Event popularity calculation method and device and related product

Abstract

The invention discloses an event popularity calculation method and device and a related product. Screening the plurality of object interaction data sets according to the to-be-calculated event to obtain interaction data subsets respectively corresponding to the to-be-calculated event in the plurality of object interaction data sets; performing calculation processing on the plurality of object interaction data sets and the plurality of interaction data subsets to obtain set attention degrees corresponding to the plurality of object sets for the to-be-calculated event; and performing calculation processing on the set weight coefficients respectively corresponding to the plurality of object sets and the set attention respectively corresponding to the to-be-calculated event by the plurality of object sets to obtain an event popularity result corresponding to the to-be-calculated event. Thus, according to the method and the device, the activeness of each object set and the set attention of each object set on the to-be-calculated event can be utilized to realize the heat calculation of the to-be-calculated event, so that the condition that the heat calculation of the event has deviation in related technologies can be avoided.

Description

Technical Field

[0001] This application relates to the technical field of heat calculation, and particularly to a method, apparatus, and related products for calculating the heat of an event. Background Art

[0002] The heat of an event can represent the degree of attention and influence of the event in the public eye. Recommending events based on the heat of the event can facilitate the public to more quickly understand the latest events and maintain a close connection with society. In related technologies, the heat of an event is usually calculated based on the interaction data of the public with respect to the event. However, due to the differences in the preferences of the public for events, there may be some members of the public who exhibit excessive interaction behavior towards the event, resulting in a deviation in the calculation of the heat of the event.

[0003] Therefore, how to avoid deviation in the calculation of the heat of an event has become an urgent technical problem in the current field. Summary of the Invention

[0004] Embodiments of this application provide a method, apparatus, and related products for calculating the heat of an event, aiming to avoid deviation in the calculation of the heat of an event.

[0005] The first aspect of this application provides a method for calculating the heat of an event, including:

[0006] Obtain an event to be calculated, a plurality of object sets, and the object interaction data sets respectively corresponding to the plurality of object sets, where each object set has a different interest in the event;

[0007] Obtain the set weight coefficients respectively corresponding to the plurality of object sets according to the plurality of object sets or the plurality of object interaction data sets, where each set weight coefficient represents the activity of each object set;

[0008] Perform a screening process on the plurality of object interaction data sets according to the event to be calculated to obtain the interaction data subsets respectively corresponding to the event to be calculated in the plurality of object interaction data sets;

[0009] Perform a calculation process on the plurality of object interaction data sets and the plurality of interaction data subsets to obtain the set attentions respectively corresponding to the plurality of object sets for the event to be calculated;

[0010] Perform a calculation process on the set weight coefficients respectively corresponding to the plurality of object sets and the set attentions respectively corresponding to the plurality of object sets for the event to be calculated to obtain the event heat result corresponding to the event to be calculated.

[0011] The second aspect of this application provides a device for calculating the heat of an event, including:

[0012] An event set acquisition unit for acquiring events to be calculated, multiple object sets, and object interaction data sets respectively corresponding to the multiple object sets, where each object set has different interests in the event;

[0013] A weight coefficient acquisition unit for obtaining set weight coefficients respectively corresponding to the multiple object sets according to the multiple object sets or multiple object interaction data sets, where each set weight coefficient represents the activity of each object set;

[0014] A data subset acquisition unit for screening and processing the multiple object interaction data sets according to the events to be calculated, and obtaining interaction data subsets respectively corresponding to the events to be calculated in the multiple object interaction data sets;

[0015] A set attention degree acquisition unit for performing calculation processing on the multiple object interaction data sets and multiple interaction data subsets, and obtaining set attention degrees respectively corresponding to the multiple object sets for the events to be calculated;

[0016] A heat result acquisition unit for performing calculation processing on the set weight coefficients respectively corresponding to the multiple object sets and the set attention degrees respectively corresponding to the multiple object sets for the events to be calculated, and obtaining an event heat result corresponding to the events to be calculated.

[0017] The third aspect of the present application provides a computer device, and the device includes a processor and a memory:

[0018] The memory is used for storing a computer program and transmitting the computer program to the processor;

[0019] The processor is used for executing the steps of the event heat calculation method provided in the first aspect according to the instructions in the computer program.

[0020] The fourth aspect of the present application provides a computer-readable storage medium, and the computer-readable storage medium is used for storing a computer program, and when the computer program is executed by a computer device, the steps of the event heat calculation method provided in the first aspect are implemented.

[0021] The fifth aspect of the present application provides a computer program product, including a computer program, and when the computer program is executed by a computer device, the steps of the event heat calculation method provided in the first aspect are implemented.

[0022] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0023] In the technical solution of this application, first, the event to be calculated, multiple object sets, and the object interaction data sets respectively corresponding to the multiple object sets can be obtained. After that, based on the multiple object sets or the multiple object interaction data sets, the set weight coefficients respectively corresponding to the multiple object sets can be obtained, and the multiple object interaction data sets can be filtered according to the event to be calculated to obtain the interaction data subsets respectively corresponding to the event to be calculated in the multiple object interaction data sets. Finally, the multiple object interaction data sets and the multiple interaction data subsets can be calculated to obtain the set attentions respectively corresponding to the multiple object sets for the event to be calculated, and the set weight coefficients respectively corresponding to the multiple object sets and the set attentions respectively corresponding to the multiple object sets for the event to be calculated are calculated to obtain the event heat result corresponding to the event to be calculated. It should be noted that each object set has different interests in the event, and each set weight coefficient represents the activity of each object set.

[0024] It can be seen that in this application, first, the set weight coefficient corresponding to each object set can be calculated, and the set attention of each object set for the event to be calculated can be calculated. After that, the set weight coefficients and set attentions respectively corresponding to the multiple object sets are calculated to obtain the event heat result corresponding to the event to be calculated. In this way, in this application, the activity of each object set and the set attention of each object set for the event to be calculated can be used to calculate the heat of the event to be calculated, so as to avoid the situation of deviation in the heat calculation of the event in the related technology. Description of the Drawings

[0025] Figure 1 It is a scenario architecture diagram of a method for calculating the heat of an event provided by an embodiment of this application;

[0026] Figure 2 It is a flowchart of a method for calculating the heat of an event provided by an embodiment of this application;

[0027] Figure 3 It is a flowchart for obtaining an object set in a method for calculating the heat of an event provided by an embodiment of this application;

[0028] Figure 4 It is an application schematic diagram of recommended events in a method for calculating the heat of an event provided by an embodiment of this application;

[0029] Figure 5 It is an application schematic diagram of recommended events in another method for calculating the heat of an event provided by an embodiment of this application;

[0030] Figure 6 It is a full flowchart of event heat calculation in a method for calculating the heat of an event provided by an embodiment of this application;

[0031] Figure 7 This is the full flow chart of event processing in a method for calculating event heat provided by an embodiment of the present application;

[0032] Figure 8 This is a schematic structural diagram of a device for calculating event heat provided by an embodiment of the present application;

[0033] Figure 9 This is a schematic structural diagram of a server in an embodiment of the present application;

[0034] Figure 10 This is a schematic structural diagram of a terminal device in an embodiment of the present application. Detailed implementation manners

[0035] The embodiments of the present application will be described below with reference to the accompanying drawings.

[0036] First, several noun terms that may be involved in the following embodiments of the present application will be explained.

[0037] Breaking the circle: It means that a certain event is no longer limited to a specific small group, but has gained wider attention and dissemination.

[0038] U2I2U mode: U2I2U (User to Item to User) is a similarity object discovery method based on collaborative filtering.

[0039] UCB: (Upper Confidence Bound) is a concept commonly used in online learning and exploration-exploitation algorithms, which represents the upper bound of the confidence interval of a certain index. In the present application, UCB is used to estimate the maximum possible value of the event attention of an object set to a certain event.

[0040] As described above, event heat can characterize the degree of attention and influence of an event in the public eye. Recommending events based on event heat can facilitate the public to learn about the latest events more quickly to maintain close contact with society. In the related art, event heat is usually calculated based on the interaction data of the public with the event. However, due to the differences in the public's preferences for events, there may be some members of the public who have excessive interaction behaviors with the event.

[0041] For example, fans may interact with a celebrity's blog or video by giving a lot of likes and forwarding. This may result in a large amount of interactive data for the celebrity's blog or video. When calculating the heat of the blog or video, the event heat of the blog or video is too high. However, when recommending the blog or video based on the calculated event heat, it is found that there is no increase in more interactive data. Therefore, the relevant technical solution is likely to lead to deviations in the heat calculation of the event. Therefore, how to avoid deviations in the heat calculation of events has become a technical problem that needs to be solved urgently in the current field.

