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

Abstract

Embodiments of the present disclosure disclose information generation methods, devices, electronic equipment and computer readable media. A specific embodiment of the information generation method comprises: obtaining a pre-constructed graph model, wherein the graph model represents the association relationship between a first user set, a second user set and an item set; generating first data according to each node and each edge associated with the node representing a first target user in the first user set in the graph model; obtaining second data of the first target user performing a second value operation on the item set; and generating first number information of the first target user performing a second value operation on the target item in a single session at a future predetermined time according to the first data and the second data. The embodiment efficiently and accurately generates the first number information of the first target user performing the second value operation on the target item in a single session at a future predetermined time by introducing the pre-constructed graph model.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of computer technology, and in particular, to an information generation method and apparatus, an electronic device, and a computer readable medium. BACKGROUND

[0002] At present, network delivery can enable consumers to more intuitively perceive the use value of an item. For a service provider, network delivery can bring more sales channels. When selecting which user to deliver, the quantity of single transactions of the user in which the item is subjected to a second value operation (for example, sales) is often referred to. For the prediction of the quantity of single transactions of the item, a common method is to determine the quantity of single transactions of the item subjected to the second value operation by using the Delphi method (determined by aggregating the experience of several experts). However, the Delphi method often requires a large amount of professional labor costs. In addition, the generated quantity of single transactions of the item is often not accurate enough. More importantly, the data information generated by using the Delphi method to generate the quantity of single transactions of the item is often not complete enough. SUMMARY

[0003] The summary of the present disclosure is used to briefly introduce the concepts, which will be described in detail in the specific embodiments part. The summary of the present disclosure is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to be used to limit the scope of the claimed technical solutions.

[0004] Some embodiments of the present disclosure provide an information generation method, apparatus, device and computer readable medium to solve the technical problems mentioned in the background section.

[0005] In a first aspect, some embodiments of the present disclosure provide an information generation method, comprising: obtaining a pre-constructed graph model, wherein the graph model represents an association relationship between a first user set, a second user set and an item set; generating first data according to each node and each edge associated with a node representing a first target user in the first user set in the graph model, wherein the first data comprises feature information representing an influence of the first target user on the second user set to perform a first value operation on a target item in the item set; obtaining second data of the first target user performing a second value operation on the item set; and generating first quantity information of the first target user performing a second value operation on the target item in a single transaction at a future predetermined time according to the first data and the second data.

[0006] Optionally, the generating, according to the first data and the second data, the first number information of the first target user performing the second value operation on the target item in a single session at a future scheduled time, comprises: obtaining attribute information of the target item, second number information of the target item performing the second value operation on a target brand in a single session, third number information of the target item performing the second value operation in a single session, and fourth number information of an item belonging to a category of the target item performing the second value operation in a single session; performing data preprocessing on the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information to obtain processed first data, processed second data, processed attribute information, processed second number information, processed third number information, and processed fourth number information; and generating the first number information of the first target user performing the second value operation on the target item in a single session at a future scheduled time based on the processed first data, the processed second data, the processed attribute information, the processed second number information, the processed third number information, and the processed fourth number information.

[0007] Optionally, the generating the first number information of the first target user performing the second value operation on the target item in a single session at a future scheduled time comprises: inputting the processed data set into a pre-trained regression network to obtain the first number information of the first target user performing the second value operation on the target item in a single session at a future scheduled time.

[0008] Optionally, the performing data preprocessing on the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information to obtain processed first data, processed second data, processed attribute information, processed second number information, processed third number information, and processed fourth number information comprises: performing data completion on data with missing data in the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information to obtain completed first data, completed second data, completed attribute information, completed second number information, completed third number information, and completed fourth number information; and performing data type conversion on the completed first data, the completed second data, the completed attribute information of the target item, the completed second number information, the completed third number information, and the completed fourth number information to obtain converted first data, converted second data, converted attribute information of the target item, converted second number information, converted third number information, and converted fourth number information.

[0009] Optionally, the data type conversion is performed on the completed first data, the completed second data, the completed attribute information of the target item, the completed second number information, the completed third number information and the completed fourth number information to obtain converted first data, converted second data, converted attribute information of the target item, converted second number information, converted third number information and converted fourth number information, including: converting the completed first data, the completed second data, the completed second number information, the completed third number information and the completed fourth number information into floating-point data; in response to the presence of continuous data in the completed attribute information, converting the continuous data into floating-point data; in response to the presence of ordered categorical variable data in the completed attribute information, performing label encoding on the ordered categorical variable data to obtain label-encoded data; and in response to the presence of unordered categorical variable data in the completed attribute information, performing one-hot encoding on the unordered categorical variable data to obtain one-hot encoded data.

[0010] Optionally, the first data is generated according to each node and each edge associated with the node representing the first target user in the graph model, including: determining a first subgraph associated with the first target user in the graph model according to each node and each edge associated with the node representing the first target user in the graph model; determining first information of each edge associated with the node representing the first target user in the first subgraph; generating first score information representing influence of the first target user according to the first subgraph; determining a second subgraph associated with the target item in the graph model according to each node and each edge associated with the node representing the target item in the graph model; determining second information of each edge associated with the node representing the target item in the second subgraph; generating second score information representing influence of the target item according to the second subgraph; and generating the first data according to the first information, the first score information, the second information and the second score information.

