Information recommendation method and device, equipment, storage medium and program product

Abstract

The invention provides an information recommendation method and device, equipment, a storage medium and a program product, and the method comprises the steps: obtaining object association information and call moment information of a call object, the object association information comprises object attribute information and call information, and the call moment information comprises call moment information; the call information comprises a connection state, a call duration and an information recommendation record of the call object, and the call moment information comprises a first number of call moments; determining the call completing rate of the call object at the first number of call moments according to the object association information and the call moment information; determining the call moment of the call object according to the call completing rate of the first number of call moments; and recommending information to the calling object based on the calling moment of the calling object. According to the embodiment of the invention, the call completing rate of the call object and the accuracy of determining the call moment can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to an information recommendation method and device, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] In common recommendation campaign scenarios, it's often necessary to consider personalized recommendation timing that suits users, namely, user-preferred timing, to maximize campaign effectiveness and maximize benefits. A user's preferred timing primarily depends on their availability at a given moment and their willingness to purchase the recommended product. In current applications, before making outbound recommendations to users, relevant user information is typically obtained to predict the user's connection rate. Therefore, determining the user's connection rate for recommended calls and the timing of outbound calls based on their availability and willingness to purchase the recommended product is crucial. However, current methods for predicting user connection rates often lack accuracy, further reducing the accuracy of determining the timing of outbound calls and increasing the failure rate of recommendations. Summary of the Invention

[0003] The present disclosure provides an information recommendation method and apparatus, an electronic device, a computer-readable storage medium, and a computer program product.

[0004] In a first aspect, the present disclosure provides an information recommendation method, comprising:

[0005] Obtaining object association information and call time information of a call object, wherein the object association information includes object attribute information and call information, the call information includes a connection status, call duration, and information recommendation record of the call object, and the call time information includes a first number of call times;

[0006] determining, based on the object association information and the call time information, a connection rate of the call object at the first number of call times;

[0007] determining a calling time of the calling party according to connection rates of the first number of calling times;

[0008] Information is recommended to the calling party based on the calling time of the calling party.

[0009] In a second aspect, the present disclosure provides an information recommendation device, comprising:

[0010] an acquisition module configured to acquire object association information and call time information of a call object, wherein the object association information includes object attribute information and call information, the call information includes a connection status, call duration, and information recommendation record of the call object, and the call time information includes a first number of call times;

[0011] a first determining module configured to determine a connection rate of the call object at the first number of call moments according to the object association information and the call moment information;

[0012] A second determining module is configured to determine a calling time of the calling object according to the connection rates of the first number of calling times;

[0013] The recommendation module is configured to recommend information to the calling party based on the calling time of the calling party.

[0014] In a third aspect, the present disclosure provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores one or more computer programs executable by the at least one processor, and the one or more computer programs are executed by the at least one processor so that the at least one processor can execute the above-mentioned information recommendation method.

[0015] In a fourth aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the above-mentioned information recommendation method when executed by a processor.

[0016] In a fifth aspect, the present disclosure provides a computer program product comprising a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above-mentioned information recommendation method.

[0017] The information recommendation method provided by the embodiment of the present disclosure can, after obtaining the object association information and call time information of the call object, predict the connection rate of the call object at the first number of call times based on the object association information and call time information, thereby obtaining the connection rate of the call object at the first number of call times, and determining the call time of the call object in the first number of call times based on the connection rate, so as to call the call object in the subsequent process based on the call time of the call object, thereby calling the call object according to the call time of the call object, which can improve the connection rate of the call object and the accuracy of determining the call time of the call object, and further improve the success rate of information recommendation for the call object based on the call time.

[0018] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. The above and other features and advantages will become more apparent to those skilled in the art by describing detailed example embodiments with reference to the accompanying drawings. In the accompanying drawings:

[0020] Figure 1 This is an application scenario diagram of an information recommendation method and device provided by an embodiment of the present disclosure;

[0021] Figure 2 is a flow chart of an information recommendation method provided by an embodiment of the present disclosure;

[0022] Figure 3 is a model architecture diagram of a first model provided by an embodiment of the present disclosure;

[0023] Figure 4 is a block diagram of an information recommendation device provided by an embodiment of the present disclosure;

[0024] Figure 5 It is a block diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] To enable those skilled in the art to better understand the technical solutions of the present disclosure, exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0026] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.

[0027] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0028] The terms used herein are only used to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a" and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "comprising" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof is not excluded. Similar words such as "connected" or "connected" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0029] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless expressly defined as such herein.

[0030] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution complies with relevant national laws and regulations (for example, the "Information Security Technology Personal Information Security Specification", etc.). For example: corresponding prescribed measures are taken to control access to personal information; the display of personal information is subject to prescribed restrictions; the purpose of using personal information does not exceed the scope of direct or reasonable connection; when using personal information, clear identity reference is eliminated to avoid precise positioning of specific individuals.

