Information generation method, device, electronic device and computer readable medium

Abstract

The embodiments of the present disclosure disclose information generation methods, devices, electronic devices, and computer-readable media. A specific implementation of the method includes: extracting feature information representing the target user's execution of a target value operation from a first item information sequence to obtain a first vector, wherein each item information corresponding to the item in the first item information sequence is an item for which the target user performs a target value operation within a first predetermined time; generating a second vector based on a second item information sequence within a second predetermined time; generating a third vector based on an item information sequence related to a target item category; and generating probability information representing the execution of the target value operation on an item of the target item category within a third predetermined time in the future. The information generation method of the embodiments of the present disclosure can accurately and effectively predict the probability of the target user performing the target value operation on an item of the target item category within a predetermined time in the future.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to an information generation method, device, electronic device, and computer-readable medium. Background Art

[0002] At present, with the popularization of the Internet, online shopping has become an indispensable part of people's daily life. Therefore, most shopping platforms want to predict whether users will buy a certain type of items in order to push information about such items to users. Existing prediction methods often include: since each user behavior corresponds to a timestamp, historical user behaviors are arranged in chronological order to form a time series, and then the probability information of users performing target value operations on objects of the target item category can be predicted through time series prediction algorithms (such as regression models, recurrent neural networks, etc.).

[0003] However, when the above method is used to generate predicted probability information, the following technical problems often occur: the time span information between user behaviors is ignored and it is difficult to capture long-term user information, resulting in inaccurate predicted probability information. Summary of the invention

[0004] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0005] Some embodiments of the present disclosure propose information generation methods, devices, equipment, and computer-readable media to solve the technical problems mentioned in the above background technology section.

[0006] In a first aspect, some embodiments of the present disclosure provide an information generation method, the method comprising: extracting feature information representing a target user performing a target value operation from a first item information sequence to obtain a first vector, wherein each item information corresponding to an item in the first item information sequence is an item on which the target user performs a target value operation within a first predetermined time; generating a second vector based on a second item information sequence within a second predetermined time, wherein each item information corresponding to an item in the second item information sequence is an item on which the target user performs at least one value operation among various value operations within the second predetermined time, and the second vector represents the target user's preference information for various item categories within the second predetermined time; generating a third vector based on an item information sequence related to the target item category, wherein the third vector represents the target user's preference information for performing at least one value operation among various value operations on items of the target item category, and each item information corresponding to an item in the item information sequence related to the target item category is an item on which the target user performs at least one value operation among various value operations; generating probability information representing the execution of the target value operation on items of the target item category within a third predetermined time in the future based on the first vector, the second vector and the third vector.

[0007] Optionally, the above-mentioned extracting feature information representing the target user's execution of the target value operation from the first item information sequence to obtain the first vector includes: performing one-hot encoding on each first item information in the above-mentioned first item information sequence to obtain a fourth vector set; performing word embedding processing on each first item information in the above-mentioned first item information sequence to obtain a fifth vector set; correspondingly concatenating the above-mentioned fourth vector set and the above-mentioned fifth vector set to obtain a sixth vector set; and inputting the above-mentioned sixth vector set into a pre-trained fully connected network to obtain the above-mentioned first vector set.

[0008] Optionally, generating the second vector based on the second item information sequence within the second predetermined time includes: performing word embedding processing on each second item information in the second item information sequence to obtain a seventh vector set; inputting the seventh vector set into a pre-trained first long short-term memory network to obtain the second vector.

[0009] Optionally, the above-mentioned generating a third vector based on the item information sequence related to the target item category includes: determining the time interval between two adjacent item information in the item information sequence related to the target item category to obtain a time interval set; inputting the item information sequence related to the target item category and the above-mentioned time interval set into a pre-trained second long short-term memory network to obtain the above-mentioned third vector, wherein the pre-trained second long short-term memory network learns the time interval information of the item information sequence of each item category for which the target value operation has been performed by the above-mentioned target user.

[0010] Optionally, the above-mentioned probability information representing the execution of the above-mentioned target value operation on the items of the above-mentioned target item category within the third predetermined time in the future is generated based on the above-mentioned first vector, the above-mentioned second vector and the above-mentioned third vector, including: multiplying the above-mentioned first vector with a first coefficient variable to obtain a first value; multiplying the above-mentioned second vector with a second coefficient variable to obtain a second value; multiplying the above-mentioned third vector with a third coefficient variable to obtain a third value; adding the above-mentioned first value, the above-mentioned second value, the above-mentioned third value and a predetermined value to obtain an added value; generating the above-mentioned probability information based on the above-mentioned added value.