[0042] In view of the above problems, a method, device and related products for calculating the heat of an event are provided in the present application, the purpose of which is to avoid deviations in the heat calculation of an event. In the technical solution provided in the present application, first, the event to be calculated, multiple object sets and object interaction data sets corresponding to the multiple object sets can be obtained. Thereafter, the set weight coefficients corresponding to the multiple object sets can be obtained according to the multiple object sets or the multiple object interaction data sets, and the multiple object interaction data sets can be screened and processed according to the event to be calculated to obtain the interaction data subsets corresponding to the event to be calculated in the multiple object interaction data sets. Finally, the multiple object interaction data sets and the multiple interaction data subsets can be calculated and processed to obtain the set attention corresponding to the event to be calculated by the multiple object sets, and the set weight coefficients corresponding to the multiple object sets and the set attention corresponding to the event to be calculated by the multiple object sets can be calculated and processed to obtain the event heat result corresponding to the event to be calculated. Each object set has different interests in the event, and each set weight coefficient represents the activity of each object set.

[0043] It can be seen that in the present application, first, the set weight coefficient corresponding to each object set can be calculated based on multiple object sets or multiple object interaction data sets, and the set attention of each object set to the event to be calculated can be calculated based on multiple object interaction data sets and multiple interaction data subsets. After that, the set weight coefficients and set attention corresponding to the multiple object sets are calculated and processed, and the event heat result corresponding to the event to be calculated can be obtained. In this way, in the present application, the activity of each object set and the set attention of each object set to the event to be calculated can be used to realize the heat calculation of the event to be calculated, so as to avoid the deviation of the heat calculation of the event in the related technology, thereby ensuring the accuracy of the calculated event heat result to a certain extent.

[0044] The execution subject of the event heat calculation method provided by the embodiments of the present application can be a terminal device. For example, obtain the event to be calculated, multiple object sets, and the object interaction data sets corresponding to the multiple object sets on the terminal device. As an example, the terminal device may specifically include, but is not limited to, mobile phones, desktop computers, tablet computers, laptop computers, handheld computers, intelligent voice interaction devices, intelligent home appliances, vehicle-mounted terminals, aircraft, etc. The execution subject of the event heat calculation method provided by the embodiments of the present application can also be a server, that is, the event to be calculated, multiple object sets, and the object interaction data sets corresponding to the multiple object sets can be obtained on the server. In addition, the event heat calculation method provided by the embodiments of the present application can also be executed jointly by the terminal device and the server. Among them, the terminal and the server can be directly or indirectly connected through wired or wireless communication methods, and the present application does not make any restrictions in this regard. Therefore, the embodiments of the present application do not limit the implementation subject of implementing the technical solution of the present application.

[0045] Figure 1 Exemplarily shows a scenario architecture diagram of an event heat calculation method. The figure includes a server and various forms of terminal devices. For example, first, the terminal device can obtain the event to be calculated, multiple object sets, and the object interaction data sets corresponding to the multiple object sets. Thereafter, the server can process the multiple object sets, the multiple object interaction data sets, and the event to be calculated to obtain the set weight coefficients and set attentions corresponding to the multiple object sets. Finally, the server can determine the event heat result corresponding to the event to be calculated based on the set weight coefficients and set attentions corresponding to the multiple object sets. In this way, the situation of deviation in the heat calculation of events in the related art can be avoided. Figure 1 The server shown can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. In addition, the server can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0046] See Figure 2 , this figure is a flowchart of an event heat calculation method provided by the embodiments of the present application. As Figure 2 shown in the event heat calculation method, the following steps are included:

[0047] S201: Obtain the event to be calculated, multiple object sets, and the object interaction data sets corresponding to the multiple object sets.

[0048] In this step, the objects in the object set include the objects in the application that publishes the event to be calculated. The object set is obtained by aggregating multiple objects, and each object set has different interests in the event. For example: Object set A is more interested in events of the photography type, and object set B is more interested in events of the news type.

[0049] The event to be calculated can be a blog post published by an object in the application, or it can be a video published by an object in the application. The event to be calculated is only a partial example, and it can also be determined in actual applications. For example: Object A publishes a blog post about social news in the application. At this time, the blog post about social news can be used as the event to be calculated. It should be noted that in this application, the event to be calculated can be determined by the interaction volume of the published event. For example, an event with a high interaction volume can be selected as the event to be calculated.

[0050] Next, the process of obtaining multiple object sets and the object interaction data sets respectively corresponding to the multiple object sets in this application is introduced. That is, before this application executes the operations of obtaining the event to be calculated, multiple object sets, and the object interaction data sets respectively corresponding to the multiple object sets, it is also possible to obtain multiple objects to be grouped and the object interaction data respectively corresponding to the multiple objects to be grouped. The objects to be grouped include the objects in the application platform that publishes the event to be calculated. The multiple objects to be grouped can be understood as all the objects existing in the application. It should be noted that the relevant data collection and processing in this application book should strictly comply with the requirements of relevant regional laws and regulations when applied in practice, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing behaviors within the scope authorized by laws and regulations and the personal information subject.

[0051] Object interaction data includes object click data, object exposure data, and object like data. Among them, object click data can be understood as the data of object click events, such as the data of object A clicking on event a and the data of object A clicking on event b; object exposure data can be understood as the data of object browsing events, such as the data of object A browsing event a and the data of object A browsing event b; object like data can be understood as the data of object like events, such as the data of object A liking event a and the data of object A liking event b. It should be noted that object interaction data can also include object forwarding data and object comment data, and this object interaction data can also be selected according to the actual situation in actual applications.

[0052] Specifically, in the present application, multiple objects to be grouped can be grouped based on multiple object interaction data to obtain multiple object sets. At this time, the multiple object interaction data can be filtered based on the multiple object sets to obtain object interaction data sets corresponding to the multiple object sets respectively. That is, based on the objects in each object set, the object interaction data corresponding to the multiple objects to be grouped can be filtered to obtain the object interaction data corresponding to the objects in each object set. At this time, the object interaction data in each object set is aggregated to obtain object interaction data sets corresponding to the multiple object sets respectively. In this way, in the present application, multiple objects to be grouped can be divided into multiple object sets, laying a foundation for calculating the popularity of the event to be calculated later.

[0053] Next, the process of obtaining multiple object sets in the present application will be further introduced. As Figure 3 shown, Figure 3 is a flowchart for obtaining an object set in a method for calculating event popularity provided by an embodiment of the present application, Figure 3 which shows steps S2011 - S2014. The specific implementation of steps S2011 - S2014 is as follows:

[0054] S2011: Perform calculation processing on multiple objects to be grouped based on multiple object interaction data to obtain multiple similarity results.

[0055] Specifically, in this step, first, similar object retrieval can be performed on multiple objects to be grouped based on multiple object interaction data to obtain multiple similar object sets, where the interests of each similar object set are different. Thereafter, the object similarity can be calculated for every two objects to be grouped in the multiple similar object sets to obtain multiple similarity results, where one similarity result is calculated based on one object to be grouped and the objects to be grouped that are similar to this object to be grouped among the multiple objects to be grouped.

[0056] For example, the multiple objects to be grouped include a first object to be grouped, a second object to be grouped, a third object to be grouped, and a fourth object to be grouped. At this time, based on the object interaction data of the first object to be grouped and the object interaction data of the third object to be grouped, the first object to be grouped and the third object to be grouped can be divided into a first similar object set (that is, the first object to be grouped and the third object to be grouped watched the same video), and based on the object interaction data of the second object to be grouped and the object interaction data of the fourth object to be grouped, the second object to be grouped and the fourth object to be grouped can be divided into a second similar object set (that is, the second object to be grouped and the fourth object to be grouped read the same article).

[0057] Thereafter, the object similarity calculation is performed on the first object to be grouped and the third object to be grouped in the first set of similar objects to obtain the similarity result of the first object to be grouped and the third object to be grouped, and the object similarity calculation is performed on the second object to be grouped and the fourth object to be grouped in the second set of similar objects to obtain the similarity result of the second object to be grouped and the fourth object to be grouped. In this way, the calculation cost for object similarity calculation can be reduced in the present application.

[0058] It should be noted that in the present application, the similar object retrieval operation can be implemented through the U2I2U method, the keyword retrieval method, and the vector similarity calculation method. For the above three optional implementation methods, the terminal device can select one or combine multiple methods to implement, and the present application does not limit this. Next, the U2I2U method, the keyword retrieval method, and the vector similarity calculation method will be introduced.