[0031] In typical telephone recommendation campaigns, it's often necessary to consider personalized recommendation timing that's tailored to the user—that is, the user's preferred timing—to maximize the effectiveness of the recommendation and maximize its value. The user's preferred timing primarily depends on whether they're available at a given moment and their willingness to purchase the recommended product. In practice, product introductions, communication, recommendations, and sales are often conducted through non-face-to-face methods like phone or the internet. However, in these scenarios, a successful product recommendation requires the user to accept the call. For example, a busy weekday user might hang up immediately upon receiving a sales call.

[0032] Therefore, in the above scenario, the success rate and efficiency of product recommendations largely depend on the timing of personalized recommendations to users, that is, the moment when a call is made to the user.

[0033] Based on this, embodiments of the present disclosure provide an information recommendation method, an information recommendation device, an electronic device, a computer-readable storage medium, and a computer program product, which are described in detail one by one in the following embodiments.

[0034] Figure 1 The following schematically illustrates an application scenario of the information recommendation method and device provided by an embodiment of the present disclosure.

[0035] like Figure 1 As shown, an application scenario of an embodiment of the present disclosure may include a terminal device 101, a network 103, and a server 102. The network 103 is used as a medium for providing a communication link between the terminal device 101 and the server 102. The network 103 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0036] The user can use the terminal device 101 to interact with the server 102 via the network 103 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0037] The terminal device 101 may be any electronic device having a display screen and supporting web browsing, including but not limited to a smart phone, a tablet computer, a laptop computer, a desktop computer, and the like.

[0038] The server 102 may be a server that provides various services, such as a background management server (for example only) that supports websites browsed by users using the terminal device 101. The background management server may analyze and process received user requests and other data, and feed back the processing results (e.g., web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0039] It should be noted that the information recommendation method and apparatus provided in the embodiments of the present disclosure can be executed by the server 102. Accordingly, the information recommendation method and apparatus provided in the embodiments of the present disclosure can be set in the server 102. The information recommendation method and apparatus provided in the embodiments of the present disclosure can also be executed by a server or server cluster that is different from the server 102 and can communicate with the terminal device 101 and / or the server 102. Accordingly, the information recommendation method and apparatus provided in the embodiments of the present disclosure can also be set in a server or server cluster that is different from the server 102 and can communicate with the terminal device 101 and / or the server 102.

[0040] It should be understood that Figure 1The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0041] See also Figure 2 , Figure 2 A flowchart of an information recommendation method according to an embodiment of the present disclosure is shown, which specifically includes the following steps:

[0042] Step 202: Obtain object association information and call time information of the call object.

[0043] The call object refers to the object to be called, specifically the user to be called. Object-related information refers to information associated with the call object, specifically including object attribute information and call information. Object attribute information includes but is not limited to the name, gender, age, occupation, etc. of the call object (i.e., the calling user). Call information specifically refers to the call object's historical call information, including the call object's connection status, call duration, and information recommendation records within a historical time period. Connection status refers to the call object's call record status at multiple historical call moments, specifically including connected or disconnected status. Call duration refers to the call object's call duration at multiple historical call moments. Information recommendation records refer to the call object's acceptance of recommended information at multiple historical call moments, specifically including whether information recommendations were accepted or not. In practical applications, the recommended information may include products or product information that need to be recommended to the call object. Historical call moments refer to multiple historical call moments within a historical time period. For example, if the historical time period is the past month, the month can be divided into two-hour historical call moments to obtain multiple historical call moments within that month. The call time information includes a first number of call times, which may be at least two. The call times are multiple future call times within a future time interval. For example, if a month is divided into two-hour intervals, multiple call times within the month can be obtained.

[0044] Specifically, after determining the person to be called, i.e., the calling party, object-related information and calling time information of the calling party are obtained. For example, the name, gender, age, and other information of user A, as well as user A's historical calling information, are obtained, along with user A's calling time information j, where j = 0, 1...T-1, representing times 1, 2,...T, respectively. In subsequent processes, based on the obtained object-related information and calling time information, the call connection rate of the calling party at multiple calling times is predicted, thereby determining the calling time for calling the calling party.

[0045] Step 204: Determine the connection rate of the call object at the first number of call times according to the object association information and the call time information.

[0046] After obtaining the object association information and call time information, the connection rate of the call object at multiple call times can be predicted based on the object association information and call time information. In the embodiment provided by the present disclosure, the connection rate of the call object at multiple call times can be predicted based on the first model.

[0047] The first model is a model used to predict the connection rate of the call recipient at a first number of call moments. The first model can be a causal model, specifically a multi-causal relationship model or a unidirectional causal relationship model. In this disclosure, the multi-causal relationship model is preferably used as the first model. The connection rate refers to the probability that the call recipient will answer the call at the call moment.