[0011] Optionally, generating the above-mentioned probability information according to the above-mentioned added values ​​includes: inputting the above-mentioned added values ​​into a multi-layer activation function to obtain the above-mentioned probability information.

[0012] In a second aspect, some embodiments of the present disclosure provide an information generating device, the device comprising: an extraction unit, configured to extract characteristic information representing a target user performing a target value operation from a first item information sequence to obtain a first vector, wherein an item corresponding to each item information in the first item information sequence is an item on which the target user performs a target value operation within a first predetermined time; a first generating unit, configured to generate a second vector based on a second item information sequence within a second predetermined time, wherein an item corresponding to each item information in the second item information sequence is an item on which the target user performs at least one value operation among various value operations within the second predetermined time, and the second vector represents the target user The preference information for each item category within the above-mentioned second predetermined time; the second generating unit is configured to generate a third vector according to the item information sequence related to the target item category, wherein the above-mentioned third vector represents the preference information of the above-mentioned target user to perform at least one of the various value operations on the items of the above-mentioned target item category, and the item corresponding to each item information in the item information sequence related to the above-mentioned target item category is the item on which the above-mentioned target user performs at least one of the various value operations; the third generating unit is configured to generate probability information representing the execution of the above-mentioned target value operation on the items of the above-mentioned target item category within the third predetermined time in the future according to the above-mentioned first vector, the above-mentioned second vector and the above-mentioned third vector.

[0013] Optionally, the extraction unit is further configured to: perform one-hot encoding on each first item information in the above-mentioned first item information sequence to obtain a fourth vector set; perform word embedding processing on each first item information in the above-mentioned first item information sequence to obtain a fifth vector set; perform corresponding splicing on the above-mentioned fourth vector set and the above-mentioned fifth vector set to obtain a sixth vector set; input the above-mentioned sixth vector set into a pre-trained fully connected network to obtain the above-mentioned first vector set.

[0014] Optionally, the first generating unit is further configured to: perform word embedding processing on each second item information in the above-mentioned second item information sequence to obtain a seventh vector set; input the above-mentioned seventh vector set into a pre-trained first long short-term memory network to obtain the above-mentioned second vector.

[0015] Optionally, the second generation unit is further configured to: determine the time interval between two adjacent item information in the item information sequence related to the above-mentioned target item category to obtain a time interval set; input the item information sequence related to the above-mentioned target item category and the above-mentioned time interval set into a pre-trained second long short-term memory network to obtain the above-mentioned third vector, wherein the above-mentioned pre-trained second long short-term memory network learns the time interval information of the item information sequence of each item category for which the target value operation has been performed by the above-mentioned target user.

[0016] Optionally, the third generating unit is further configured to: multiply the first vector by the first coefficient variable to obtain a first value; multiply the second vector by the second coefficient variable to obtain a second value; multiply the third vector by the third coefficient variable to obtain a third value; add the first value, the second value, the third value and a predetermined value to obtain an added value; generate the probability information according to the added value.

[0017] Optionally, the third generating unit is further configured to: input the above-added numerical value into a multi-layer activation function to obtain the above-mentioned probability information.

[0018] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any method as described in the first aspect.

[0019] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, any method in the first aspect is implemented.