[0011] Optionally, the second score information representing influence of the target item is generated according to the second subgraph, including: generating the second score information representing influence of the target item according to the second subgraph by using a page rank algorithm.

[0012] In a second aspect, some embodiments of the present disclosure provide an information generation apparatus, the apparatus comprising: a first acquisition unit configured to acquire a pre-constructed graph model, wherein the graph model represents an association relationship among a first user set, a second user set and an item set; a first generation unit configured to generate first data according to each node and each edge associated with a node representing a first target user in the first user set in the graph model, wherein the first data comprises feature information representing that the first target user influences the second user set to perform a first value operation on a target item in the item set; a second acquisition unit configured to acquire second data representing that the first target user performs a second value operation on the item set; and a second generation unit configured to generate first number information of the first target user performing the second value operation on the target item in a single session at a future predetermined time according to the first data and the second data.

[0013] Optionally, the second generation unit is further configured to: acquire attribute information of the target item, second number information of the target item of a target brand performing the second value operation in a single session, third number information of the target item performing the second value operation in a single session, and fourth number information of an item belonging to a category of the target item performing the second value operation in a single session; perform data preprocessing on the first data, the second data, the attribute information of the target item, the second number information, the third number information and the fourth number information to obtain processed first data, processed second data, processed attribute information, processed second number information, processed third number information and processed fourth number information; and generate the first number information of the first target user performing the second value operation on the target item in a single session at the future predetermined time based on the processed first data, the processed second data, the processed attribute information, the processed second number information, the processed third number information and the processed fourth number information.

[0014] Optionally, the second generation unit is further configured to: input the processed data set into a pre-trained regression network to obtain the first number information of the first target user performing the second value operation on the target item in a single session at the future predetermined time.

[0015] Optionally, the second generating unit is further configured to: perform data completion on the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information that have data missing, to obtain completed first data, completed second data, completed attribute information, completed second number information, completed third number information, and completed fourth number information; and perform data type conversion on the completed first data, the completed second data, the completed attribute information of the target item, the completed second number information, the completed third number information, and the completed fourth number information, to obtain converted first data, converted second data, converted attribute information of the target item, converted second number information, converted third number information, and converted fourth number information.

[0016] Optionally, the second generating unit is further configured to: convert the completed first data, the completed second data, the completed second number information, the completed third number information, and the completed fourth number information into floating-point data; convert continuous data in the completed attribute information into floating-point data in response to the continuous data; perform label encoding on ordered categorical variable data in the completed attribute information, to obtain label-encoded data; and perform one-hot encoding on unordered categorical variable data in the completed attribute information, to obtain one-hot-encoded data.

[0017] Optionally, the first generating unit is further configured to: determine a first subgraph associated with the first target user in the graph model according to each node and each edge associated with a node representing the first target user in the graph model; determine first information of each edge associated with the node representing the first target user in the first subgraph; generate first score information representing influence of the first target user according to the first subgraph; determine a second subgraph associated with the target item in the graph model according to each node and each edge associated with a node representing the target item in the graph model; determine second information of each edge associated with the node representing the target item in the second subgraph; generate second score information representing influence of the target item according to the second subgraph; and generate the first data according to the first information, the first score information, the second information, and the second score information.

[0018] Optionally, the first generating unit is further configured to: generate the second score information representing influence of the target item according to the second subgraph by using a page rank algorithm.

[0019] 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, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of the first aspect.

[0020] In a fourth aspect, some embodiments of the present disclosure provide a computer readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method of any one of the first aspect.

[0021] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: the information generation method of some embodiments of the present disclosure can efficiently and accurately generate the first number information of the first target user performing the second value operation on the target item at a future predetermined time by introducing a pre-constructed graph model. Specifically, using the Delphi method often requires a large amount of professional labor cost. In addition, the generated item quantity information often has the problem of not being accurate enough. More importantly, the data information generated by using the Delphi method to generate item quantity reference is often not complete enough. Based on this, the information generation method of some embodiments of the present disclosure can first obtain a pre-constructed graph model. The graph model represents the association relationship between the first user set, the second user set and the item set. Here, the pre-constructed graph model can be used for subsequent first data generation, so that the subsequently generated number information takes into account the feature information of the first target user influencing the second user set performing the first value operation on the target item in the item set. In this way, the generated first number information is more accurate. Then, according to each node and each edge associated with the node representing the first target user in the first user set in the graph model, the first data can be generated simply and quickly. The first data includes feature information representing the influence of the first target user on the second user set performing the first value operation on the target item in the item set. Further, the second data of the first target user performing the second value operation on the item set is obtained. The second data provides data support for the generation of the first number information. Finally, according to the first data and the second data, the first number information of the first target user performing the second value operation on the target item at a future predetermined time can be accurately and conveniently generated. BRIEF DESCRIPTION OF DRAWINGS

[0022] The above and other features, aspects, and advantages of embodiments of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Throughout the drawings, like or similar reference numerals are used to refer to like or similar elements throughout the several views. It should be understood that the drawings are schematic and elements depicted are not necessarily to scale.