[0048] The following describes a specific implementation process of predicting the connection rate of the call object at the first number of call moments based on the first model.

[0049] In step 204, determining the connection rate of the call object at the first number of call moments based on the object association information and the call moment information includes:

[0050] Encoding the object association information and the call time information respectively to obtain an object feature vector of the object association information and a time feature vector of the call time information;

[0051] Performing linear transformation on the object feature vector and the moment feature vector respectively to obtain object linear features of the object feature vector and moment linear features of the moment feature vector;

[0052] Based on the object linear feature and the time linear feature, a connection rate of the call object at the first number of call times is determined.

[0053] See also Figure 3 , Figure 3 FIG. 1 shows a model architecture diagram of a first model provided according to an embodiment of the present disclosure. Figure 3 As shown, the first model includes a coding layer, a linear layer and an activation layer, wherein the coding layer is used to encode object association information and call time information to obtain object feature vectors and time feature vectors; the linear layer is used to perform linear transformation on the object feature vectors and the time feature vectors to obtain object linear features and time linear features; the activation layer is used to activate the object linear features and the time linear features (that is, perform nonlinear processing) to obtain the connection rate of the call object at different call times.

[0054] Specifically, based on the encoding layer of the first model, the acquired object association information and call time information are encoded respectively, so as to obtain the object feature vector of the object association information and the moment feature vector of the call time information; based on the linear layer of the first model, the object feature vector and the moment feature vector are linearly transformed respectively, so as to obtain the object linear features and the moment linear features after the linear transformation; based on the activation layer of the first model, the object linear features and the moment linear features are activated using the activation function, so as to obtain the connection rate of the call object at the first number of call moments.

[0055] Furthermore, if Figure 3 As shown, the linear layer specifically includes a multilayer perceptron (MLP) and a linear transformation layer (Linear). The MLP neural network is a type of feedforward neural network. During the model training process, the gradient needs to be calculated by the backpropagation algorithm to update the model parameters. Based on the linear layer of the first model, the object feature vector and the moment feature vector are linearly transformed respectively to obtain the object linear feature and the moment linear feature. Specifically, the object feature vector and the moment feature vector are input into the MLP neural network. After the object feature vector and the moment feature vector are nonlinearly processed by the MLP neural network, they are input into the linear transformation layer for linear transformation to obtain the object linear feature and the moment linear feature.

[0056] In actual applications, whether the user is idle at a certain moment and the user's willingness to accept the product or product information will affect the user's connection rate at a certain moment. Therefore, in order to improve the accuracy of the predicted connection rate, in the embodiment provided by the present disclosure, a causal relationship can be established between the impact of the two influencing factors of whether the user is idle at a certain moment and the user's willingness to accept the product or product information on the connection rate, and the connection rate can be predicted based on the first model according to the constructed causal relationship. Therefore, the activation layer of the first model provided by the present disclosure specifically includes two activation layers, which are used to obtain information willingness and time idleness, as well as the connection rate. The specific implementation process of the activation layer is as follows:

[0057] In a specific embodiment provided by the present disclosure, determining the connection rate of the call object at the first number of call moments based on the object linear feature and the moment linear feature includes:

[0058] activating the linear feature of the object to obtain the information willingness of the call object at the first number of call moments, wherein the information willingness is used to represent the probability of the call object accepting the recommended information;

[0059] activating the object linear feature and the moment linear feature to obtain the moment availability of the call object at the first number of call moments, wherein the moment availability is used to represent the probability of the call object being idle at the call moment;

[0060] Based on the information willingness and the time availability, a connection rate of the call object at the first number of call times is determined.

[0061] Continue to see Figure 3 The activation layer of the first model specifically includes a first activation layer and a second activation layer. The first activation layer is used to activate the linear features of the object and the linear features of the time to obtain information willingness and time availability. The second activation layer is used to activate information willingness and time availability to obtain the connection rate. Information willingness represents the probability that the callee will accept the recommended information. In practical applications, the recommended information can be a product or product information about the product to be recommended to the user. Time availability represents the probability that the callee will be available at the time of the call.

[0062] Specifically, after obtaining the object linear features and the time linear features, the object linear features are activated based on the first activation layer of the first model using the activation function of the first activation layer to obtain the information willingness of the call object at the first number of call moments. Based on the first activation layer of the first model, the object linear features and the time linear features are activated using the activation function of the first activation layer to obtain the time availability of the call object at the first number of call moments. After obtaining the information willingness and time availability, the information willingness and time availability are activated based on the second activation layer of the first model using the activation function of the second activation layer to obtain the connection rate of the call object at the first number of call moments.