[0020] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the information generation method of some embodiments of the present disclosure, it is possible to accurately and effectively predict the probability of the target user to perform the target value operation on the items of the target item category in the future scheduled time. Specifically, the current prediction method often ignores the time span information between user behaviors and is difficult to capture long-term user information, resulting in inaccurate predicted probability information. Based on this, the information generation method of some embodiments of the present disclosure can first extract feature information representing the target user performing the target value operation from the first item information sequence to obtain a first vector. Among them, each item information corresponding to the item in the above-mentioned first item information sequence is the item on which the above-mentioned target user performs the target value operation within the first scheduled time. Here, the first vector is used for subsequent probability information prediction so that the prediction result can take into account the feature information of the target user performing the target value operation, which can further improve the accuracy of probability information prediction. Then, according to the second item information sequence within the second scheduled time, a second vector is generated. Among them, each item information corresponding to the item in the above-mentioned second item information sequence is the item on which the above-mentioned target user performs at least one value operation in each value operation within the above-mentioned second scheduled time, and the above-mentioned second vector represents the preference information of the above-mentioned target user for each item category within the above-mentioned second scheduled time. Similarly, the second vector is used for subsequent probability information prediction, which can make the prediction result take into account the target user's preference information for each item category within the second predetermined time, and can further improve the accuracy of probability information prediction. Then, a third vector is generated based on the item information sequence related to the target item category. Among them, the third vector represents the preference information of the target user for performing at least one value operation in each value operation on the items of the target item category, and each item information corresponding to the item in the item information sequence related to the target item category is an item for which the target user performs at least one value operation in each value operation. Similarly, the third vector is used for subsequent probability information prediction, which can make the prediction result take into account the target user's preference information for performing the target value operation on the target item category, and can further improve the accuracy of probability information prediction. Finally, based on the first vector, the second vector and the third vector, the probability information representing the execution of the target value operation on the items of the target item category within the third predetermined time in the future is generated. It can be obtained that the above method can accurately and effectively predict the probability of the target user performing the target value operation on the items of the target item category in the future predetermined time. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0022] Figure 1 is a schematic diagram of an application scenario diagram of the information generation method of some embodiments of the present disclosure;

[0023] Figure 2 is a flow chart of some embodiments of the information generation method according to the present disclosure;

[0024] Figure 3 is a schematic diagram of a long short-term memory network structure of an information generation method according to some embodiments of the present disclosure;

[0025] Figure 4 are flow charts of other embodiments of the information generating method according to the present disclosure;

[0026] Figure 5 is a schematic diagram of the structure of some embodiments of the information generating device according to the present disclosure;

[0027] Figure 6 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0028] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0029] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0030] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0031] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0032] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0033] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0034] Figure 1 It is a schematic diagram of an application scenario diagram of the information generation method of some embodiments of the present disclosure.

[0035] like Figure 1As shown, the electronic device 101 can first extract feature information representing the target user performing the target value operation from the first item information sequence 102 to obtain a first vector 104. The item corresponding to each item information in the first item information sequence 102 is the item on which the target user performs the target value operation within the first scheduled time 103. In this application scenario, the first item information sequence 102 can be: "first brand washing machine, second brand dishwasher, third brand shower gel, fourth brand milk, fifth brand laundry detergent, sixth brand toilet paper, seventh brand toilet paper". The first vector 104 can be: "[2,3,1,22,44,45,34,21,22,32,12,32]". The first scheduled time 103 can be: "January 23, 2134-January 23, 2135". Then, according to the second item information sequence 106 within the second scheduled time 105, a second vector 107 is generated. The corresponding item of each item information in the second item information sequence 106 is an item for which the target user performs at least one value operation in each value operation within the second scheduled time 105, and the second vector 107 represents the preference information of the target user for each item category within the second scheduled time 105. In this application scenario, the second item information sequence 106 may be: "fourth brand milk, fifth brand laundry detergent, eighth brand paper cup, ninth brand flower pot, sixth brand toilet paper, tenth brand table, seventh brand toilet paper". The second vector 107 may be: "[12,34,2,62,87,34,23,12,47,90,2,32]". The second scheduled time 105 may be: "December 23, 2134 - January 23, 2135". Further, the third vector 109 is generated according to the item information sequence 108 related to the target item category. Among them, the third vector 109 represents the preference information of the target user to perform at least one value operation among various value operations on the items of the target item category, and each item information corresponding to the item in the item information sequence 108 related to the target item category is an item for which the target user performs at least one value operation among various value operations. In this application scenario, the item information sequence 108 related to the target item category may be: "sixth brand toilet paper, eleventh brand toilet paper, sixth brand toilet paper, seventh brand toilet paper". The third vector 109 may be: "[9,0,1,2,89,49,45,23,56,22]". Finally, according to the first vector 104, the second vector 107 and the third vector 109, the probability information 110 representing the execution of the target value operation on the items of the target item category within the third predetermined time in the future is generated. In this application scenario, the probability information may be "0.73".

[0036] It should be noted that the electronic device 101 can be hardware or software. When the electronic device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or it can be implemented as a single server or a single terminal device. When the electronic device is embodied as software, it can be installed in the hardware devices listed above. It can be implemented as multiple software or software modules for providing distributed services, for example, or it can be implemented as a single software or software module. No specific limitation is made here.

[0037] It should be understood that Figure 1 The number of electronic devices in the embodiment is only for illustration. Any number of electronic devices may be provided according to implementation requirements.