[0023] Figure 1is a schematic diagram of one application scenario of the information generation method of some embodiments of the disclosure;

[0024] Figure 2 is a flowchart of some embodiments of the information generation method according to the disclosure;

[0025] Figure 3 is a schematic diagram of generating a first subgraph in the information generation method according to some embodiments of the disclosure;

[0026] Figure 4 is a schematic diagram of generating a second subgraph in the information generation method according to some embodiments of the disclosure;

[0027] Figure 5 is a flowchart of another embodiment of the information generation method according to the disclosure;

[0028] Figure 6 is a structural schematic diagram of some embodiments of the information generation apparatus according to the disclosure;

[0029] Figure 7 is a structural schematic diagram of an electronic device suitable for implementing some embodiments of the disclosure. DETAILED DESCRIPTION

[0030] Embodiments of the disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the disclosure are shown in the drawings, it should be understood that the disclosure can be implemented in various forms, and should not be interpreted as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to make the disclosure more thorough and complete. It should be understood that the drawings and embodiments of the disclosure are only for exemplary purposes, and are not intended to limit the scope of protection of the disclosure.

[0031] In addition, it should be noted that only parts related to the application are shown in the drawings for ease of description. The embodiments in the disclosure and the features in the embodiments can be combined with each other without conflict.

[0032] It should be noted that the concepts of "first", "second", etc. mentioned in the disclosure are only used to distinguish different devices, modules or units, and are not intended to limit the order or interdependence of the functions performed by these devices, modules or units.

[0033] It should be noted that the modification of "one" or "multiple" in the disclosure is illustrative rather than limiting, and those skilled in the art should understand that unless otherwise explicitly stated in the context, it should be understood as "one or more".

[0034] Names of messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0035] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0036] Figure 1 is a schematic diagram of one application scenario of the information generation method of some embodiments of the present disclosure.

[0037] As Figure 1 shown, the electronic device 101 can first obtain a pre-constructed graph model 102. The graph model 102 represents the association relationship between a first user set, a second user set and an item set. In this application scenario, the first user set includes a first user 1021 and a first target user 1022. The second user set includes a second user 1023 and a second user 1024. The item set includes an item 1025 and a target item 1026. Then, according to the nodes and edges associated with the first target user 1022 in the first user set in the graph model 102, first data 103 is generated. The first data 103 includes feature information representing the influence of the first target user 1022 on the second user set performing a first value operation on the target item 1026 in the item set. In this application scenario, the nodes associated with the first target user 1022 can include an edge from the first target user 1022 to the item 1025, an edge from the second target user 1023 to the first target user 1022, an edge from the second target user 1024 to the first target user 1022, and an edge from the first target user 1022 to the target item 1026. Further, second data 104 is obtained, which represents the second value operation performed by the first target user 1022 on the item set. Finally, according to the first data 103 and the second data 104, first number information 105 is generated, which represents the number of times the first target user 1022 performs a second value operation on the target item 1026 at a future predetermined time. In this application scenario, the first number information 105 can be "2387".

[0038] 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 composed of multiple servers or terminal devices, or as a single server or a single terminal device. When the electronic device is 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, or as a single software or software module. No specific limitation is made herein.

[0039] It should be understood that Figure 1The number of electronic devices shown is merely illustrative. Any number of electronic devices can be used depending on the implementation requirements.

[0040] Continue to refer to Figure 2 The flowchart 200 illustrates some embodiments of the information generation method according to this disclosure. The information generation method includes the following steps:

[0041] Step 201: Obtain the pre-built graph model.

[0042] In some embodiments, the entity executing the information generation method (e.g. Figure 1 The electronic device 101 shown can acquire a pre-built graph model via a wired or wireless connection. This graph model represents the relationships between a first user set, a second user set, and an item set. In practice, the first user set can be at least one livestreamer. The second user set can be a group of viewers watching a livestream sales event. The items in the item set can be items for which a second value operation (sale) is to be performed. The edges in the graph model can have directions. For example, if an edge in the graph model points from a first user to an item, it indicates that the first user in the graph model is performing the second value operation (sale) on the item. If an edge in the graph model points from a second user to a first user, it indicates that the second user in the graph model frequently watches the first user and / or has followed the first user. If an edge in the graph model points from a second user to an item, it indicates that the second user in the graph model is performing a first value operation (e.g., purchasing) on ​​the item. If an edge in the graph model points from a second user to a first user, the value corresponding to the edge can be the number of likes the second user has given to the first user.

[0043] It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections, and other currently known or future wireless connection methods.

[0044] Step 202: Generate first data based on the nodes and edges associated with the nodes representing the first target user in the first user set in the graph model described above.

[0045] In some embodiments, the executing entity may generate first data based on the nodes and edges associated with the nodes representing the first target user in the first user set in the graph model. The first data includes feature information representing the influence of the first target user on the second user set to perform a first value operation on a target item in the item set.