[0059] The U2I2U method is specifically embodied as follows: an interaction matrix of the object to be grouped - object interaction data can be constructed to obtain the data interaction degree of each object to be grouped through this interaction matrix. This interaction matrix can be expressed as R(u, i), where R(u, i) represents the data interaction degree between the object to be grouped u and the object interaction data i. Thereafter, the similarity between any two objects to be grouped is calculated to determine multiple sets of similar objects based on the similarity.

[0060] In an implementable embodiment, the similarity between any two objects to be grouped can be obtained through formula (1), and formula (1) is specifically embodied as follows:

[0061] sim(u,v)=R(u,:)*R(v,:) / (||R(u,:)||*||R(v,:)||) Formula (1)

[0062] Among them, u represents the object to be grouped u, v represents the object to be grouped v, R(u,:) represents the interaction vector between the object to be grouped u and the object interaction data, R(v,:) represents the interaction vector between the object to be grouped v and the object interaction data, ||·|| represents the L2 norm of the interaction vector, and the higher the similarity between any two objects to be grouped, the more similar the interests between the any two objects to be grouped.

[0063] The keyword retrieval method is specifically embodied as follows: First, keyword extraction can be performed on the interactive data of multiple objects to obtain the data keywords corresponding to the interactive data of multiple objects respectively. Thereafter, based on the data keywords corresponding to the interactive data of multiple objects respectively, keyword vectors corresponding to multiple objects to be grouped can be constructed. Finally, vector calculations can be performed on the keyword vectors corresponding to any two objects to be grouped respectively until all objects to be grouped are calculated, obtaining multiple similarities (this similarity can be the Jacard similarity). These multiple similarities can determine multiple sets of similar objects. Among them, the object interactive data also includes object-related materials and object historical browsing records. The higher the similarity between any two objects to be grouped, the more similar the interests between these two objects to be grouped.

[0064] The vector similarity calculation method can be the similarity calculation method of the embedding vector. The similarity calculation method of this embedding vector is specifically embodied as follows: In this application, through deep learning technology, multiple objects to be grouped can be mapped into a low-dimensional continuous vector space. In this vector space, the objects to be grouped with similar interests can be close to each other, so as to finally determine multiple sets of similar objects. Or in this application, technologies such as Word2Vec and Node2Vec can also be used to perform embedded processing on the interactive data of multiple objects to obtain the similarities corresponding to multiple objects to be grouped respectively. These multiple similarities can determine multiple sets of similar objects. The higher the similarity, the more similar the interests between any two objects to be grouped corresponding to the similarity.

[0065] It should also be noted that in this application, the object similarity calculation operation can be implemented through a similarity calculation model. Next, the implementation of the object similarity calculation operation through the similarity calculation model will be introduced. First, the interactive data of multiple objects can be constructed to obtain the interactive behavior sequences corresponding to multiple objects to be grouped respectively. For example, the object interactive data of object A includes the data of liking video a and the data of forwarding article b. At this time, the interactive behavior sequence corresponding to object A includes {"liking video a", "forwarding article b"}.

[0066] Thereafter, the interactive behavior sequences corresponding to multiple objects to be grouped respectively can be input into the similarity calculation model to perform similarity calculation on the interactive behavior sequences corresponding to multiple objects to be grouped respectively through the similarity calculation model, so as to obtain multiple similarity results. The value range of this similarity result is [-1, 1]. The higher the similarity, the more similar the interests between any two objects to be grouped corresponding to the similarity. And in this application, by setting a threshold, any object to be grouped with a similarity result higher than the threshold can be determined as a pair of similar objects.

[0067] Further, it should be noted that the similarity calculation model in this application can be obtained through model fine-tuning. That is, multiple sample behavior sequences can be input into a pre-trained model (such as BERT or LLaMA) to fine-tune the pre-trained model until the model fine-tuning cut-off condition is met, and a similarity calculation model is obtained. This similarity calculation model can capture the semantic information and potential patterns contained in the object behavior sequence.

[0068] S2012: Perform a construction process on multiple objects to be clustered and multiple similarity results to obtain a matrix of similar object pairs.

[0069] In this step, each point in the matrix of similar object pairs represents the similarity result between any two objects to be clustered. The number of rows of the matrix of similar object pairs is equal to the number of columns of the matrix of similar object pairs, and the number of rows of the matrix of similar object pairs is equal to the total number of multiple objects to be clustered. In this way, in subsequent processes, multiple objects to be clustered can be clustered based on the similar object pairs to obtain sets of objects with different interests.

[0070] S2013: Perform a solution process on the matrix of similar object pairs to obtain the optimal solution of the matrix corresponding to the matrix of similar object pairs.

[0071] In this step, first, a weighted undirected graph can be constructed based on the matrix of similar object pairs. Then, the weighted undirected graph can be maximally solved to obtain the optimal solution of the undirected graph corresponding to the weighted undirected graph. This optimal solution of the undirected graph can be used as the optimal solution of the matrix. In this way, in subsequent steps, multiple object sets can be determined based on this optimal solution of the matrix. Next, the specific process of constructing the weighted undirected graph in this application will be introduced.

[0072] In this application, clustering can be performed in the weighted connected graph based on the matrix of similar object pairs through a community discovery algorithm, and the community discovery algorithm includes the Louvain algorithm. Specifically, each object to be clustered can be used as a node, and the non-zero elements in the matrix of similar object pairs can be used as the weights of the edges between the nodes, so as to construct a weighted undirected graph G=(V, E, W). Where V represents the set of nodes, each node corresponds to an object to be clustered, E represents the set of edges. If S(u, v)>0, there is an edge between u and v. S(u, v) represents the similarity result between the object to be clustered u and the object to be clustered v. W represents the set of edge weights, W(u, v)=S(u, v), and W(u, v) represents the weight corresponding to the similarity result between the object to be clustered u and the object to be clustered v.

[0073] In an implementable embodiment, the optimal solution of the undirected graph corresponding to the weighted undirected graph can be obtained through formula (2), which can be understood as the Modularity function. Formula (2) is specifically as follows:

[0074] Q = 1 / 2m * Σ(u,v)[W(u,v) - k(u)*k(v) / 2m] * δ(c(u), c(v)) Formula (2)

[0075] Among them, m represents the sum of the weights of all edges in the weighted undirected graph, k(u) represents the degree of node u (i.e., the object to be clustered u) (i.e., the sum of the weights of the edges connected to node u), c(u) represents the community (i.e., the set) to which node u belongs, and δ(c(u), c(v)) represents the indicator function. When node u and node v (i.e., the object to be clustered v) belong to the same community, this indicator function is 1, otherwise it is 0. It can be understood that in this application, through the community discovery algorithm, nodes can be iteratively moved to different communities to maximize the result of formula (2), so as to obtain the optimal solution of the undirected graph (i.e., the final community division result), where each community corresponds to an object set.

[0076] S2014: Obtain multiple object sets according to the optimal matrix solution.

[0077] It should be noted that in this application, the process of clustering multiple objects to be clustered to obtain multiple object sets can be a real-time processing process or a timed processing process. The real-time processing process can be understood as performing the clustering operation only after obtaining the event to be calculated, and the timed processing process can be understood as obtaining the event to be calculated only after performing the clustering operation. This timed processing operation can be performed once a day or once every n hours.

[0078] S202: Obtain the set weight coefficients corresponding to the multiple object sets respectively according to the multiple object sets or multiple object interaction data sets.

[0079] In this step, the set weight coefficient corresponding to each object set can be obtained according to each object set or the object interaction data set corresponding to each object set, where the set weight coefficient can characterize the activity of each object set. In this way, in this application, the activity degree of each object set can be reflected based on the set weight coefficient, so as to avoid individual object sets having too much influence on the heat calculation, and further avoid poor event recommendation effects in the end.

[0080] In the embodiments of the present application, there are various possible implementation manners for the above-mentioned S202, which will be introduced separately below. It should be noted that the implementation manners given in the following introduction are only exemplary descriptions and do not represent all the implementation manners of the embodiments of the present application. For the following multiple optional implementation manners, the terminal device can choose one or combine multiple to implement, and the present application does not make any restrictions on this.

[0081] The first optional implementation manner of S202 is specifically as follows: First, for each object set, the object interaction data set corresponding to the object set can be counted to obtain the first total interaction quantity corresponding to the object set. For example, the object interaction data set corresponding to object set A includes the object like data of to-be-clustered object a and the object like data of to-be-clustered object b. At this time, the object like data of to-be-clustered object a and the object like data of to-be-clustered object b can be counted, and the like quantity corresponding to object set A can be used as the first total interaction quantity corresponding to object set A.