[0038] Continue to refer Figure 2 , shows a process 200 of some embodiments of the information generation method according to the present disclosure. The information generation method comprises the following steps:

[0039] Step 201: extract feature information representing a target user performing a target value operation from a first item information sequence to obtain a first vector.

[0040] In some embodiments, the execution subject of the information generation method (eg Figure 1 The electronic device 101 shown) can extract characteristic information characterizing the target user's execution of the target value operation from the first item information sequence to obtain a first vector. Among them, the item corresponding to each item information in the above-mentioned first item information sequence is the item on which the above-mentioned target user performs the target value operation within the first predetermined time. It should be noted that the above-mentioned target value operation may be a "purchase operation". Furthermore, the above-mentioned first item information sequence may be an information sequence of multiple items that have been purchased. As an example, the above-mentioned execution subject may input the above-mentioned first item information sequence into a pre-trained recurrent neural network (RNN) to obtain a first vector. As another example, the above-mentioned execution subject may input the above-mentioned first item information sequence into a long short-term memory network (LSTM, Long Short-Term Memory) to obtain a first vector.

[0041] In some optional implementations of some embodiments, extracting feature information representing the target user performing the target value operation from the first item information sequence to obtain the first vector may include the following steps:

[0042] In the first step, each first item information in the first item information sequence is uniquely encoded to obtain a fourth vector set.

[0043] In the second step, word embedding processing is performed on each first item information in the first item information sequence to obtain a fifth vector set.

[0044] The third step is to correspondingly connect the fourth vector set and the fifth vector set to obtain a sixth vector set.

[0045] The sixth vector set is input into a pre-trained fully connected network to obtain the first vector set.

[0046] Step 202: Generate a second vector according to a second item information sequence within a second predetermined time.

[0047] In some embodiments, the execution subject may generate a second vector based on a second item information sequence within a second predetermined time. The item corresponding to each item information in the second item information sequence is an item on which the target user performs at least one value operation among various value operations within the second predetermined time. The second vector represents the preference information of the target user for various item categories within the second predetermined time. For example, the various item categories may include: household appliances, toiletries, office supplies, etc. The various value operations may include but are not limited to at least one of the following: purchase operation, collection operation, add to shopping cart operation, and browsing operation. It should be noted that the length of the second predetermined time may be less than the length of the first predetermined time. For example, the second predetermined time may be a short period of time. The first predetermined time may be a long period of time.

[0048] As an example, the execution entity may input the second item information sequence into a pre-trained recurrent neural network to obtain the second vector.

[0049] In some optional implementations of some embodiments, the generating of the second vector according to the second item information sequence within the second predetermined time may include the following steps:

[0050] In the first step, word embedding processing is performed on each second item information in the second item information sequence to obtain a seventh vector set.

[0051] The second step is to input the seventh vector set into the pre-trained first long short-term memory network to obtain the second vector.

[0052] As an example, Figure 3 As shown, the above long short-term memory network can adopt a network structure with multiple inputs and a single output, and the sampling frame feature X in the above sampling frame feature set can be obtained. t-1 and X t Among them, the input of the first cell 301 of the long short-term memory network is X t-1 , the output is the long short-term memory network output value H at time t-1 t-1 and the cell state C at time t-1 t-1The input of the second cell 302 of the LSTM network is X t , H t-1 and C t-1 The output is the long short-term memory network output value H at time t t and the cell state C at time t t It should be noted that the cells in the long short-term memory network include three gates: forget gate, input gate and output gate.

[0053] The forget gate can decide which information should be discarded or retained. The information from the previous hidden state and the current input information are simultaneously input into the S-shaped growth curve (Sigmoid) (i.e., σ in the figure), and the output value is between 0 and 1. The closer to 0, the more information should be forgotten, and the closer to 1, the more information should be retained. The output f of the forget gate t Determines which data to forget. Among them, the output f of the forget gate t It can be obtained by the following formula:

[0054] f t =σ(W f ·[h t-1 , x t ]+b f )

[0055] Here, W f represents the weight matrix of the forget gate, b f represents the bias term of the forget gate, [h t-1 , x t ] indicates that two vectors are connected to form a longer vector (i.e., vector concatenation).