[0046] The first target user is the user in the first user set whose number of information is to be predicted.

[0047] As an example, the execution subject can determine at least one second target user associated with the first target user, the number of likes received by the first target user, and at least one item associated with the first target user from the item set according to the nodes and edges associated with the node representing the first target user in the graph model by querying.

[0048] In some optional implementations of some embodiments, the generating the first data according to the nodes and edges associated with the node representing the first target user in the graph model can include the following steps:

[0049] In the first step, a first subgraph associated with the first target user in the graph model is determined according to the nodes and edges associated with the node representing the first target user in the graph model.

[0050] As an example, the graph model can be Figure 1 the graph model 102. The first subgraph can refer to Figure 3 .

[0051] In the second step, the first information of the edges associated with the node representing the first target user in the first subgraph is determined. The first target user corresponds to at least one second user in the first subgraph. The first target user corresponds to at least one item in the first subgraph. The first information of the edges associated with the node representing the first target user can include the number of edges between the node representing the first target user and the node set representing the at least one second user, the first average value corresponding to each edge between the node representing the first target user and the node set representing the at least one second user, the number of edges between the node representing the first target user and the node set representing the at least one item, the second average value corresponding to each edge between the node representing the first target user and the node set representing the at least one item, and the longest node path information of the first subgraph.

[0052] Here, the first average value corresponding to each edge between the node representing the first target user and the node set representing the at least one second user can be determined by the following steps: first, determining each edge between the node representing the first target user and the node set representing the at least one second user; then, determining the first value set corresponding to each edge; finally, determining the average value of the first value set as the first average value.

[0053] Here, the second average value corresponding to each edge between the node representing the first target user and the node set representing the at least one item can be determined by: first, determining each edge between the node representing the first target user and the node set representing the at least one item; then, determining a second value set corresponding to each edge; and finally, determining an average number corresponding to the second value set as the second average value.

[0054] Thirdly, generating first score information representing the influence of the first target user according to the first subgraph. As an example, the execution subject can generate the first score information representing the influence of the first target user according to the first subgraph by using a PageRank algorithm.

[0055] Fourthly, determining a second subgraph associated with the target item in the graph model according to each node and each edge associated with the node representing the target item in the graph model.

[0056] As an example, the graph model can be a graph model 102. The second subgraph can refer to a graph model 102. Figure 1 . Figure 4 .

[0057] Fifthly, determining second information of each edge associated with the node representing the target item in the second subgraph. The target item corresponds to at least one second user in the second subgraph. The target item corresponds to at least one first user in the second subgraph. Further, the second information of each edge associated with the node representing the target item can include: the number of edges between the node representing the target item and a node set representing the at least one second user, a third average value corresponding to each edge between the node representing the target item and the node set representing the at least one second user, the number of edges between the node representing the target item and a node set representing the at least one first user, a fourth average value corresponding to each edge between the node representing the target item and the node set representing the at least one first user, and longest node path information of the second subgraph.

[0058] Here, the third average value corresponding to each edge between the node representing the target item and the node set representing the at least one second user can be determined by: first, determining each edge between the node representing the target item and the node set representing the at least one second user; then, determining a third value set corresponding to each edge; and finally, determining an average number corresponding to the third value set as the third average value.

[0059] Here, the fourth average value corresponding to each edge between the node representing the target item and the set of nodes representing the at least one first user can be determined by first determining each edge between the node representing the target item and the set of nodes representing the at least one first user. Then, a set of fourth values corresponding to each edge is determined. Finally, an average value of the set of fourth values is determined as the fourth average value.

[0060] In the sixth step, the second score information representing the influence of the target item is generated according to the second subgraph. As an example, the second score information representing the influence of the target item can be generated in various ways according to the second subgraph.

[0061] In the seventh step, the first data is generated according to the first information, the first score information, the second information, and the second score information. As an example, the execution subject can combine the first information, the first score information, the second information, and the second score information to obtain the first data.

[0062] Optionally, the second score information representing the influence of the target item is generated by using a PageRank algorithm according to the second subgraph.

[0063] In step 203, the second data of the first target user performing a second value operation on the set of items is obtained.

[0064] In some embodiments, the execution subject can obtain the second data of the first target user performing a second value operation on the set of items. The second value operation can be "sales". The second data can be historical data of the first target user performing a second value operation on each item in the set of items.

[0065] As an example, the second data can include: the average value of the first target user performing a second value operation on the set of items in a historical single session, the average value of the first target user performing a second value operation on the target item in a historical single session, the average number of viewers in a single session of the first target user, the average number of categories of the set of items in a single session of the first target user performing a second value operation, and the historical number information of performing a second value operation on the target item in a single session. Among them, the viewers in a single session of the first target user are users whose viewing time is greater than or equal to a predetermined threshold.

[0066] In step 204, the first number information of the first target user performing a second value operation on the target item in a single session at a future predetermined time is generated according to the first data and the second data.