[0082] After that, for multiple object sets, the object interaction data sets respectively corresponding to the multiple object sets can be counted to obtain the second total interaction quantities corresponding to the multiple object sets. For example, the object interaction data set corresponding to object set A includes the object like data of to-be-clustered object a and the object like data of to-be-clustered object b, and the object interaction data set corresponding to object set B includes the object comment data of to-be-clustered object c and the object comment data of to-be-clustered object d. At this time, the object like data of to-be-clustered object a and the object like data of to-be-clustered object b can be counted, and the object comment data of to-be-clustered object c and the object comment data of to-be-clustered object d can be counted, and the like quantity corresponding to object set A and the click quantity corresponding to object set B can be used as the second total interaction quantities of object set A and object set B.

[0083] Finally, for each first total interaction quantity, the first total interaction quantity and the second total interaction quantity can be calculated to obtain the set weight coefficient of the object set corresponding to the first total interaction quantity, and for multiple first total interaction quantities, the set weight coefficients respectively corresponding to the multiple object sets can be obtained. In this way, the set weight coefficient of the object can be determined based on the object interaction data set of the object set, that is, the higher the activity degree of the object set, the greater the impact on the overall popularity.

[0084] In an implementable embodiment, the set weight coefficients respectively corresponding to the multiple object sets can be obtained through formula (3), and formula (3) is specifically as follows:

[0085]

[0086] where, w iCharacterize the set weight coefficient corresponding to the $i$-th object set, $u\in c$ i Characterize the object $u$ to be grouped in the $i$-th object set, $u\in c$ j Characterize the object $u$ to be grouped in the $j$-th object set, $f(u)$ represents the object interaction data of the object $u$ to be grouped, and $N$ represents all object sets. It should be noted that in practical applications of this application, different weights can be assigned to different types of object interaction data, and then weighted and summed to obtain the comprehensive object interaction data $f(u)$. For example: $f(u)=\alpha\times$ number of likes $(u)+\beta\times$ number of comments $(u)+\gamma\times$ number of forwards $(u)$, where $\alpha$, $\beta$, and $\gamma$ are preset weight coefficients and can be adjusted according to specific scenarios.

[0087] The second alternative implementation of S202 is specifically as follows: First, multiple object sets can be counted to obtain the total number of the first objects corresponding to each of the multiple object sets and the total number of the second objects corresponding to the multiple object sets. For example: The multiple object sets include object set A, object set B, and object set C. Object set A includes the object $a$ to be grouped and the object $b$ to be grouped. Object set B includes the object $c$ to be grouped and the object $d$ to be grouped. Object set C includes the object $d$ to be grouped and the object $e$ to be grouped. At this time, the total number of the first objects corresponding to object set A can be 2, the total number of the first objects corresponding to object set B can be 2, the total number of the first objects corresponding to object set C can be 2, and the total number of the second objects corresponding to object set A, object set B, and object set C can be 6.

[0088] After that, for each total number of the first objects, the total number of the first objects and the total number of the second objects can be calculated to obtain the set weight coefficient of the object set corresponding to the total number of the first objects, and for the multiple total numbers of the first objects, the set weight coefficients corresponding to each of the multiple object sets can be obtained. In this way, the set weight coefficient of the object can be determined based on the object set, that is, the higher the number of objects in the object set, the greater the impact on the overall popularity.

[0089] In an implementable embodiment, the set weight coefficients corresponding to each of the multiple object sets can be obtained through formula (4), and formula (4) is specifically as follows:

[0090]

[0091] where $c$ i_m represents the total number of the $i$-th object set, $c$ j_m represents the total number of the $j$-th object set.

[0092] The third alternative implementation of S202 is specifically as follows: First, multiple object sets can be filtered to obtain specific object sets corresponding to the multiple object sets respectively. The specific objects in the specific object set include objects with the number of fans greater than a preset number in a specific application. It can be understood that such specific objects usually have a relatively high influence in the speech in the object set.

[0093] After that, multiple specific object sets can be counted to obtain the first specific total quantity corresponding to each of the multiple specific object sets and the second specific total quantity corresponding to the multiple specific object sets. For example: The multiple specific object sets include specific object set A and specific object set B. Specific object set A includes specific object a and specific object b. Specific object set B includes specific object c and specific object d. At this time, the first specific total quantity corresponding to specific object set A is 2, the first specific total quantity corresponding to specific object set B is 2, and the second specific total quantity corresponding to specific object set A and specific object set B is 4.

[0094] Finally, for each first specific total quantity, the first specific total quantity and the second specific total quantity can be calculated to obtain the set weight coefficient of the object set corresponding to the first specific total quantity. And for multiple first specific total quantities, the set weight coefficients corresponding to the multiple object sets can be obtained. In this way, the set weight coefficient of the object can be determined based on the specific object set in the object set. That is, in some scenarios, the specific object set may have a greater impact on the overall popularity.

[0095] In an implementable embodiment, the set weight coefficients corresponding to the multiple object sets can be obtained through formula (5). Formula (5) is specifically as follows:

[0096]

[0097] where c i_tm represents the total quantity of the specific object set in the i-th object set, and c j_tm represents the total quantity of the specific object set in the j-th object set.

[0098] It should be further noted that in actual applications, if it is necessary to combine the above three alternative implementations to determine the set weight coefficients corresponding to the multiple object sets, the weighted average method can be used to achieve this (that is, set corresponding weight coefficients for each alternative implementation).

[0099] S203: Filter the multiple object interaction data sets according to the to-be-calculated event to obtain interaction data subsets corresponding to the to-be-calculated event in the multiple object interaction data sets respectively.

[0100] In this step, the subset of interaction data includes the subset of data corresponding to the event to be calculated in the object interaction dataset, such as the data of operations like giving a like or making a comment on the event to be calculated. It can be understood that for each object interaction dataset, the object interaction dataset can be filtered based on the event to be calculated to obtain the subset of interaction data corresponding to the object interaction dataset. In this way, for multiple object interaction datasets, subsets of interaction data corresponding to the multiple object interaction datasets can be obtained. Thus, in this application, the interaction data of different object sets for an event can be pre-screened, so that the degree of attention of different object sets to the event can be calculated subsequently.

[0101] S204: Perform calculation processing on the multiple object interaction datasets and the multiple subsets of interaction data to obtain the set attention degrees corresponding to the multiple object sets for the event to be calculated.

[0102] In this step, under the action of the multiple object interaction datasets and the multiple subsets of interaction data, the set attention degree of each object set for the event to be calculated can be calculated. Thus, in this application, the degree of attention of different object sets to the event can be considered, so that the true popularity of the event can be calculated more accurately subsequently.

[0103] In the embodiments of this application, there are multiple possible implementation manners for the above-mentioned S204, which are introduced separately below. It should be noted that the implementation manners given in the following introduction are only for illustrative purposes and do not represent all the implementation manners of the embodiments of this application. For the following two alternative implementation manners, the terminal device can choose one or combine multiple to implement, and this application does not make any restrictions.

[0104] The first alternative implementation manner of S204 is specifically as follows: First of all, it should be noted that the object interaction dataset includes the total click data and the total like data, and the subset of interaction data includes the click data and the like data for the event to be calculated. The total click data can be understood as the total click data of the objects in the object set corresponding to the object interaction dataset for all events. At this time, the click can be understood as an object clicking on an event. For example, the object click data of object a and object b in object set A can be combined into the total click data. The click data for the event to be calculated can be understood as the total click data of the objects in the object set corresponding to the object interaction dataset for the event to be calculated. At this time, the click can be understood as an object clicking on the event to be calculated. For example, the object click data of object a in object set A for the event to be calculated and the object click data of object b for the event to be calculated can be combined into the click data for the event to be calculated.

[0105] The first like data includes the total like data of the objects in the object set corresponding to the object interaction data set for all events. At this time, the like can be understood as an object liking an event. For example, the object like data of object a and object b in object set A can be combined into the total like data. The like data for the event to be calculated includes the total like data of the objects in the object set corresponding to the object interaction data set for the event to be calculated. At this time, the like can be understood as an object liking the event to be calculated. For example, the object like data of object a in object set A for the event to be calculated and the object like data of object b for the event to be calculated can be combined into the like data for the event to be calculated.