[0056] The input gate is used to update the unit state. The information of the previous hidden state and the current input is first input into the S-shaped growth curve function. The output value is adjusted between 0 and 1 to determine which information to update. 0 means unimportant and 1 means important. The hidden state and the current input can also be transmitted to the hyperbolic tangent (Tanh) function, and the value is compressed between -1 and 1 to adjust the network. Then the output of the hyperbolic tangent function is multiplied by the output of the S-shaped growth curve function. The output of the S-shaped growth curve function will determine which information in the output of the hyperbolic tangent function is important and needs to be retained. The output of the input gate i t It can be obtained by the following formula:

[0057] i t =σ(W i ·[h t-1 , x t ]+b i )

[0058] Here, I tCan determine the input X of the network at the current moment t How many cell states C are saved to time t? t .W i represents the weight matrix of the input gate, b i Represents the bias term of the input gate.

[0059] Here, the cell state used to describe the input at time t can be determined by the following formula:

[0060]

[0061] Here, represents the cell state at time t, W C represents the weight matrix of the input gate, b C Represents the bias term of the input gate.

[0062] Finally, the cell state C at time t t It is determined by the following formula:

[0063]

[0064] Among them, the output gate can determine the value of the next hidden state, and the hidden state contains relevant information about the previous input. Of course, the hidden state can also be used for prediction. First, the previous hidden state and the current input are passed to the S-type growth curve function; then the newly obtained unit state is passed to the hyperbolic tangent function; then the output of the hyperbolic tangent function is multiplied by the output of the S-type growth curve function to determine the information that the hidden state should carry; finally, the hidden state is used as the current unit output, and the new unit state and the new hidden state are transmitted to the next time step. The output of the output gate H t The formula is as follows:

[0065] o t =σ(W o [h t-1 , x t ]+b o )

[0066] h t =o t *tanh(C t )

[0067] Here, t Indicates which ones will be considered as output at time t, W o represents the weight matrix of the output gate, b o represents the bias term of the output gate, h t It can be expressed as the output of the output gate, that is, the sampling features output between time t-1 and time t.

[0068] Step 203: Generate a third vector according to the item information sequence related to the target item category.

[0069] In some embodiments, the execution subject may generate a third vector based on the sequence of item information related to the target item category. The third vector represents the preference information of the target user for performing at least one value operation among various value operations on the items in the target item category. Each item information corresponding to the item in the sequence of item information related to the target item category is an item for which the target user performs at least one value operation among various value operations. As an example, the execution subject may input the sequence of item information related to the target item category into a pre-trained long short-term memory network to obtain the third vector.

[0070] Step 204: Generate probability information representing the execution of the target value operation on the items of the target item category within a third predetermined time in the future based on the first vector, the second vector and the third vector.

[0071] In some embodiments, the execution subject may generate probability information representing the execution of the target value operation on the items of the target item category within a third predetermined time in the future based on the first vector, the second vector, and the third vector. As an example, the execution subject may input the first vector, the second vector, and the third vector into a pre-trained fully connected network to obtain the probability information.

[0072] In some optional implementations of some embodiments, generating, according to the first vector, the second vector and the third vector, probability information representing the execution of the target value operation on the items of the target item category within a third predetermined time in the future may include the following steps:

[0073] The first step is to multiply the first vector by the first coefficient variable to obtain a first value.

[0074] The second step is to multiply the second vector by the second coefficient variable to obtain a second value.

[0075] The third step is to multiply the third vector by the third coefficient variable to obtain a third value.

[0076] The fourth step is to add the first value, the second value, the third value and the predetermined value to obtain the added value.

[0077] The fifth step is to generate the above probability information according to the above added values.

[0078] It should be noted that the first coefficient variable, the second coefficient variable and the third coefficient variable can be obtained by model training.

[0079] Optionally, the execution entity may input the added value into a multi-layer activation function to obtain the probability information.