[0067] In some embodiments, the execution subject can generate, according to the first data and the second data, first quantity information of the first target user performing a second value operation on the target item at a future predetermined time. As an example, the execution subject can input the first data and the second data into a pre-trained time sequence neural network to obtain the first quantity information of the first target user performing a second value operation on the target item at a future predetermined time. The time sequence neural network can be one of a recurrent neural network (RNN), a long short-term memory (LSTM) network, and a gated recurrent unit (GRU) network.

[0068] The above various embodiments of the present disclosure have the following beneficial effects: the information generation method of some embodiments of the present disclosure can efficiently and accurately generate first quantity information of the first target user performing a second value operation on the target item at a future predetermined time by introducing a pre-constructed graph model. Specifically, using the Delphi method often requires a large amount of professional labor cost. In addition, the generated item quantity information often has the problem of being not accurate enough. More importantly, the data information generated by using the Delphi method to generate item quantity reference is often not complete enough. Based on this, the information generation method of some embodiments of the present disclosure can first obtain a pre-constructed graph model. The graph model represents the association relationship between the first user set, the second user set, and the item set. Here, the pre-constructed graph model can be used for subsequent generation of first data, so that the subsequently generated quantity information takes into account the characteristic information of the first target user influencing the second user set performing a first value operation on the target item in the item set. This makes the generated first quantity information more accurate. Then, according to each node and each edge associated with the node representing the first target user in the first user set in the graph model, the first data can be generated simply and quickly. The first data includes characteristic information representing the first target user influencing the second user set performing a first value operation on the target item in the item set. Furthermore, the second data of the first target user performing a second value operation on the item set is obtained. The second data provides data support for the generation of the first quantity information. Finally, according to the first data and the second data, the first quantity information of the first target user performing a second value operation on the target item at a future predetermined time can be accurately and conveniently generated.

[0069] Continuing to refer to Figure 5 , another flow 500 of the information generation method according to some other embodiments of the present disclosure is shown. The information generation method includes the following steps:

[0070] Step 501, obtaining a pre-constructed graph model.

[0071] Step 502, generating first data according to each node and each edge associated with a node representing a first target user in the first user set in the graph model.

[0072] Step 503, obtaining second data of the first target user performing a second value operation on the set of items.

[0073] In some embodiments, the specific implementation of steps 501-503 and the resulting technical effects can refer to Figure 2 Steps 201-203 in the corresponding embodiments, which will not be repeated here.

[0074] Step 504, obtaining attribute information of the target item, second number information of performing a second value operation on the target item of the target brand in a single field, third number information of performing a second value operation on the target item in a single field, and fourth number information of performing a second value operation on the item of the category to which the target item belongs in a single field.

[0075] In some embodiments, the execution subject (e.g. Figure 1 The electronic device 101 shown) can obtain attribute information of the target item, second number information of performing a second value operation on the target item of the target brand in a single field, third number information of performing a second value operation on the target item in a single field, and fourth number information of performing a second value operation on the item of the category to which the target item belongs in a single field. As an example, the attribute information of the target item can include: price information of the target item, weight information of the target item, volume information of the target item, and summary information of the target item.

[0076] Step 505, data preprocessing on the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information, to obtain processed first data, processed second data, processed attribute information, processed second number information, processed third number information, and processed fourth number information.

[0077] In some embodiments, the execution subject can perform data preprocessing on the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information, to obtain processed first data, processed second data, processed attribute information, processed second number information, processed third number information, and processed fourth number information.

[0078] As an example, the execution subject can directly perform One-Hot encoding processing on the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information to obtain processed first data, processed second data, processed attribute information, processed second number information, processed third number information, and processed fourth number information.

[0079] In some optional implementations of some embodiments, the data preprocessing of the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information to obtain processed first data, processed second data, processed attribute information, processed second number information, processed third number information, and processed fourth number information can include the following steps:

[0080] First, data missing in the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information is completed to obtain completed first data, completed second data, completed attribute information, completed second number information, completed third number information, and completed fourth number information.

[0081] Second, the completed first data, the completed second data, the completed attribute information of the target item, the completed second number information, the completed third number information, and the completed fourth number information are converted to obtain converted first data, converted second data, converted attribute information of the target item, converted second number information, converted third number information, and converted fourth number information.

[0082] Optionally, the data type conversion of the completed first data, the completed second data, the completed attribute information of the target item, the completed second number information, the completed third number information, and the completed fourth number information to obtain converted first data, converted second data, converted attribute information of the target item, converted second number information, converted third number information, and converted fourth number information can include the following steps:

[0083] First, the completed first data, the completed second data, the completed second number information, the completed third number information, and the completed fourth number information are converted to floating-point data.

[0084] The second step is to convert the continuous data in the completed attribute information into floating-point data.

[0085] The third step involves assigning labels to data containing ordered categorical variables in the completed attribute information. This results in labeled data. Ordered categorical variables can be those where there is a degree of difference between categories. For example, urine glucose test results can be categorized using -, ±, +, ++, and +++. Treatment efficacy can be categorized as cured, significantly effective, improved, or ineffective.