[0063] It should be noted that, in actual applications, one or two first activation layers can be deployed in the first model. The specific number can be determined based on the actual application situation. The embodiment of the present disclosure does not limit the number of first activation layers. To reduce the memory space occupied by the first model, one first activation layer can be deployed in the first model. When the first model includes one first activation layer, the object linear features can be first input into the first activation layer to obtain the information willingness of the call object at the time of the call, and then the object linear features and the time linear features can be input into the first activation layer to obtain the time availability of the call object at the time of the call. Alternatively, the object linear features and the time linear features can be first input into the first activation layer to obtain the time availability of the call object at the time of the call, and then the object linear features can be input into the first activation layer to obtain the information willingness of the call object at the time of the call. The embodiment of the present disclosure does not limit the input order of the object linear features and the time linear features. To improve the efficiency of obtaining information willingness and time availability, two first activation layers can be deployed in the first model, so that the object linear features, object linear features, and time linear features can be simultaneously input into the corresponding first activation layers to obtain information willingness and time availability.

[0064] As described above, in the embodiments provided by the present disclosure, a causal relationship can be established between the influence of two influencing factors, namely, whether the user is idle at a certain moment (i.e., momentary idleness) and the user's willingness to accept products or product information (i.e., information willingness), on the connection rate, and the connection rate can be predicted based on the first model according to the established causal relationship. The causal relationship between the influence of momentary idleness and information willingness on the connection rate is shown in Table 1 below:

[0065] Table 1

[0066] Time availability Information willingness Connection rate high high high high Low middle Low high middle Low Low Low

[0067] The levels of time availability, information willingness, and connection rate can be determined by their corresponding specific values. For example, 0%-30% can be determined as a low probability, 31%-60% can be determined as a medium probability, and 61%-100% can be determined as a high probability. The probabilities of time availability, information willingness, and connection rate can be set based on actual application conditions and are not limited in this disclosure.

[0068] According to Table 1 above, when the probability of both the time idleness and the information willingness is low, the probability of the connection rate will also be low. Based on this, in the embodiment provided by the present disclosure, the connection rate is defined as:

[0069] Connection rate = 1-(1-time availability) × (1-information willingness) Formula 1

[0070] Expanding the above formula 1, we can obtain the following formula 2:

[0071] Connection rate = time availability + information willingness - time availability × information willingness Formula 2

[0072] After obtaining the above formula 2, the information willingness and the time availability can be activated in the second activation layer according to the above formula 2 to obtain a more accurate connection rate.

[0073] Based on this, in a specific embodiment provided by the present disclosure, determining the connection rate of the call object at the first number of call times based on the information willingness and the time availability includes:

[0074] Multiplying the willingness to process the information and the availability of the pending time to obtain a causal correlation between the willingness to process the information and the availability of the pending time, wherein the willingness to process the information is the willingness of the callee at the pending call time, the availability of the pending time is the availability of the callee at the pending call time, and the pending call time is any one of the first number of call times;

[0075] According to the first weight parameter, the second weight parameter and the third weight parameter, the willingness of the information to be processed, the idleness of the waiting time and the causal correlation are weightedly calculated to determine the connection rate of the call object at the waiting call time.

[0076] The causal correlation is specifically the product of the willingness to process information and the availability at the time of processing. The first weight parameter is used to represent the influence of the availability at the time of processing on the connection rate; the second weight parameter is used to represent the influence of the willingness to process information on the connection rate; and the third weight parameter is used to represent the influence of the product of the availability at the time of processing and the willingness to process information (i.e., the causal correlation) on the connection rate.

[0077] In practical applications, in order to make the impact of time availability and information willingness on the connection rate learnable, we can set weight parameters based on the above formula 2 to obtain the following formula 3:

[0078]

[0079] in, is the connection rate of call object i, is the idleness of the calling object i, is the information willingness of call recipient i, w1, w2, and w3 are three learnable weight parameters: the first, second, and third weight parameters, respectively. w1 represents the influence of time availability on the connection rate of call recipient i; w2 represents the influence of information willingness on the connection rate of call recipient i; and w3 represents the influence of the product of time availability and information willingness on the connection rate of call recipient i.

[0080] Specifically, for the time availability and information willingness at any calling moment, the time availability and information willingness are multiplied to obtain the causal correlation between the time availability and information willingness. According to the first weight parameter, the second weight parameter and the third weight parameter, according to the above formula 3, the time availability, information willingness and causal correlation are weightedly calculated to obtain the connection rate of the calling object at any calling moment.

[0081] The embodiments provided by the present disclosure construct a causal relationship between time idleness, information willingness and the connection rate according to the influence of time idleness and information willingness on the connection rate, and based on the constructed causal relationship, predict the connection rate of the call object at multiple call times based on the first model, thereby improving the accuracy of the connection rate prediction.