[0080] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the information generation method of some embodiments of the present disclosure, it is possible to accurately and effectively predict the probability of the target user to perform the target value operation on the items of the target item category in the future scheduled time. Specifically, the current prediction method often ignores the time span information between user behaviors and is difficult to capture long-term user information, resulting in inaccurate predicted probability information. Based on this, the information generation method of some embodiments of the present disclosure can first extract feature information representing the target user performing the target value operation from the first item information sequence to obtain a first vector. Among them, each item information corresponding to the item in the above-mentioned first item information sequence is the item on which the above-mentioned target user performs the target value operation within the first scheduled time. Here, the first vector is used for subsequent probability information prediction so that the prediction result can take into account the feature information of the target user performing the target value operation, which can further improve the accuracy of probability information prediction. Then, according to the second item information sequence within the second scheduled time, a second vector is generated. Among them, each item information corresponding to the item in the above-mentioned second item information sequence is the item on which the above-mentioned target user performs at least one value operation in each value operation within the above-mentioned second scheduled time, and the above-mentioned second vector represents the preference information of the above-mentioned target user for each item category within the above-mentioned second scheduled time. Similarly, the second vector is used for subsequent probability information prediction, which can make the prediction result take into account the target user's preference information for each item category within the second predetermined time, and can further improve the accuracy of probability information prediction. Then, a third vector is generated based on the item information sequence related to the target item category. Among them, the third vector represents the preference information of the target user for performing at least one value operation in each value operation on the items of the target item category, and each item information corresponding to the item in the item information sequence related to the target item category is an item for which the target user performs at least one value operation in each value operation. Similarly, the third vector is used for subsequent probability information prediction, which can make the prediction result take into account the target user's preference information for performing the target value operation on the target item category, and can further improve the accuracy of probability information prediction. Finally, based on the first vector, the second vector and the third vector, the probability information representing the execution of the target value operation on the items of the target item category within the third predetermined time in the future is generated. It can be obtained that the above method can accurately and effectively predict the probability of the target user performing the target value operation on the items of the target item category in the future predetermined time.

[0081] Continue to refer Figure 4, shows a process 400 of another embodiment of the information generation method according to the present disclosure. The information generation method comprises the following steps:

[0082] Step 401: extract feature information representing a target user performing a target value operation from a first item information sequence to obtain a first vector.

[0083] Step 402: Generate a second vector according to a second item information sequence within a second predetermined time.

[0084] Step 403: determine the time interval between two adjacent item information in the item information sequence related to the target item category to obtain a time interval set.

[0085] In some embodiments, an execution entity (e.g. Figure 1 The electronic device 101 shown in the figure can determine the time interval between two adjacent item information in the above-mentioned item information sequence related to the target item category to obtain a time interval set. As an example, the item information related to the target item category and the corresponding time information stored in the target database can be used to obtain the item information related to the target item category.

[0086] Step 404: input the above-mentioned target item category-related item information sequence and the above-mentioned time interval set into a pre-trained second long short-term memory network to obtain the above-mentioned third vector.

[0087] In some embodiments, the execution subject may input the target item category-related item information sequence and the time interval set into a pre-trained second long short-term memory network to obtain the third vector. The second long short-term memory network may be a time-aware long short-term memory (T-LSTM).

[0088] Step 405 , generating, based on the first vector, the second vector and the third vector, probability information representing the execution of the target value operation on the items of the target item category within a third predetermined time in the future.

[0089] In some embodiments, the specific implementation of steps 401, 402 and 405 and the technical effects brought about can be referred to Figure 2 The corresponding steps 201, 202 and 205 in the embodiments are not described in detail here.

[0090] from Figure 4 It can be seen that Figure 2 Compared with the description of some corresponding embodiments, Figure 4The process 400 of the information generation method in some corresponding embodiments reflects the step of considering the time interval information between two adjacent item information in the item information sequence related to the target item category. Therefore, the schemes described in these embodiments can more accurately and effectively predict the probability of the target user performing the target value operation on the items of the target item category at a predetermined time in the future.

[0091] Continue to refer Figure 5 As an implementation of the above methods in the above figures, the present disclosure provides some embodiments of an information generating device, and these device embodiments are Figure 2 Corresponding to the above method embodiments, the device can be specifically applied to various electronic devices.

[0092] like Figure 5 As shown, the information generating device 500 of some embodiments includes: an extraction unit 501, a first generating unit 502, a second generating unit 503 and a third generating unit 504. The extraction unit 501 is configured to extract feature information representing the target user performing the target value operation from the first item information sequence to obtain a first vector, wherein the item corresponding to each item information in the first item information sequence is the item on which the target user performs the target value operation within the first predetermined time. The first generating unit 502 is configured to generate a second vector based on the second item information sequence within the second predetermined time, wherein the item corresponding to each item information in the second item information sequence is the item on which the target user performs at least one value operation among the various value operations within the second predetermined time, and the second vector represents the preference information of the target user for each item category within the second predetermined time. The second generating unit 503 is configured to generate a third vector based on the item information sequence related to the target item category, wherein the third vector represents the preference information of the target user to perform at least one value operation among various value operations on the items of the target item category, and each item information corresponding to the item in the item information sequence related to the target item category is an item on which the target user performs at least one value operation among various value operations. The third generating unit 504 is configured to generate probability information representing the execution of the target value operation on the items of the target item category within a third predetermined time in the future based on the first vector, the second vector and the third vector.