[0086] The fourth step involves responding to the presence of unordered categorical variables in the completed attribute information. These unordered categorical variables are then one-hot encoded to obtain the one-hot encoded data. An unordered categorical variable refers to a category or attribute where there is no difference in degree or order. For example, it could be a binary category, such as gender (male, female) or drug response (negative, positive).

[0087] Step 506: Input the processed dataset into the pre-trained regression network to obtain the first number of times the first target user performs the second value operation on the target item in a single session at a predetermined time in the future.

[0088] In some embodiments, the execution entity may input the processed dataset into a pre-trained regression network to obtain information on the first number of times the first target user performs a second value operation on the target item in a single session at a predetermined future time. The regression network may be an extreme gradient boosting (XGBoost) or support vector machine (SVM) neural network used for regression tasks.

[0089] Here, the loss function of the regression network can be the squared loss function:

[0090] L=(y i -r i ) 2 ,

[0091] Where L is the loss value. i can be the target item. y i r represents the actual number of second-value operations performed by the first target user on the target item. i This can refer to the output of the regression network for the target item.

[0092] from Figure 5 It can be seen from this that, with Figure 2 Compared to the description of some corresponding embodiments,Figure 5 The flow 500 of the information generation method in some embodiments corresponds to the steps of obtaining the aspect data associated with the target item, processing the aspect data associated with the target item, and how to generate the first number of information. Thus, the schemes described by these embodiments can more comprehensively and accurately generate the first number of information of the first target user performing the second value operation on the target item at the future predetermined time by introducing aspect data and regression networks.

[0093] With reference to the foregoing Figure 6 , as an implementation of the foregoing methods, the present disclosure provides some embodiments of an information generation device, which device embodiments correspond to Figure 2 the foregoing method embodiments, and the device can be specifically applied to various electronic devices.

[0094] As shown in Figure 6 , the information generation device 600 of some embodiments includes a first obtaining unit 601, a first generating unit 602, a second obtaining unit 603, and a second generating unit 604. The first obtaining unit 601 is configured to obtain a pre-constructed graph model, wherein the graph model represents the association relationship between the first user set, the second user set, and the item set. The first generating unit 602 is configured to generate first data according to each node and each edge associated with the node representing the first target user in the first user set in the graph model, wherein the first data includes feature information representing the influence of the first target user on the second user set performing the first value operation on the target item in the item set. The second obtaining unit 603 is configured to obtain second data of the first target user performing the second value operation on the item set. The second generating unit 604 is configured to generate the first number of information of the first target user performing the second value operation on the target item at a future predetermined time according to the first data and the second data.

[0095] In some optional implementations of some embodiments, the second generation unit 604 of the information generation apparatus 600 can be further configured to: obtain attribute information of the target item, second number information of performing the second value operation on the target item by the target brand, third number information of performing the second value operation on the target item, and fourth number information of performing the second value operation on an item belonging to the category of the target item; perform data preprocessing on the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information to obtain processed first data, processed second data, processed attribute information, processed second number information, processed third number information, and processed fourth number information; and generate the first number information of performing the second value operation on the target item by the target user at a future predetermined time based on the processed first data, the processed second data, the processed attribute information, the processed second number information, the processed third number information, and the processed fourth number information.

[0096] In some optional implementations of some embodiments, the second generation unit 604 of the information generation apparatus 600 can be further configured to: input the processed data set into a pre-trained regression network to obtain the first number information of performing the second value operation on the target item by the target user at a future predetermined time.

[0097] In some optional implementations of some embodiments, the second generation unit 604 of the information generation apparatus 600 can be further configured to: perform data completion on the data with missing data in the first data, the second data, the attribute information of the target item, the second number information, the third number information, and the fourth number information to obtain completed first data, completed second data, completed attribute information, completed second number information, completed third number information, and completed fourth number information; and perform data type conversion on the completed first data, the completed second data, the completed attribute information of the target item, the completed second number information, the completed third number information, and the completed fourth number information to obtain converted first data, converted second data, converted attribute information of the target item, converted second number information, converted third number information, and converted fourth number information.

[0098] In some optional implementations of some embodiments, the second generating unit 604 of the information generating apparatus 600 can be further configured to: convert the first completed data, the second completed data, the second number information, the third number information and the fourth number information into floating point data; convert continuous data in the completed attribute information into floating point data; perform label encoding on ordered categorical variable data in the completed attribute information to obtain label encoded data; and perform one-hot encoding on unordered categorical variable data in the completed attribute information to obtain one-hot encoded data.

[0099] In some optional implementations of some embodiments, the first generating unit 602 of the information generating apparatus 600 can be further configured to: determine a first subgraph associated with a first target user in the graph model according to each node and each edge associated with the node representing the first target user in the graph model; determine first information of each edge associated with the node representing the first target user in the first subgraph; generate first score information representing influence of the first target user according to the first subgraph; determine a second subgraph associated with a target item in the graph model according to each node and each edge associated with the node representing the target item in the graph model; determine second information of each edge associated with the node representing the target item in the second subgraph; generate second score information representing influence of the target item according to the second subgraph; and generate the first data according to the first information, the first score information, the second information and the second score information.

