the N-dimensional indicator, the maximum pooling layer is used to perform maximum pooling processing on the first feature, and input the first feature after the maximum pooling processing into the first Inception model layer;

[0144] The first Inception model layer is used to extract the second feature from the first feature after the maximum pooling process;

[0145] The second Inception model layer is used to extract a third feature from the second feature;

[0146] The global average pooling layer is used to perform global average pooling processing on the second feature to obtain a first prediction result; the global average pooling layer is also used to perform global average pooling processing on the third feature to obtain a second prediction result;

[0147] The output layer is used to determine a first service person matching the consultation request data according to the first prediction result and the second prediction result, and output identification information of the first service person.

[0148] Optionally, the Internet service information determining device 600 further includes:

[0149] A receiving module, configured to receive a Hypertext Markup Language (HTML) template of the target Internet page;

[0150] An updating module is used to update the target Internet page according to the HTML template of the target Internet page, wherein the updated target Internet page includes a consulting service entrance, and the consulting request data is data obtained from the consulting service entrance.

[0151] The Internet service information determination device 600 provided by the embodiment of the present invention can achieve Figure 1 or Figure 5 The various processes implemented in the method embodiment shown can achieve the same beneficial effects, and will not be described again here to avoid repetition.

[0152] Optional, such as Figure 7 As shown, an embodiment of the present invention further provides an electronic device 700, including a processor 701, a memory 702, a program or instruction stored in the memory 702 and executable on the processor 701, and the program or instruction is executed by the processor 701 to implement the following Figure 1The various processes of the method embodiment shown can achieve the same technical effect, and to avoid repetition, they will not be described here.

[0153] The embodiment of the present invention further provides a computer-readable storage medium, wherein a program or instruction is stored on the computer-readable storage medium, and when the program or instruction is executed by a processor, the following is realized: Figure 1 The various processes of the method embodiment shown can achieve the same technical effect, and to avoid repetition, they will not be described here.

[0154] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk.

[0155] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present invention is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0156] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0157] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

Claims

1. A method for determining Internet service information, characterized in that: include: receiving a first input; In response to the first input, obtaining behavioral characteristic data of a foreground user on a target Internet page, the behavioral characteristic data including consultation request data; Determine, based on the behavior characteristic data, a value of each indicator in an N-dimensional indicator, where N is an integer greater than 1, and the N-dimensional indicator is used to quantify the behavior characteristic from N dimensions; Convert the values ​​of all the indicators in the N-dimensional indicators into a two-dimensional single-channel image; The two-dimensional single-channel image is calculated based on a preset network model to predict a first service personnel matching the consultation request data.

2. The method according to claim 1, characterized in that After calculating the two-dimensional single-channel image based on the preset network model to predict a first service personnel matching the consultation request data, the method further includes: Obtaining a pre-stored first communication address corresponding to the first service personnel; A first prompt message is sent to the first communication address, where the first prompt message is used to prompt the first service personnel to respond to the consultation request data.

3. The method according to claim 2, characterized in that The method further comprises: Obtaining a pre-configured first association relationship, wherein the first association relationship includes an association relationship between each pre-configured role and a corresponding consultation column; In a case where the consultation request data is data input in a target consultation column, determining a target role corresponding to the target consultation column according to the first association relationship; Determining a second service personnel pre-configured with attributes of the target role; Obtaining a pre-stored second communication address corresponding to the second service personnel; A second prompt message is sent to the second communication address, where the second prompt message is used to prompt the second service personnel to respond to the consultation request data.

4. The method according to any one of claims 1 to 3, characterized in that Determining the value of each indicator in the N-dimensional indicators based on the behavioral characteristic data includes: Determining a preset M-dimensional index based on the behavioral characteristic data, where M is an integer greater than N; Based on the random forest algorithm, the M-dimensional indicators are screened to obtain N-dimensional indicators; Determine the value of each indicator in the N indicator.

5. The method according to claim 4, characterized in that The random forest algorithm is used to screen the M-dimensional indicators to obtain the N-dimensional indicators, including: Determine the feature importance coefficient of each indicator in the M-dimensional indicators based on the random forest algorithm, wherein the feature importance coefficient is positively correlated with the Gini coefficient of the corresponding indicator; An N-dimensional indicator whose feature importance coefficient is greater than or equal to a preset value is selected from the M-dimensional indicators.

6. The method according to claim 4, characterized in that The preset network model includes a preset Inception network model, and the calculation of the two-dimensional single-channel image based on the preset network model to predict a first service personnel matching the consultation request data includes: Input the two-dimensional single-channel image into the preset Inception network model; According to the output result of the preset Inception network model, a first service personnel matching the consultation request data is predicted.

7. The method according to claim 6, characterized in that The preset Inception network model includes a convolutional layer, a maximum pooling layer, a first Inception model layer, a second Inception model layer, a global average pooling layer and an output layer; The convolution layer is used to extract the first feature of the N-dimensional indicator, the maximum pooling layer is used to perform maximum pooling processing on the first feature, and input the first feature after the maximum pooling processing into the first Inception model layer; The first Inception model layer is used to extract the second feature from the first feature after the maximum pooling process; The second Inception model layer is used to extract a third feature from the second feature; The global average pooling layer is used to perform global average pooling processing on the second feature to obtain a first prediction result; the global average pooling layer is also used to perform global average pooling processing on the third feature to obtain a second prediction result; The output layer is used to determine a first service person matching the consultation request data according to the first prediction result and the second prediction result, and output identification information of the first service person.

8. The method according to claim 1, characterized in that Before obtaining the behavior characteristic data of the foreground user on the target Internet page, wherein the behavior characteristic data includes consultation request data, the method further includes: Receiving a Hypertext Markup Language (HTML) template of the target Internet page; The target Internet page is updated according to the HTML template of the target Internet page, wherein the updated target Internet page includes a consulting service portal, and the consulting request data is data obtained from the consulting service portal.

9. An Internet service information determination device, characterized in that: include: A receiving module, configured to receive a first input; a first acquisition module, configured to acquire, in response to the first input, behavioral characteristic data of a foreground user on a target Internet page, the behavioral characteristic data including consultation request data; A first determining module is configured to determine, based on the behavior feature data, a value of each indicator in an N-dimensional indicator, where N is an integer greater than 1, and the N-dimensional indicator is configured to quantify the behavior feature from N dimensions; A conversion module, configured to convert the values ​​of all the indicators in the N-dimensional indicators into a two-dimensional single-channel image; A calculation module is used to calculate the two-dimensional single-channel image based on a preset network model to predict a first service personnel matching the consultation request data.

10. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps in the method for determining Internet service information as claimed in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the method for determining Internet service information according to any one of claims 1 to 8.

Images