[0082] It should be noted that in actual applications, the causal relationship between the impact of time idleness on the connection rate and the causal relationship between the impact of information willingness on the connection rate can also be constructed separately. However, in order to save computing resources and simplify the implementation process, the present disclosure preferably constructs the causal relationship between time idleness, information willingness and connection rate.

[0083] In actual applications, before predicting the connection rate of a call object based on the first model, the first model needs to be trained so as to use the trained first model to predict the connection rate of the call object. The following explains the training process of the first model.

[0084] In a specific embodiment provided by the present disclosure, the first model is trained based on the following method:

[0085] Acquiring sample object association information and sample call time information, wherein the sample call time information includes multiple sample call times and sample time availability, sample information willingness, and sample connection rate of the multiple sample call times;

[0086] Determining predicted call time information of the plurality of sample call times based on the sample object association information and the sample call time information, wherein the predicted call time information includes predicted time availability, predicted information willingness, and predicted connection rate;

[0087] Calculating a loss value based on the sample call time information and the predicted call time information;

[0088] A first model is trained based on the loss value.

[0089] Among them, sample object association information refers to object association information used to train the first model; sample object association information includes sample object attribute information and sample historical call information; sample call time information refers to call time information used to train the first model; sample call time information includes multiple sample call times and sample time idleness, sample information willingness and sample connection rate of multiple sample call times.

[0090] In practical applications, the sample moment availability, sample information willingness, and sample connection rate can be determined based on the historical call duration, historical information recommendation records, and historical connection status in the sample historical call information. Specifically, calls with a duration greater than M in the historical call duration can be identified as idle, while calls with a duration less than M can be identified as busy. The sample moment availability is determined based on the idle status of the callee; the sample information willingness is determined based on the acceptance records of received and rejected messages in the historical information recommendation records; and the sample connection rate is determined based on the call records of connected and unconnected states in the historical connection status.

[0091] Predicted call time information refers to the call time information predicted by the first model by inputting sample object association information and sample call time information into the first model. Specifically, it includes predicted time availability, predicted information willingness, and predicted connection rate.

[0092] Specifically, sample object association information and sample call time information are obtained, and the sample object association information and sample call time information are input into the first model to obtain the predicted time idleness, predicted information willingness, and predicted connection rate of multiple sample call times. Since the first model at this time is not a trained model, the predicted time idleness, predicted information willingness, and predicted connection rate obtained by prediction will deviate from the actual sample time idleness, sample information willingness, and sample connection rate, and the model parameters of the first model need to be adjusted accordingly. Specifically, the loss value of the first model is calculated based on the sample call time information and the predicted call time information. The loss function used to calculate the loss value can be a cross-entropy loss function, a square loss function, etc. In the embodiment provided in the present disclosure, the cross-entropy loss function is preferably used as the loss function for calculating the loss value. The model parameters of the first model are adjusted according to the calculated loss value. The first model is continued to be trained based on the adjusted model parameters for the next batch of sample object association information and sample call time information until the stop condition of the model training is reached.

[0093] Both time availability and information willingness will affect the connection rate of the call object. Based on this, in the embodiment provided by the present disclosure, the loss value of the first model can be calculated by combining time availability, information willingness and connection rate.

[0094] In a specific embodiment provided by the present disclosure, calculating the loss value according to the sample call time information and the predicted call time information includes:

[0095] Calculating a first loss value according to the idleness at the sample moment and the idleness at the predicted moment;

[0096] Calculating a second loss value according to the sample information willingness and the predicted information willingness;

[0097] Calculating a third loss value according to the sample connection rate and the predicted connection rate;

[0098] The first loss value, the second loss value and the third loss value are weightedly calculated according to the idle weight parameter of the first loss value and the willingness weight parameter of the second loss value to determine a loss value.

[0099] The idle weight parameter is used to represent the degree of influence of the first loss value on the loss value of the first model; the willingness weight parameter is used to represent the degree of influence of the second loss value on the loss value of the first model. Both the idle weight parameter and the willingness weight parameter can be customized according to the actual application.

[0100] Specifically, after obtaining the predicted time availability, predicted information willingness, and predicted connection rate, a first loss value is calculated based on the sample time availability and the predicted time availability. A second loss value is calculated based on the sample information willingness and the predicted information willingness. A third loss value is calculated based on the sample connection rate and the predicted connection rate. The first, second, and third loss values ​​are weighted using the availability weight parameter for the first loss value and the willingness weight parameter for the second loss value to obtain the loss value of the first model.