[0100] In some optional implementations of some embodiments, the first generating unit 602 of the information generating apparatus 600 can be further configured to: generate the second score information representing influence of the target item according to the second subgraph by using a page rank algorithm.

[0101] It can be understood that the units described in the apparatus 600 correspond to the respective steps in the method described with reference to Figure 2 Thus, the operations, features and advantages described above for the method also apply to the apparatus 600 and the units included therein, which will not be described here again.

[0102] Reference is made below to Figure 7 which shows a structural schematic diagram of an electronic device 700 suitable for implementing some embodiments of the present disclosure. Figure 7 The electronic device shown is merely an example and should not impose any limitation on the functions and use range of the embodiments of the present disclosure.

[0103] like Figure 7 As shown, the electronic device 700 may include a processing unit (e.g., a central processing unit, a graphics processor, etc.) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage device 708 into a random access memory (RAM) 703. The RAM 703 also stores various programs and data required for the operation of the electronic device 700. The processing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0104] Typically, the following devices can be connected to I / O interface 705: input devices 706 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 707 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 708 including, for example, magnetic tapes, hard disks, etc.; and communication devices 709. Communication device 709 allows electronic device 700 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 7 An electronic device 700 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 7 Each box shown can represent a device or multiple devices as needed.

[0105] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 709, or installed from storage device 708, or installed from ROM 702. When the computer program is executed by processing device 701, it performs the functions defined in the methods of some embodiments of this disclosure.

[0106] Note that the computer-readable medium or media used to provide the computer program sequence to the computer system can be embedded in a computer program product, which comprises all the respective features, which are provided with the computer program sequence, and which are enumerated above. It is understood that the computer-readable medium or media described herein are included in the computer program product, or are a component of the computer program product. In some embodiments of the disclosure, the computer-readable storage medium can be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In some embodiments of the disclosure, a computer-readable storage medium can be any tangible medium that contains, or stores a program for use by or in connection with an instruction execution system, apparatus, or device. In some embodiments of the disclosure, a computer-readable signal medium can include a computer-readable storage medium in baseband or propagated as a carrier wave in a propagated data signal, which contains a computer-readable program code. Such a propagated signal can take a wide variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium that is not a computer-readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0107] In some embodiments, the client, server, or both can communicate using any current known or future developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet, and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any current known or future developed networks.

[0108] The computer readable medium can be included in the apparatus; or can exist separately from the electronic device. The computer readable medium bears one or more programs, and when the one or more programs are executed by the electronic device, the electronic device is caused to: obtain a pre-constructed graph model, wherein the graph model represents an association relationship among a first user set, a second user set and an item set; generate first data according to each node and each edge associated with a node representing a first target user in the first user set in the graph model, wherein the first data includes feature information representing that the first target user influences the second user set to perform a first value operation on a target item in the item set; obtain second data representing that the first target user performs a second value operation on the item set; and generate first number information representing that the first target user performs the second value operation on the target item in a single session at a future predetermined time according to the first data and the second data.

[0109] Computer program code for carrying out operations of some embodiments of the present disclosure can be written in any of one or more programming languages, including object oriented programming languages such as Java, Smalltalk, C++, or conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0110] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other processing devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0111] The units described in some embodiments of the present disclosure can be implemented by means of software, or by means of hardware. The units described can also be provided in a processor, for example, a processor can be described as comprising a first obtaining unit, a first generating unit, a second obtaining unit and a second generating unit. In some cases, the names of the units do not constitute a limitation on the units themselves, for example, the first obtaining unit can also be described as a unit for obtaining a pre-constructed graph model.

[0112] The functions described above can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that can be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system on a chip (SOCs), complex programmable logic devices (CPLDs), etc.

[0113] The above description is merely exemplary of some preferred embodiments of the present disclosure and of the application of the principles of the present disclosure. It is therefore intended that the scope of the present disclosure be limited only by the scope of the claims appended hereto and that the scope of the claims should encompass modifications and variations of the exemplary embodiments which have been disclosed herein as well as other implementations that come within the scope of the appended claims and their equivalents. For example, the features of the above described embodiments can be combined with other features disclosed in the present disclosure (but not limited to) to form other technical solutions.

Claims

1. An information generation method, comprising: obtaining a pre-constructed graph model, wherein the graph model represents an association relationship among a first user set, a second user set and an item set, edges in the graph model are directional, the first user set is at least one anchor, the second user set is a set of audiences watching live goods, and an item in the item set is an item to be executed with a second value operation, in response to an edge pointing from a first user to an item, it represents that the first user is executing a second value operation on the item, in response to an edge pointing from a second user to a first user, it represents that the second user often watches the first user and / or has followed the first user, in response to an edge pointing from a second user to an item, it represents that the second user executes a first value operation on the item, in response to an edge pointing from a second user to a first user, a numerical value corresponding to the edge is a like amount of the second user to the first user, the first value operation is a sales operation, and the second value operation is a purchase operation; generating first data according to each node and each edge associated with a node representing a first target user in the first user set in the graph model, wherein the first data includes feature information representing that the first target user influences the second user set to execute a first value operation on a target item in the item set; obtaining second data of the first target user executing a second value operation on the item set; generating first number information of the first target user executing a second value operation on the target item in a single session at a future predetermined time according to the first data and the second data.