[0106] Specifically, for each object set, the total click data and click data corresponding to the object set can be divided to obtain the average click-through rate corresponding to the object set, and the total like data and like data corresponding to the object set can be divided to obtain the average like rate corresponding to the object set. Thereafter, for multiple object sets, the average click-through rates and average like rates respectively corresponding to the multiple object sets can be summed to obtain the set attention degrees respectively corresponding to the multiple object sets for the event to be calculated. That is, each object set can obtain the set attention degree for the event to be calculated based on the average click-through rate and average like rate corresponding to the object set. In this way, in the present application, the attention degree of the object set can be determined based on the interaction intensity of the objects in the object set for the event to be calculated, so as to improve the accuracy of subsequent popularity calculation.

[0107] In a realizable implementation manner, the set attention degrees respectively corresponding to multiple object sets for the event to be calculated can be obtained through formula (6), and formula (6) is specifically as follows:

[0108] K i =μ 1 *CTR(c i ,e)+μ 2 *LTR(c i ,e) Formula (6)

[0109] Wherein, K i represents the set attention degree corresponding to the i-th object set for the event to be calculated, e represents the event to be calculated, CTR represents the average click-through rate, LTR represents the average like rate, and μ 1 and μ 2 represent preset weight coefficients. It should be noted that other interaction indicators, such as average comment rate and average repost rate, can also be introduced in the present application to obtain a more comprehensive set attention degree.

[0110] The second alternative implementation of S204 is specifically as follows: First of all, it should be noted that the object interaction dataset includes total exposure data, and the interaction data subset includes click data and exposure data for the event to be calculated. Among them, the total exposure data can be understood as the total browsing data of objects in the object set corresponding to this object interaction dataset for all events. At this time, the exposure can be understood as an object browsing an event. For example, the object browsing data of object a in object set A for event a and the object browsing data of object b for event b can be combined into the total exposure data. The exposure data for the event to be calculated can be understood as the total browsing data of objects in the object set corresponding to this object interaction dataset for the event to be calculated. At this time, the exposure can be understood as an object browsing the event to be calculated. For example, the object browsing data of to-be-mined object a in object set A for the event to be calculated and the object browsing data of to-be-mined object b for the event to be calculated can be combined into the exposure data for the event to be calculated.

[0111] Specifically, for each object set, the click data and exposure data corresponding to this object set can be divided to obtain the average click-through rate of this object set for the event to be calculated. Thereafter, for each object set, the average click-through rate and total exposure data corresponding to this object set can be subjected to confidence processing to obtain the maximum confidence value of this object set for the event to be calculated. Further, for multiple object sets, the maximum confidence values corresponding to the multiple object sets respectively can be used as the set attentions corresponding to the multiple object sets for the event to be calculated respectively. That is, each object set can obtain the set attention corresponding to this object set based on the average click-through rate and total exposure data corresponding to this object set. In this way, in the present application, the attention of the object set can be determined based on the interaction intensity of the objects in the object set for the event to be calculated, so as to improve the accuracy of subsequent popularity calculation.

[0112] In an implementable embodiment, the set attentions corresponding to the multiple object sets for the event to be calculated respectively can be obtained through formulas (7)-(8). Formulas (7)-(8) are specifically as follows:

[0113] CTR i =∑ u∈i d u / ∑ u∈i p u Formula (7)

[0114]

[0115] Among them, CTR i represents the average click-through rate corresponding to the i-th object set for the event to be calculated respectively, d uCharacterize the click data of the object u to be recommended for the event to be calculated, p u Characterize the exposure data of the object u to be recommended for the event to be calculated, UCB CTR Characterize the upper bound of the confidence interval for estimating the CTR (i.e., the set attention). It should be noted that in this application, other interaction indicators, such as the average comment rate and the average repost rate, can also be introduced to obtain a more comprehensive set attention.

[0116] S205: Perform calculation processing on the set weight coefficients corresponding to the multiple object sets and the set attentions corresponding to the multiple object sets for the event to be calculated, to obtain the event heat result corresponding to the event to be calculated.

[0117] Specifically, in this application, first for each object set, the set weight coefficient corresponding to the object set and the set attention corresponding to the object set for the event to be calculated can be multiplied to obtain the event heat value corresponding to the object set, and for multiple object sets, the multiple event heat values can be summed to obtain the total event heat value corresponding to the event to be calculated. At this time, the event heat result corresponding to the event to be calculated can be obtained according to the total event heat value corresponding to the event to be calculated. In this way, in this application, the event to be calculated can be calculated based on the set weight coefficient of the object set and the set attention for the event to be calculated, to obtain a more accurate event heat result, thereby avoiding the problem of deviation in event heat calculation caused by only determining the event heat of the event based on the interaction data of the event in the related art.

[0118] In an implementable embodiment, the total event heat value can be obtained through formula (9), and formula (9) is specifically as follows:

[0119]

[0120] Among them, E represents the total event heat value, UCB i Represents the set attention corresponding to the i-th object set for the event to be calculated. In this way, the total event heat value can comprehensively consider the importance and attention degree of different object sets, so as to more accurately calculate the true heat of the event to be calculated.

[0121] It should also be noted that before the present application performs the operation of obtaining the event heat result corresponding to the event to be calculated based on the total event heat value corresponding to the event to be calculated, a level prediction model can also be obtained. The level prediction model is used to predict the event level of an event, and the level prediction model is pre-constructed. Specifically, in the present application, the total event heat value corresponding to the event to be calculated can be predicted and processed according to the level prediction model to obtain the event heat level of the event to be calculated, where the event heat level includes S+ level, S level, A level, B level, and below B level. Thereafter, the event heat level of the event to be calculated is used as the event heat result corresponding to the event to be calculated. In this way, it is convenient to more accurately implement the recommendation or push of the event to be calculated in the subsequent process.

[0122] In an implementable embodiment, in the present application, the construction of the level prediction model can be achieved through formulas (10)-(12). The specific expressions of formulas (10)-(12) are as follows:

[0123]

[0124] Among them, D represents the training sample set, Q represents the number of the training sample set, represents the feature vector of the i-th event, represents the heat level label corresponding to the i-th event, represents the cross-entropy loss of the model (the goal of building the model is to minimize the cross-entropy loss of the model on the training sample set), θ represents the model parameters, and θ * represents the optimized model parameters, represents, for each sample i, the probability that the sample i belongs to y i It should be noted that the level prediction model in the present application can not only predict the event heat level corresponding to the event to be calculated based on the total event heat value, but also be continuously adjusted based on the feedback opinions of the object to continuously improve the model performance.

[0125] It should also be noted that in the present application, the event heat level corresponding to the event to be calculated can be determined by means of threshold determination. For example, if the total event heat value is 10,000, it can be determined that the event heat level corresponding to the event to be calculated is A level. In this way, by determining the event heat level of the event to be calculated, the present application can provide more accurate data support for personalized content recommendation and improve the accuracy of event recommendation and the experience of the object.

[0126] Further, it should be noted that after the event heat level of the event to be calculated is used as the event heat result corresponding to the event to be calculated, the event recommendation list can be obtained. The event recommendation list is used to place recommended events, and the object using the specific application can learn about the event to be calculated based on this event recommendation list. And the event recommendation order on the event recommendation list can also be sorted according to the event heat level. For example, events of S+ level are at the top of the event recommendation list, followed by events of S level and A level, and finally events of B level.

[0127] Specifically, if the event heat result indicates that the event heat level is the target level, the event to be calculated can be recommended to the event recommendation list; if the event heat result indicates that the event heat level is not the target level, an operation of increasing the exposure of the event to be calculated can be performed. The target level can be S+ level, and the target level can also be S level. There is no specific limitation on the target level here, and it can also be determined in actual applications. In addition, in this application, in addition to placing the event to be calculated on the event recommendation list, events with a high event heat level (such as S+ level) can also be pushed to multiple object sets or preferentially pushed to the object set of interest. In this way, this application can calculate the true heat of events in the public set more accurately, thereby improving the event recommendation effect.

[0128] In a feasible implementation manner, in this application, the heat prediction of events can also be realized based on a heat prediction model. The heat prediction model is constructed based on a historical sample event set and a historical heat result set corresponding to the historical sample event set. One historical sample event corresponds to one historical heat result, and the heat prediction model can continuously obtain changes in object interests in combination with an online learning mechanism and an adaptive adjustment mechanism, and timely adjust the model prediction strategy. In this way, the heat prediction of events can be realized quickly and conveniently.

[0129] As Figure 4 shown, Figure 4 is an application schematic diagram of recommended events in a method for calculating event heat provided by an embodiment of this application. Figure 4 An event recommendation list is shown therein. The event recommendation list can be a blog post list about society and culture and entertainment. At this time, based on the event heat level of the event to be calculated, the event to be calculated can be placed at the top of the event recommendation list for the public to understand current affairs more quickly.