[0101] Furthermore, the calculation of the first loss value can refer to the following formula 4:

[0102]

[0103] Among them, L1 is the first loss value, is the idleness at the sample time, is the idleness at the predicted moment. According to the above formula 4, the first loss value of the first model can be calculated. The calculation of the second loss value can refer to the following formula 5:

[0104]

[0105] Among them, L2 is the second loss value, is the sample information willingness, To predict information willingness, the second loss value of the first model can be calculated according to the above formula 5. The calculation of the third loss value can be referred to the following formula 6:

[0106]

[0107] Among them, L3 is the third loss value, is the sample connection rate, To predict the connection rate. According to the above formula 6, the third loss value of the first model can be calculated. After calculating the first loss value, the second loss value, and the third loss value, the loss value of the first model can be calculated based on the first loss value, the second loss value, and the third loss value. For details, see the following formula 7:

[0108] L= w4*L1+ w5*L2+L3 Formula 7

[0109] Where L is the loss value of the first model, w4 is the idle weight parameter, and w5 is the willingness weight parameter. According to Formula 7 above, the first loss value, the second loss value, and the third loss value are weighted based on the idle weight parameter and the willingness weight parameter to obtain the loss value of the first model.

[0110] The embodiments provided by the present disclosure respectively determine a first loss value of momentary idleness, a second loss value of information willingness, and a third loss value of connection rate, and determine the loss value of the first model based on the first loss value, the second loss value, and the third loss value, thereby improving the accuracy and reliability of determining the loss value of the first model, thereby improving the accuracy of the first model in predicting the connection rate.

[0111] Step 206: Determine the calling time of the calling party according to the connection rates of the first number of calling times.

[0112] In the embodiment provided by the present disclosure, the first model can be used to predict the connection rate of the call object at multiple call times. After obtaining the connection rate of the call object at multiple call times, the call time of the call object can be determined among the multiple call times based on the connection rate of each call time, so as to facilitate making a call to the call object at the call time of the call object.

[0113] In a specific embodiment provided by the present disclosure, determining the calling time of the calling party according to the connection rates of the first number of calling times includes:

[0114] determining the maximum connection rate among the connection rates of the first number of call moments as the target connection rate;

[0115] The calling time corresponding to the target connection rate is determined as the calling time of the calling object.

[0116] Specifically, among the connection rates of the first number of call moments, the maximum connection rate is determined, and the maximum connection rate is determined as the target connection rate, and the call moment corresponding to the target connection rate is determined as the call moment of the call object.

[0117] By determining the maximum connection rate as the target connection rate and determining the call time corresponding to the target connection rate as the call time of the callee, the probability that the callee will answer the call at the call time is increased.

[0118] Furthermore, the target connection rate represents a high probability that the callee will answer the call at the time of the call. However, in actual applications, probability is contingent. While there is a high probability that the callee will answer the call at the time of the call, there may still be cases where the callee does not answer the call at the time of the call. Based on this, a preset connection rate threshold can be set. The connection rates at multiple call times that exceed the preset connection rate threshold are all determined as the target connection rates. The call times corresponding to these multiple target connection rates are then determined as the call times of the callee. This allows multiple call times to be determined while ensuring a high connection rate, increasing the probability that the callee will answer the call.

[0119] Step 208: Recommend information to the callee based on the call time of the callee.

[0120] After determining the call time for calling the callee based on the connection rates corresponding to multiple call times, a telephone call can be made to the callee based on the determined call time, and information can be recommended to the callee after the callee answers the call.

[0121] The information recommendation method provided by the present disclosure includes: obtaining object association information and call time information of a call object, wherein the object association information includes object attribute information and call information, the call information includes the connection status, call duration and information recommendation record of the call object, and the call time information includes a first number of call times; determining the connection rate of the call object at the first number of call times based on the object association information and the call time information; determining the call time of the call object based on the connection rate of the first number of call times; and recommending information to the call object based on the call time of the call object.

[0122] The embodiment of the present disclosure realizes that after obtaining the object association information and call time information of the call object, the connection rate of the call object at different call times is predicted through the first model, so that the connection rate of the call object at different call times can be obtained, and the call time for calling the call object at different call times can be determined according to the connection rate, so that the call object can be called in the subsequent process based on the call time, thereby calling the call object according to the call time, which can improve the connection rate of the call object and the accuracy of determining the call time, and recommend information to the call object based on the call time.

[0123] It is understood that the above-mentioned various method embodiments mentioned in this disclosure can be combined with each other to form combined embodiments without violating the principle logic. Due to space limitations, this disclosure will not go into details. It is understood by those skilled in the art that in the above-mentioned methods of specific implementation, the specific execution order of each step should be determined by its function and possible internal logic.

[0124] In addition, the present disclosure also provides an information recommendation device, an electronic device, a computer-readable storage medium, and a computer program product, all of which can be used to implement any information recommendation method provided by the present disclosure. The corresponding technical solutions and descriptions are referred to the corresponding records in the method section and will not be repeated here.