[0093] In some optional implementations of some embodiments, the extraction unit 501 of the information generating device 500 may be further configured to: perform one-hot encoding on each first item information in the above-mentioned first item information sequence to obtain a fourth vector set; perform word embedding processing on each first item information in the above-mentioned first item information sequence to obtain a fifth vector set; perform corresponding splicing on the above-mentioned fourth vector set and the above-mentioned fifth vector set to obtain a sixth vector set; and input the above-mentioned sixth vector set into a pre-trained fully connected network to obtain the above-mentioned first vector set.

[0094] In some optional implementations of some embodiments, the first generation unit 502 of the information generating device 500 may be further configured to: perform word embedding processing on each second item information in the above-mentioned second item information sequence to obtain a seventh vector set; input the above-mentioned seventh vector set into a pre-trained first long short-term memory network to obtain the above-mentioned second vector.

[0095] In some optional implementations of some embodiments, the second generation unit 503 of the information generating device 500 may be further configured to: determine the time interval between two adjacent item information in the item information sequence related to the above-mentioned target item category to obtain a time interval set; input the item information sequence related to the above-mentioned target item category and the above-mentioned time interval set into a pre-trained second long short-term memory network to obtain the above-mentioned third vector, wherein the above-mentioned pre-trained second long short-term memory network learns the time interval information of the item information sequence of each item category for which the target value operation has been performed by the above-mentioned target user.

[0096] In some optional implementations of some embodiments, the third generation unit 504 of the information generating device 500 can be further configured to: multiply the above-mentioned first vector by the first coefficient variable to obtain a first numerical value; multiply the above-mentioned second vector by the second coefficient variable to obtain a second numerical value; multiply the above-mentioned third vector by the third coefficient variable to obtain a third numerical value; add the above-mentioned first numerical value, the above-mentioned second numerical value, the above-mentioned third numerical value and a predetermined numerical value to obtain an added numerical value; generate the above-mentioned probability information according to the above-mentioned added numerical value.

[0097] In some optional implementations of some embodiments, the third generating unit 504 of the information generating device 500 may be further configured to: input the above-added numerical value into a multi-layer activation function to obtain the above-mentioned probability information.

[0098] It is understood that the units described in the device 500 are similar to those described in the reference Figure 2 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the device 500 and the units included therein, and will not be described in detail here.

[0099] Reference below Figure 6 , which shows a structural schematic diagram of an electronic device 600 suitable for implementing some embodiments of the present disclosure. Figure 6 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0100] like Figure 6 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0101] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 6 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0102] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0103] It should be noted that the computer-readable medium in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0104] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0105] The computer-readable medium may be included in the device, or may exist independently without being incorporated into the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: extracts characteristic information representing the target user performing a target value operation from a first item information sequence to obtain a first vector, wherein each item information corresponding to the item in the first item information sequence is an item on which the target user performs a target value operation within a first predetermined time; generates a second vector according to a second item information sequence within a second predetermined time, wherein each item information corresponding to the item in the second item information sequence is an item on which the target user performs at least one value operation among various value operations within the second predetermined time, and the second vector represents the target user's preference information for various item categories within the second predetermined time; generates a third vector according to an item information sequence related to the target item category, wherein the third vector represents the target user's preference information for performing at least one value operation among various value operations on items of the target item category, and each item information corresponding to the item in the item information sequence related to the target item category is an item on which the target user performs at least one value operation among various value operations; generates probability information representing the execution of the target value operation on items of the target item category within a third predetermined time in the future according to the first vector, the second vector, and the third vector.

[0106] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving 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., via the Internet using an Internet service provider).

[0107] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession 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 square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0108] The units described in some embodiments of the present disclosure may be implemented by software or hardware. The units described may also be set in a processor, for example, it may be described as: a processor includes an extraction unit, a first generation unit, a second generation unit, and a third generation unit. The names of these units do not constitute a limitation on the units themselves in some cases, for example, the first generation unit may also be described as "a unit for generating a second vector according to a second item information sequence within a second predetermined time".