2. The method of claim 1, wherein, The generating of the first number information of the first target user executing a second value operation on the target item in a single session at a future predetermined time according to the first data and the second data comprises: obtaining attribute information of the target item, second number information of a target item of a target brand executed in a single session, third number information of the target item executed in a single session, and fourth number information of an item of a category to which the target item belongs executed in a single session; performing data preprocessing on the first data, the second data, the attribute information of the target item, the second number information, the third number information and the fourth number information to obtain processed first data, processed second data, processed attribute information, processed second number information, processed third number information and processed fourth number information; generating the first number information of the first target user executing a second value operation on the target item in a single session at a future predetermined time based on the processed first data, the processed second data, the processed attribute information, the processed second number information, the processed third number information and the processed fourth number information.

3. The method of claim 2, wherein, The generating of the first number information of the first target user executing a second value operation on the target item in a single session at a future predetermined time comprises: inputting the processed data set into a pre-trained regression network to obtain the first number information of the first target user executing a second value operation on the target item in a single session at a future predetermined time.

4. The method of claim 2, wherein, The step of preprocessing the first data, the second data, the attribute information of the target item, the second quantity information, the third quantity information, and the fourth quantity information to obtain processed first data, processed second data, processed attribute information, processed second quantity information, processed third quantity information, and processed fourth quantity information includes: Data missing data in the first data, the second data, the attribute information of the target item, the second quantity information, the third quantity information, and the fourth quantity information are filled in to obtain the filled first data, the filled second data, the filled attribute information, the filled second quantity information, the filled third quantity information, and the filled fourth quantity information. The data types of the completed first data, the completed second data, the completed target item attribute information, the completed second number information, the completed third number information, and the completed fourth number information are converted to obtain the converted first data, the converted second data, the converted target item attribute information, the converted second number information, the converted third number information, and the converted fourth number information.

5. The method of claim 4, wherein, The process of converting the data types of the completed first data, the completed second data, the completed target item attribute information, the completed second quantity information, the completed third quantity information, and the completed fourth quantity information to obtain the converted first data, the converted second data, the converted target item attribute information, the converted second quantity information, the converted third quantity information, and the converted fourth quantity information includes: Convert the completed first data, the completed second data, the completed second number information, the completed third number information, and the completed fourth number information into floating-point data; In response to the presence of continuous data in the completed attribute information, the continuous data is converted into floating-point data; In response to the presence of ordered categorical variables in the completed attribute information, the ordered categorical variable data is labeled and encoded to obtain the labeled data. In response to the presence of unordered categorical variables in the completed attribute information, the unordered categorical variables are one-hot encoded to obtain the one-hot encoded data.

6. The method of claim 1, wherein, The step of generating first data based on the nodes and edges associated with the nodes representing the first target user in the graph model includes: Based on the nodes and edges associated with the nodes representing the first target user in the graph model, determine the first subgraph associated with the first target user in the graph model; Determine the first information of each edge in the first subgraph associated with the node representing the first target user; Based on the first subgraph, generate first score information representing the influence of the first target user; determine, according to the nodes and the edges associated with the node representing the target item in the graph model, a second subgraph associated with the target item in the graph model; determine second information of the edges associated with the node representing the target item in the second subgraph; generate, according to the second subgraph, second score information representing influence of the target item; generate the first data according to the first information, the first score information, the second information, and the second score information.

7. The method of claim 6, wherein, The generating, according to the second subgraph, of the second score information representing influence of the target item includes: generating, according to the second subgraph, the second score information representing influence of the target item by using a page rank algorithm. 8.An information generation apparatus, comprising: a first obtaining unit configured to obtain a pre-constructed graph model, wherein the graph model represents an association relationship among a first user set, a second user set, and an item set, edges in the graph model are directed, the first user set is at least one anchor, the second user set is a set of audiences watching live goods, and an item in the item set is an item to be subjected to a second value operation, in response to an edge being directed from a first user to an item, it is represented that the first user is performing a second value operation on the item, in response to an edge being directed from a second user to a first user, it is represented that the second user frequently watches the first user and / or has followed the first user, in response to an edge being directed from a second user to an item, it is represented that the second user performs a first value operation on the item, in response to an edge being directed from a second user to a first user, a value corresponding to the edge is a number of likes of the second user to the first user, the first value operation is a sales operation, and the second value operation is a purchase operation; a first generation unit configured to generate, according to nodes and edges associated with a node representing a first target user in the first user set in the graph model, first data, wherein the first data includes feature information representing that the first target user influences the second user set to perform a first value operation on a target item in the item set; a second obtaining unit configured to obtain second data of a second value operation performed by the first target user on the item set; a second generation unit configured to generate, according to the first data and the second data, first number information of a single session of the first target user performing the second value operation on the target item in a future predetermined time. 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 implement the method of any one of claims 1-7.

10. A computer readable medium having stored thereon a computer program, wherein, The program is executed by the processor to implement the method of any one of claims 1-7.

Images