[0125] Figure 4 FIG2 shows a block diagram of an information recommendation device provided according to an embodiment of the present disclosure. Figure 4 , an embodiment of the present disclosure provides an information recommendation device, the information recommendation device comprising:

[0126] an acquisition module 402 configured to acquire object association information and call time information of a call object, wherein the object association information includes object attribute information and call information, the call information includes a connection status, call duration, and information recommendation record of the call object, and the call time information includes a first number of call times;

[0127] A first determining module 404 is configured to determine a connection rate of the call object at the first number of call moments based on the object association information and the call moment information;

[0128] A second determining module 406 is configured to determine a calling time of the calling object according to the connection rates of the first number of calling times;

[0129] The recommendation module 408 is configured to recommend information to the callee based on the call time of the callee.

[0130] Optionally, the first determining module 404 is further configured to:

[0131] Encoding the object association information and the call time information respectively to obtain an object feature vector of the object association information and a time feature vector of the call time information;

[0132] Performing linear transformation on the object feature vector and the moment feature vector respectively to obtain object linear features of the object feature vector and moment linear features of the moment feature vector;

[0133] Based on the object linear feature and the time linear feature, a connection rate of the call object at the first number of call times is determined.

[0134] Optionally, the first determining module 404 is further configured to:

[0135] activating the linear feature of the object to obtain the information willingness of the call object at the first number of call moments, wherein the information willingness is used to represent the probability of the call object accepting the recommended information;

[0136] activating the object linear feature and the moment linear feature to obtain the moment availability of the call object at the first number of call moments, wherein the moment availability is used to represent the probability of the call object being idle at the call moment;

[0137] Based on the information willingness and the time availability, a connection rate of the call object at the first number of call times is determined.

[0138] Optionally, the first determining module 404 is further configured to:

[0139] Multiplying the willingness to process the information and the availability of the pending time to obtain a causal correlation between the willingness to process the information and the availability of the pending time, wherein the willingness to process the information is the willingness of the callee at the pending call time, the availability of the pending time is the availability of the callee at the pending call time, and the pending call time is any one of the first number of call times;

[0140] According to the first weight parameter, the second weight parameter and the third weight parameter, the willingness of the information to be processed, the idleness of the waiting time and the causal correlation are weightedly calculated to determine the connection rate of the call object at the waiting call time.

[0141] Optionally, the second determining module 406 is further configured to:

[0142] determining the maximum connection rate among the connection rates of the first number of call moments as the target connection rate;

[0143] The calling time corresponding to the target connection rate is determined as the calling time of the calling object.

[0144] Optionally, the information recommendation device further includes a training module configured to:

[0145] Acquiring sample object association information and sample call time information, wherein the sample call time information includes multiple sample call times and sample time availability, sample information willingness, and sample connection rate of the multiple sample call times;

[0146] Determining predicted call time information of the plurality of sample call times based on the sample object association information and the sample call time information, wherein the predicted call time information includes predicted time availability, predicted information willingness, and predicted connection rate;

[0147] Calculating a loss value based on the sample call time information and the predicted call time information;

[0148] A first model is trained based on the loss value.

[0149] Optionally, the training module is further configured to:

[0150] Calculating a first loss value according to the idleness at the sample moment and the idleness at the predicted moment;

[0151] Calculating a second loss value according to the sample information willingness and the predicted information willingness;

[0152] Calculating a third loss value according to the sample connection rate and the predicted connection rate;

[0153] The first loss value, the second loss value and the third loss value are weightedly calculated according to the idle weight parameter of the first loss value and the willingness weight parameter of the second loss value to determine a loss value.

[0154] The information recommendation device provided by the present disclosure includes: an acquisition module, configured to acquire object association information and call time information of a call object, wherein the object association information includes object attribute information and call information, the call information includes the connection status, call duration and information recommendation record of the call object, and the call time information includes a first number of call times; a first determination module, configured to determine the connection rate of the call object at the first number of call times based on the object association information and the call time information; a second determination module, configured to determine the call time of the call object based on the connection rate of the first number of call times; and a recommendation module, configured to recommend information to the call object based on the call time of the call object.

[0155] The embodiment of the present disclosure realizes that after obtaining the object association information and call time information of the call object, the object association information and call time information are input into the first model, and the connection rate of the call object at multiple call times is predicted by the first model, so that the connection rate of the call object at multiple call times can be obtained, and the call time is determined among the multiple call times according to the connection rate, so that the call object is called in the subsequent process based on the call time, and the call object is called according to the call time, which can improve the connection rate of the call object and the accuracy of determining the call time, and improve the success rate of recommending information to the call object based on the call time.

[0156] Each module in the above-mentioned information recommendation device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0157] Figure 5 A block diagram of an electronic device provided in an embodiment of the present disclosure.