[0130] As Figure 5 shown, Figure 5 is an application schematic diagram of recommended events in another method for calculating event heat provided by an embodiment of this application. Figure 5An event recommendation list is shown. The event recommendation list can be a video list for social and cultural and entertainment categories. In this application platform, the event recommendation list can be switched with the xxx list, can also be switched with the yyy list, and can also be switched with the zzz list. At this time, based on the event heat level of the event to be calculated, the event to be calculated can be placed at the top of the event recommendation list to facilitate the public to understand current events more quickly.

[0131] As Figure 6 shown, Figure 6 is the full flow chart of event heat calculation in a method for calculating event heat provided by an embodiment of the present application. In Figure 6 first, multiple objects to be grouped can be grouped to obtain multiple object sets. After that, based on the multiple object sets or the object interaction data sets respectively corresponding to the multiple object sets, the set weight coefficients respectively corresponding to the multiple object sets can be obtained, and based on the event to be calculated and the object interaction data sets respectively corresponding to the multiple object sets, the set attentions respectively corresponding to the multiple object sets can be obtained. Finally, by respectively calculating and processing the set weight coefficients respectively corresponding to the multiple object sets and the set attentions respectively corresponding to the multiple object sets, the event heat result corresponding to the event to be calculated can be obtained. In this way, in the present application, based on the activity degree of the object set and the attention of the object set to the event, the heat calculation of the event can be realized, avoiding the situation of deviation in the heat calculation of the event in the related art.

[0132] As Figure 7 shown, Figure 7 is the full flow chart of event processing in a method for calculating event heat provided by an embodiment of the present application. In Figure 7 first, the event heat level represented in the event heat result can be determined. If the event heat level of the event to be calculated is the target level, the event to be calculated can be recommended to the event recommendation list, or the event to be calculated can be pushed to the multiple object sets. In this way, the present application uses the calculated event heat result for event recommendation or push, and can achieve the real "breaking the circle" of the event, thereby improving the recommendation / push effect of the event.

[0133] In summary, in the embodiments of the present application, first, the set weight coefficient corresponding to each object set can be calculated based on multiple object sets or multiple object interaction data sets, and the set attention of each object set to the set to be calculated can be calculated based on multiple object interaction data sets and multiple interaction data subsets. Then, the set weight coefficients and set attentions respectively corresponding to multiple object sets are calculated and processed, and the event heat result corresponding to the event to be calculated can be obtained. In this way, in the present application, the activity of each object set and the set attention of each object set to the event to be calculated can be used to calculate the heat of the event to be calculated, thereby avoiding the deviation in the heat calculation of the event in the related art, and to a certain extent ensuring the accuracy of the calculated event heat result. Then, based on the event heat result of the event, recommendations or pushes for the event are made, which can achieve true "breaking the circle", thereby improving the recommendation / push effect of the event and further enhancing the experience of the object.

[0134] Based on the event heat calculation method provided in the foregoing embodiments, the present application also correspondingly provides an event heat calculation device. The event heat calculation device provided in the embodiments of the present application will be specifically introduced below.

[0135] See Figure 8 , which is a schematic structural diagram of an event heat calculation device provided in an embodiment of the present application. As Figure 8 shown, the event heat calculation device specifically includes:

[0136] An event set acquisition unit 801, configured to acquire an event to be calculated, multiple object sets, and the object interaction data sets respectively corresponding to the multiple object sets, where the interest of each object set in the event is different;

[0137] A weight coefficient acquisition unit 802, configured to obtain the set weight coefficients respectively corresponding to the multiple object sets according to the multiple object sets or multiple object interaction data sets, where each set weight coefficient represents the activity of each object set;

[0138] A data subset acquisition unit 803, configured to perform screening processing on the multiple object interaction data sets according to the event to be calculated, and obtain the interaction data subsets respectively corresponding to the event to be calculated in the multiple object interaction data sets;

[0139] A set attention acquisition unit 804, configured to perform calculation processing on the multiple object interaction data sets and multiple interaction data subsets, and obtain the set attentions respectively corresponding to the multiple object sets for the event to be calculated;

[0140] A heat result obtaining unit 805 is configured to perform calculation processing on the set weight coefficients corresponding to the multiple object sets and the set attentions corresponding to the multiple object sets for the event to be calculated, so as to obtain an event heat result corresponding to the event to be calculated.

[0141] In an implementable embodiment, the heat result obtaining unit 805 includes:

[0142] An event heat value obtaining unit is configured to, for each object set, perform multiplication processing on the set weight coefficient corresponding to the object set and the set attention corresponding to the object set for the event to be calculated, so as to obtain an event heat value corresponding to the object set;

[0143] A total heat value obtaining unit is configured to, for the multiple object sets, perform summation processing on the multiple event heat values, so as to obtain an event total heat value corresponding to the event to be calculated;

[0144] A total heat value processing unit is configured to obtain an event heat result corresponding to the event to be calculated according to the event total heat value corresponding to the event to be calculated.

[0145] In an implementable embodiment, the set attention obtaining unit 804 is specifically configured to:

[0146] For each object set, perform division processing on the total click data and the click data corresponding to the object set, so as to obtain an average click-through rate corresponding to the object set;

[0147] For each object set, perform division processing on the total like data and the like data corresponding to the object set, so as to obtain an average click-through rate corresponding to the object set;

[0148] For the multiple object sets, perform summation processing on the average click-through rates and the average like rates respectively corresponding to the multiple object sets, so as to obtain the set attentions respectively corresponding to the multiple object sets for the event to be calculated.

[0149] In an implementable embodiment, the set attention obtaining unit 804 is specifically configured to:

[0150] For each object set, perform division processing on the click data and the exposure data corresponding to the object set, so as to obtain an average click-through rate corresponding to the object set;

[0151] For each object set, perform confidence processing on the total exposure data and the average click-through rate corresponding to the object set, so as to obtain a maximum confidence value corresponding to the object set;

[0152] For the multiple object sets, the maximum confidence values respectively corresponding to the multiple object sets are used as the set attentions respectively corresponding to the multiple object sets for the event to be calculated.

[0153] In an implementable embodiment, the weight coefficient obtaining unit 802 is specifically configured to:

[0154] For each object set, perform a counting process on the object interaction data set corresponding to the object set to obtain the first total interaction quantity corresponding to the object set;

[0155] For the multiple object sets, perform a counting process on the object interaction data sets respectively corresponding to the multiple object sets to obtain the second total interaction quantity corresponding to the multiple object sets;

[0156] Perform a division process on the multiple first total interaction quantities and the second total interaction quantity respectively to obtain the set weight coefficients respectively corresponding to the multiple object sets.

[0157] In an implementable embodiment, the weight coefficient obtaining unit 802 is specifically configured to:

[0158] Perform a counting process on the multiple object sets to obtain the first total object quantities respectively corresponding to the multiple object sets and the second total object quantity corresponding to the multiple object sets;

[0159] Perform a division process on the multiple first total object quantities and the second total object quantity respectively to obtain the set weight coefficients respectively corresponding to the multiple object sets.

[0160] In an implementable embodiment, the weight coefficient obtaining unit 802 is specifically configured to:

[0161] Perform a screening process on the multiple object sets to obtain the specific object sets respectively corresponding to the multiple object sets, where the specific objects in the specific object sets include objects with the number of fans greater than a preset number in a specific application;

[0162] Perform a counting process on the multiple specific object sets to obtain the first specific total quantities respectively corresponding to the multiple specific object sets and the second specific total quantity corresponding to the multiple specific object sets;

[0163] Perform a division process on the multiple first specific total quantities and the second specific total quantity respectively to obtain the set weight coefficients respectively corresponding to the multiple object sets.

[0164] In an implementable embodiment, the device further includes:

[0165] A prediction model acquisition unit for acquiring a level prediction model for predicting the event level of an event.

[0166] The total heat value processing unit is specifically configured to:

[0167] Perform prediction processing on the total heat value of the event corresponding to the event to be calculated according to the level prediction model to obtain the event heat level of the event to be calculated.

[0168] Use the event heat level of the event to be calculated as the event heat result corresponding to the event to be calculated.

[0169] In an implementable embodiment, the device further includes:

[0170] An event recommendation list acquisition unit for acquiring an event recommendation list, where the event recommendation list is used to place recommended events.

[0171] A to-be-calculated event recommendation unit for recommending the to-be-calculated event to the event recommendation list if the event heat result indicates that the event heat level is a target level.

[0172] A to-be-calculated event push unit for pushing the to-be-calculated event to the multiple object sets if the event heat result indicates that the event heat level is a target level.