[0109] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0110] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A method for generating information, comprising: Extracting feature information representing the target user performing a target value operation from the first item information sequence to obtain a first vector, wherein each item information corresponding to the item in the first item information sequence is an item on which the target user performs a target value operation within a first predetermined time, the target value operation is a purchase operation, and the first item information sequence is an information sequence of multiple items that have been purchased; Generate a second vector according to a second item information sequence within a second predetermined time, wherein each item corresponding to the item information in the second item information sequence is an item for which the target user performs at least one value operation among various value operations within the second predetermined time, and the second vector represents the target user's preference information for various item categories within the second predetermined time, the length of the second predetermined time is shorter than the length of the first predetermined time, and the various value operations include: purchase operation, collection operation, add to shopping cart operation, and browsing operation; Generate a third vector according to the item information sequence related to the target item category, wherein the third vector represents the preference information of the target user for performing at least one value operation among various value operations on items of the target item category, and each item information corresponding to the item in the item information sequence related to the target item category is an item for which the target user performs at least one value operation among various value operations; Probability information representing the execution of the target value operation on the items of the target item category within a third predetermined time in the future is generated according to the first vector, the second vector and the third vector.

2. The method according to claim 1, wherein: The step of extracting feature information representing the target user performing the target value operation from the first item information sequence to obtain a first vector includes: Performing one-hot encoding on each first item information in the first item information sequence to obtain a fourth vector set; Performing word embedding processing on each first item information in the first item information sequence to obtain a fifth vector set; Correspondingly concatenating the fourth vector set and the fifth vector set to obtain a sixth vector set; The sixth vector set is input into a pre-trained fully connected network to obtain the first vector set.

3. The method according to claim 1, wherein: The generating a second vector according to the second item information sequence within the second predetermined time includes: Performing word embedding processing on each second item information in the second item information sequence to obtain a seventh vector set; The seventh vector set is input into a pre-trained first long short-term memory network to obtain the second vector.

4. The method according to claim 1, wherein: The step of generating a third vector according to the item information sequence related to the target item category includes: Determine the time interval between two adjacent item information in the item information sequence related to the target item category to obtain a time interval set; The item information sequence related to the target item category and the time interval set are input into a pre-trained second long short-term memory network to obtain the third vector, wherein the pre-trained second long short-term memory network learns the time interval information of the item information sequence of each item category for which the target value operation has been performed by the target user.

5. The method according to claim 1, wherein: The generating, according to the first vector, the second vector and the third vector, probability information representing the execution of the target value operation on the items of the target item category within a third predetermined time in the future includes: multiplying the first vector by a first coefficient variable to obtain a first value; multiplying the second vector by a second coefficient variable to obtain a second value; multiplying the third vector by a third coefficient variable to obtain a third value; Adding the first value, the second value, the third value, and a predetermined value to obtain an added value; The probability information is generated according to the added value.

6. The method according to claim 5, wherein: The generating the probability information according to the added value includes: The added values ​​are input into a multi-layer activation function to obtain the probability information.

7. An information generating device, comprising: an extraction unit configured to extract feature information representing that a target user performs a target value operation from a first item information sequence to obtain a first vector, wherein an item corresponding to each item information in the first item information sequence is an item on which the target user performs a target value operation within a first predetermined time, the target value operation is a purchase operation, and the first item information sequence is an information sequence of multiple items that have been purchased; a first generating unit, configured to generate a second vector according to a second item information sequence within a second predetermined time, wherein each item corresponding to the item information in the second item information sequence is an item for which the target user performs at least one value operation among various value operations within the second predetermined time, and the second vector represents the preference information of the target user for various item categories within the second predetermined time, the time length of the second predetermined time is shorter than the time length of the first predetermined time, and the various value operations include: a purchase operation, a favorite operation, an add-to-cart operation, and a browse operation; A second generating unit is configured to generate a third vector according to an item information sequence related to a target item category, wherein the third vector represents preference information of the target user for performing at least one value operation among various value operations on items of the target item category, and each item information corresponding item in the item information sequence related to the target item category is an item for which the target user performs at least one value operation among various value operations; The third generating unit is configured to generate probability information representing the execution of the target value operation on the items of the target item category within a third predetermined time in the future according to the first vector, the second vector and the third vector.

8. The information generating device according to claim 7, wherein: The second generating unit is further configured to: Determine the time interval between two adjacent item information in the item information sequence related to the target item category to obtain a time interval set; The item information sequence related to the target item category and the time interval set are input into a pre-trained second long short-term memory network to obtain the third vector, wherein the pre-trained second long short-term memory network learns the time interval information of the item information sequence of each item category for which the target value operation has been performed by the target user.

9. An electronic device, comprising: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method according to any one of claims 1 to 6.

10. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images