[0158] Reference Figure 5 An embodiment of the present disclosure provides an electronic device 500, which includes: at least one processor 501; at least one memory 502, and one or more I / O interfaces 503 connected between the processor 501 and the memory 502; wherein the memory 502 stores one or more computer programs that can be executed by the at least one processor 501, and the one or more computer programs are executed by the at least one processor 501 so that the at least one processor 501 can perform the above-mentioned information recommendation method.

[0159] Each module in the above-mentioned electronic device can be implemented in whole or in part through software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each of the above modules.

[0160] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor / processing core, implements the above-mentioned information recommendation method. The computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium.

[0161] An embodiment of the present disclosure also provides a computer program product, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above-mentioned information recommendation method.

[0162] It will be understood by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable storage medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium).

[0163] As is well known to those skilled in the art, the term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information (such as computer-readable program instructions, data structures, program modules or other data). Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technology, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those skilled in the art, communication media typically contains computer-readable program instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0164] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0165] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0166] The computer program product described herein may be implemented in hardware, software, or a combination thereof. In one embodiment, the computer program product is implemented as a computer storage medium. In another embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0167] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0168] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0169] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0170] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction contains one or more executable instructions for realizing the prescribed logical function. In some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the prescribed function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0171] Example embodiments have been disclosed herein, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for purposes of limitation. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly indicated, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, it will be understood by those skilled in the art that various changes in form and detail may be made without departing from the scope of the present disclosure as set forth in the appended claims.

Claims

1. An information recommendation method, characterized in that: include: Obtaining object association information and call time information of a call object, wherein the object association information includes object attribute information and call information, the call information includes a connection status, call duration, and information recommendation record of the call object, and the call time information includes a first number of call times; determining, based on the object association information and the call time information, a connection rate of the call object at the first number of call times; determining a calling time of the calling party according to connection rates of the first number of calling times; Information is recommended to the calling party based on the calling time of the calling party.

2. The method according to claim 1, wherein Determining, based on the object association information and the call time information, a connection rate of the call object at the first number of call times, including: Encoding the object association information and the call time information respectively to obtain an object feature vector of the object association information and a time feature vector of the call time information; Performing linear transformation on the object feature vector and the moment feature vector respectively to obtain object linear features of the object feature vector and moment linear features of the moment feature vector; Based on the object linear feature and the time linear feature, a connection rate of the call object at the first number of call times is determined.

3. The method according to claim 2, wherein Determining, based on the object linear feature and the time linear feature, a connection rate of the call object at the first number of call times, includes: activating the linear feature of the object to obtain the information willingness of the call object at the first number of call moments, wherein the information willingness is used to represent the probability of the call object accepting the recommended information; activating the object linear feature and the moment linear feature to obtain the moment availability of the call object at the first number of call moments, wherein the moment availability is used to represent the probability of the call object being idle at the call moment; Based on the information willingness and the time availability, a connection rate of the call object at the first number of call times is determined.

4. The method according to claim 3, wherein Determining, based on the information willingness and the time availability, a connection rate of the call object at the first number of call times, including: Multiplying the willingness to process the information and the availability of the pending time to obtain a causal correlation between the willingness to process the information and the availability of the pending time, wherein the willingness to process the information is the willingness of the callee at the pending call time, the availability of the pending time is the availability of the callee at the pending call time, and the pending call time is any one of the first number of call times; According to the first weight parameter, the second weight parameter and the third weight parameter, the willingness of the information to be processed, the idleness of the waiting time and the causal correlation are weightedly calculated to determine the connection rate of the call object at the waiting call time.

5. The method according to claim 1, wherein Determining the calling time of the calling party according to the connection rates of the first number of calling times includes: determining the maximum connection rate among the connection rates of the first number of call moments as the target connection rate; The calling time corresponding to the target connection rate is determined as the calling time of the calling object.

6. The method according to claim 2, wherein The method further comprises: Acquiring sample object association information and sample call time information, wherein the sample call time information includes multiple sample call times and sample time availability, sample information willingness, and sample connection rate of the multiple sample call times; Determining predicted call time information of the plurality of sample call times based on the sample object association information and the sample call time information, wherein the predicted call time information includes predicted time availability, predicted information willingness, and predicted connection rate; Calculating a loss value based on the sample call time information and the predicted call time information; A first model is trained based on the loss value.

7. The method according to claim 6, wherein Calculating a loss value according to the sample call time information and the predicted call time information includes: Calculating a first loss value according to the idleness at the sample moment and the idleness at the predicted moment; Calculating a second loss value according to the sample information willingness and the predicted information willingness; Calculating a third loss value according to the sample connection rate and the predicted connection rate; The first loss value, the second loss value and the third loss value are weightedly calculated according to the idle weight parameter of the first loss value and the willingness weight parameter of the second loss value to determine a loss value.

8. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores one or more computer programs executable by the at least one processor, and the one or more computer programs are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that The method comprises a computer-readable code or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the method according to any one of claims 1 to 7.

Images