[0173] The event heat calculation device provided by the embodiments of the present application has the same beneficial effects as the event heat calculation method provided by the above embodiments, so details are not described herein again.

[0174] The embodiments of the present application provide a computer device, which may be a server. Figure 9 FIG. is a schematic structural diagram of a server provided by the embodiments of the present application. The server 900 may vary greatly due to configuration or performance differences and may include one or more central processing units (CPUs) 922 (for example, one or more processors) and a memory 932, and one or more storage media 930 (for example, one or more mass storage devices) for storing application programs 942 or data 944. Among them, the memory 932 and the storage media 930 may be transient storage or persistent storage. The program stored in the storage media 930 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the server. Further, the central processing unit 922 may be configured to communicate with the storage media 930 and execute a series of instruction operations in the storage media 930 on the server 900.

[0175] The server 900 may also include one or more power supplies 926, one or more wired or wireless network interfaces 950, one or more input / output interfaces 958, and / or one or more operating systems 941, such as Windows Server TM , Mac OS X TM , Unix TM , Linux TM , FreeBSD TM and so on.

[0176] Among them, the CPU 922 is used to execute the following steps:

[0177] Obtain the event to be calculated, multiple object sets, and the object interaction data sets respectively corresponding to the multiple object sets, where each object set has different interests in the event;

[0178] According to the multiple object sets or multiple object interaction data sets, obtain the set weight coefficients respectively corresponding to the multiple object sets, where each set weight coefficient characterizes the activity of each object set;

[0179] Perform a screening process on the multiple object interaction data sets according to the event to be calculated, and obtain the interaction data subsets respectively corresponding to the event to be calculated in the multiple object interaction data sets;

[0180] Perform a calculation process on the multiple object interaction data sets and multiple interaction data subsets, and obtain the set attentions respectively corresponding to the multiple object sets for the event to be calculated;

[0181] Perform a calculation process on the set weight coefficients respectively corresponding to the multiple object sets and the set attentions respectively corresponding to the multiple object sets for the event to be calculated, and obtain the event heat result corresponding to the event to be calculated.

[0182] The embodiment of the present application also provides another computer device, and this computer device may be a terminal device. As Figure 10 shown, for the convenience of description, only the parts related to the embodiment of the present application are shown. For the specific technical details not disclosed, please refer to the method part of the embodiment of the present application. Taking this terminal device as a mobile phone as an example:

[0183] Figure 10 What is shown is a block diagram of a part of the structure of the mobile phone provided by the embodiment of the present application. Refer to Figure 10, the mobile phone includes components such as a Radio Frequency (RF) circuit 1010, a memory 1020, an input unit 1030, a display unit 1040, a sensor 1050, an audio circuit 1060, a wireless fidelity (WiFi) module 1070, a processor 1080, and a power supply 1090. Those skilled in the art can understand that Figure 10 the mobile phone structure shown in

[0184] does not limit the mobile phone and may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements. Figure 10 The following specifically introduces each component of the mobile phone:

[0185] The RF circuit 1010 can be used for receiving and sending signals during information reception or call processes. Specifically, after receiving the downlink information from the base station, it is given to the processor 1080 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit 1010 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a Low Noise Amplifier (LNA), a duplexer, etc. In addition, the RF circuit 1010 can also communicate with the network and other devices through wireless communication. The above wireless communication can use any communication standard or protocol, including but not limited to the Global System of Mobile communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Messaging Service (SMS), etc.

[0186] The memory 1020 can be used to store software programs and modules. The processor 1080 executes various functional applications and data processing of the mobile phone by running the software programs and modules stored in the memory 1020. The memory 1020 may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory 1020 may include high-speed random access memory and may also include non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0187] The input unit 1030 can be used to receive input digital or character information and generate key signal inputs related to the user settings and function controls of the mobile phone. Specifically, the input unit 1030 may include a touch panel 1031 and other input devices 1032. The touch panel 1031, also known as a touch screen, can collect touch operations of the user on or near it (such as operations of the user using a finger, a stylus, or any suitable object or accessory on or near the touch panel 1031), and drive the corresponding connection device according to a pre-set program. Optionally, the touch panel 1031 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the touch position of the user, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it to the processor 1080, and can receive and execute the commands sent by the processor 1080. In addition, various types such as resistive, capacitive, infrared, and surface acoustic wave can be used to implement the touch panel 1031. In addition to the touch panel 1031, the input unit 1030 may further include other input devices 1032. Specifically, the other input devices 1032 may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control keys, power on / off keys, etc.), a trackball, a mouse, a joystick, etc.

[0188] The display unit 1040 can be used to display information input by the user or information provided to the user, as well as various menus of the mobile phone. The display unit 1040 may include a display panel 1041. Optionally, the display panel 1041 can be configured in the form of, for example, a liquid crystal display (LCD for short), an organic light-emitting diode (OLED for short), etc. Further, the touch panel 1031 can cover the display panel 1041. When the touch panel 1031 detects a touch operation on or near it, it is transmitted to the processor 1080 to determine the type of touch event. Subsequently, the processor 1080 provides a corresponding visual output on the display panel 1041 according to the type of touch event. Although in Figure 10 , the touch panel 1031 and the display panel 1041 are implemented as two independent components to realize the input and input functions of the mobile phone, but in some embodiments, the touch panel 1031 and the display panel 1041 can be integrated to realize the input and output functions of the mobile phone.

[0189] The mobile phone may further include at least one sensor 1050, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor. Among them, the ambient light sensor can adjust the brightness of the display panel 1041 according to the brightness of the ambient light, and the proximity sensor can turn off the display panel 1041 and / or the backlight when the mobile phone is moved to the ear. As a kind of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in all directions (generally three axes). When stationary, it can detect the magnitude and direction of gravity, and can be used for applications that identify the posture of the mobile phone (such as horizontal and vertical screen switching, related games, magnetometer attitude calibration), vibration recognition related functions (such as pedometer, tapping), etc.; as for other sensors such as gyroscopes, barometers, hygrometers, thermometers, and infrared sensors that the mobile phone can also be configured with, they will not be elaborated here.

[0190] The audio circuit 1060, the speaker 1061, and the microphone 1062 can provide an audio interface between the user and the mobile phone. The audio circuit 1060 can transmit the electrical signal converted from the received audio data to the speaker 1061, and the speaker 1061 converts it into a sound signal for output; on the other hand, the microphone 1062 converts the collected sound signal into an electrical signal, which is received by the audio circuit 1060 and then converted into audio data. After the audio data is output to the processor 1080 for processing, it is sent to another mobile phone, for example, via the RF circuit 1010, or the audio data is output to the memory 1020 for further processing.

[0191] WiFi belongs to short-range wireless transmission technology. Through the WiFi module 1070, a mobile phone can help users send and receive emails, browse the web, and access streaming media, etc. It provides users with wireless broadband Internet access. Although Figure 10 the WiFi module 1070 is shown, it can be understood that it does not belong to the essential components of the mobile phone and can be completely omitted within the scope of not changing the essence of the invention according to needs.

[0192] The processor 1080 is the control center of the mobile phone, connecting various parts of the entire mobile phone through various interfaces and circuits. By running or executing software programs and / or modules stored in the memory 1020, and by calling data stored in the memory 1020, it executes various functions of the mobile phone and processes data, thereby collecting overall data and information of the mobile phone. Optionally, the processor 1080 may include one or more processing units; preferably, the processor 1080 may integrate an application processor and a modem processor. Among them, the application processor mainly processes the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communications. It can be understood that the above-mentioned modem processor may not be integrated into the processor 1080 either.

[0193] The mobile phone also includes a power source 1090 (such as a battery) for powering each component. Preferably, the power source can be logically connected to the processor 1080 through a power management system, so as to realize functions such as management of charging, discharging, and power consumption management through the power management system.

[0194] Although not shown, the mobile phone may also include a camera, a Bluetooth module, etc., which will not be elaborated here.

[0195] In the embodiments of the present application, the processor 1080 included in the mobile phone further has the following functions:

[0196] Obtain the event to be calculated, multiple object sets, and the object interaction data sets respectively corresponding to the multiple object sets, where each object set has different interests in the event;

[0197] According to the multiple object sets or multiple object interaction data sets, obtain the set weight coefficients respectively corresponding to the multiple object sets, where each set weight coefficient represents the activity of each object set;

[0198] According to the event to be calculated, perform screening processing on the multiple object interaction data sets to obtain the interaction data subsets respectively corresponding to the event to be calculated in the multiple object interaction data sets;

[0199] Perform calculation processing on the multiple object interaction data sets and multiple interaction data subsets to obtain the set attentions respectively corresponding to the multiple object sets for the event to be calculated;

[0200] Calculate the set weight coefficients corresponding to the multiple object sets and the set attentions corresponding to the multiple object sets for the event to be calculated respectively, and obtain the event heat result corresponding to the event to be calculated.

[0201] The embodiment of the present application also provides a computer-readable storage medium for storing a computer program. When the computer program runs on a computer device, the computer device is enabled to execute any one of the implementation manners of the method for calculating event heat described in the foregoing various embodiments.

[0202] The embodiment of the present application also provides a computer program product including a computer program. When it runs on a computer device, the computer device is enabled to execute any one of the implementation manners of the method for calculating event heat described in the foregoing various embodiments.

[0203] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems and devices can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0204] In several embodiments provided by the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the system is only a logical function division. In actual implementation, there may be other division methods. For example, multiple systems can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0205] The system described as a separate component may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0206] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0207] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes: various media that can store computer programs, such as USB flash drives, mobile hard disks, read-only memories (English full name: Read-Only Memory, English abbreviation: ROM), random access memories (English full name: Random Access Memory, English abbreviation: RAM), magnetic disks, or optical discs.

[0208] In the embodiments of this application, the term "module" or "unit" refers to a computer program with a predetermined function or a part of a computer program, which works together with other related parts to achieve a predetermined goal, and can be fully or partially implemented by using software, hardware (such as a processing circuit or a memory), or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of the overall module or unit that includes the function of this module or unit.

[0209] The above embodiments are only used to illustrate the technical solution of this application, rather than to limit it; although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of this application.

Claims

1. A method for calculating event heat, characterized in that: include: Obtaining an event to be calculated, a plurality of object sets, and object interaction data sets corresponding to the plurality of object sets, wherein each object set has different interests in the event; According to the multiple object sets or the multiple object interaction data sets, obtaining set weight coefficients corresponding to the multiple object sets respectively, wherein each set weight coefficient represents the activity of each object set; Screening multiple object interaction data sets according to the event to be calculated, and obtaining interaction data subsets corresponding to the event to be calculated in the multiple object interaction data sets; Calculating and processing the plurality of object interaction data sets and the plurality of interaction data subsets to obtain the set attention degrees of the plurality of object sets respectively corresponding to the events to be calculated; The set weight coefficients respectively corresponding to the multiple object sets and the set attention degrees respectively corresponding to the multiple object sets for the events to be calculated are calculated to obtain event heat results corresponding to the events to be calculated.

2. The method according to claim 1, characterized in that The calculating and processing the set weight coefficients respectively corresponding to the multiple object sets and the set attention degrees respectively corresponding to the events to be calculated by the multiple object sets to obtain the event heat results corresponding to the events to be calculated, includes: For each object set, a set weight coefficient corresponding to the object set and a set attention degree of the object set corresponding to the event to be calculated are multiplied to obtain an event heat value corresponding to the object set; For the multiple object sets, multiple event heat values ​​are summed to obtain a total event heat value corresponding to the event to be calculated; According to the total event heat value corresponding to the event to be calculated, the event heat result corresponding to the event to be calculated is obtained.

3. The method according to claim 1, characterized in that The object interaction data set includes total click data and total like data, the interaction data subset includes click data and like data for the event to be calculated, and the calculation and processing of the multiple object interaction data sets and the multiple interaction data subsets to obtain the set attentions of the multiple object sets corresponding to the event to be calculated respectively includes: For each object set, the total click data and the click data corresponding to the object set are divided to obtain the average click rate corresponding to the object set; For each object set, the total likes data and the likes data corresponding to the object set are divided to obtain the average click rate corresponding to the object set; For the multiple object sets, average click rates and average like rates corresponding to the multiple object sets are summed up to obtain the set attentions of the multiple object sets corresponding to the events to be calculated.

4. The method according to claim 1, characterized in that The object interaction data set includes total exposure data, the interaction data subset includes click data and exposure data for the event to be calculated, and the calculation and processing of the multiple object interaction data sets and the multiple interaction data subsets to obtain the set attentions corresponding to the multiple object sets for the event to be calculated includes: For each object set, the click data and exposure data corresponding to the object set are divided to obtain the average click rate corresponding to the object set; For each object set, confidence processing is performed on the total exposure data and average click-through rate corresponding to the object set to obtain the maximum confidence value corresponding to the object set; For the multiple object sets, the maximum confidence values ​​respectively corresponding to the multiple object sets are used as the set attention degrees respectively corresponding to the multiple object sets for the events to be calculated.

5. The method according to claim 1, characterized in that According to the plurality of object interaction data sets, obtaining set weight coefficients corresponding to the plurality of object sets respectively includes: For each object set, counting processing is performed on the object interaction data set corresponding to the object set to obtain the total number of first interactions corresponding to the object set; For the multiple object sets, counting processing is performed on the object interaction data sets respectively corresponding to the multiple object sets to obtain the total number of second interactions corresponding to the multiple object sets; A plurality of first interaction total numbers and the second interaction total numbers are divided respectively to obtain set weight coefficients corresponding to the plurality of object sets respectively.

6. The method according to claim 1, characterized in that Obtaining, according to the multiple object sets, set weight coefficients respectively corresponding to the multiple object sets, including: Performing counting processing on the multiple object sets to obtain a total number of first objects corresponding to the multiple object sets and a total number of second objects corresponding to the multiple object sets; The total number of the plurality of first objects and the total number of the second objects are divided respectively to obtain set weight coefficients corresponding to the plurality of object sets respectively.

7. The method according to claim 1, characterized in that Obtaining, according to the multiple object sets, set weight coefficients respectively corresponding to the multiple object sets, including: Filter the multiple object sets to obtain specific object sets corresponding to the multiple object sets, wherein the specific objects in the specific object sets include objects in a specific application program whose number of fans is greater than a preset number; Counting the multiple specific object sets to obtain first specific total quantities corresponding to the multiple specific object sets and second specific total quantities corresponding to the multiple specific object sets; A plurality of first specific total quantities and the second specific total quantities are divided respectively to obtain set weight coefficients corresponding to the plurality of object sets respectively.

8. The method according to claim 2, characterized in that: Before obtaining the event heat result corresponding to the event to be calculated according to the total event heat value corresponding to the event to be calculated, the method further includes: Acquire a grade prediction model, wherein the grade prediction model is used to predict an event grade of an event; The step of obtaining the event heat result corresponding to the event to be calculated according to the total event heat value corresponding to the event to be calculated includes: Predicting the total event heat value corresponding to the event to be calculated according to the level prediction model to obtain the event heat level of the event to be calculated; The event heat level of the event to be calculated is used as the event heat result corresponding to the event to be calculated.

9. The method according to claim 8, characterized in that Also includes: Obtain an event recommendation list, wherein the event recommendation list is used to place recommended events; If the event heat result indicates that the event heat level is the target level, recommending the event to be calculated to the event recommendation list; Alternatively, if the event heat result indicates that the event heat level is a target level, the event to be calculated is pushed to the multiple object sets.

10. A device for calculating event heat, characterized in that: include: An event set acquisition unit, used to acquire an event to be calculated, a plurality of object sets, and object interaction data sets corresponding to the plurality of object sets, wherein each object set has different interests in the event; A weight coefficient obtaining unit, configured to obtain set weight coefficients corresponding to the plurality of object sets respectively according to the plurality of object sets or the plurality of object interaction data sets, wherein each set weight coefficient represents the activity of each object set; A data subset obtaining unit, configured to filter a plurality of object interaction data sets according to the event to be calculated, and obtain interaction data subsets corresponding to the event to be calculated in the plurality of object interaction data sets; A collection attention degree obtaining unit, configured to calculate and process the plurality of object interaction data sets and the plurality of interaction data subsets, and obtain collection attention degrees corresponding to the plurality of object sets for the events to be calculated; The heat result obtaining unit is used to calculate and process the set weight coefficients respectively corresponding to the multiple object sets and the set attention degrees respectively corresponding to the multiple object sets for the events to be calculated, so as to obtain the event heat result corresponding to the events to be calculated.

11. A computer device, characterized in that: The device comprises a processor and a memory: The memory is used to store a computer program and transmit the computer program to the processor; The processor is used to execute the steps of the method for calculating the event heat according to any one of claims 1 to 9 according to the instructions in the computer program.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and when the computer program is executed by a computer device, the steps of the method for calculating the event heat described in any one of claims 1 to 9 are implemented.

13. A computer program product, characterized in that It comprises a computer program, which, when executed by a computer device, implements the steps of the method for calculating the event heat as described in any one of claims 1 to 9.

Images