Method and device for determining activity communication effect, storage medium, and terminal

Abstract

The present invention discloses a method and device for determining the effect of activity communication, a storage medium, and a terminal, which relate to the field of media communication technology, and the main purpose is to solve the problem of low quantification accuracy of the effect of activity communication. It mainly includes obtaining the index parameters of the multi-dimensional activity communication index of the sample activity; determining a regression model with the activity communication index as the independent variable and the activity effect index as the dependent variable, and determining a neural network model with the activity communication index as the input and the activity effect index as the output, using the first weight coefficient of the regression model and the second weight coefficient of the neural network model to perform a fusion calculation to determine the target weight coefficient, and based on the target weight coefficient, the activity communication index and the activity effect index, generate an activity effect index formula; using the activity effect index formula, calculate the index parameters of the multi-dimensional activity communication index of the target activity to determine the target activity effect index value.

Description

Technical Field

[0001] The present invention relates to the field of media communication technology, and in particular to a method and device for determining activity communication effects, a storage medium, and a terminal. Background Art

[0002] In order to establish brand image, expand brand recognition and reputation, brands need to carry out brand communication. In brand communication, communication activities play a very important role. In the field of communication, the effect of activity communication and whether it has met expectations need to be measured by work results.

[0003] At present, the existing evaluation of the communication effect of activities is mainly measured by the amount of media releases in public channels or approximate traffic indicators. However, since the activities in the communication environment are no longer spread between single points, but in a chain-like manner, the amount of releases or traffic indicators can only indicate the brand's exposure rate, but cannot evaluate the degree of impact on the audience. Evaluating the effectiveness of communication activities based on the above methods cannot achieve a complete and comprehensive evaluation of communication activities, resulting in low accuracy in the quantitative evaluation of the effectiveness of communication activities. Summary of the invention

[0004] In view of this, the present invention provides a method and device for determining the effect of activity communication, a storage medium, and a terminal, the main purpose of which is to solve the problem of low quantification accuracy of existing communication activity effects.

[0005] According to one aspect of the present invention, a method for determining an activity communication effect is provided, comprising:

[0006] Obtain indicator parameters of multi-dimensional activity communication indicators of sample activities, wherein the activity communication indicators include multiple ones of the activity content release volume, activity communication volume, activity reach, activity interaction volume, activity communication health rate and key information quantity;

[0007] Determine a regression model with the activity communication index as an independent variable and the activity effect index as a dependent variable, calculate the activity effect index value corresponding to the indicator parameter through the regression model, and obtain a first weight coefficient of the regression model;

[0008] Determine a neural network model with the activity propagation index as input and the activity effect index as output, and train a second weight coefficient of the neural network model using the index parameter and the activity effect index value;

[0009] Performing a fusion calculation on the first weight coefficient and the second weight coefficient to determine a target weight coefficient, and generating an activity effect index formula based on the target weight coefficient, the activity propagation index and the activity effect index;

[0010] The activity effect index formula is used to calculate the indicator parameters of the multi-dimensional activity communication indicators of the target activity to determine the target activity effect index value.

[0011] Further, the step of fusing the first weight coefficient and the second weight coefficient to determine a target weight coefficient includes:

[0012] updating the first weight coefficient according to an average of the second weight coefficient and the first weight coefficient;

[0013] Recalculate the activity effect index value of the indicator parameter through the regression model with updated weight coefficient;

[0014] The second weight coefficient corresponding to the neural network model is trained according to the indicator parameters and the recalculated activity effect index value until the second weight coefficient or the activity effect index value meets the convergence condition, and the target weight coefficient is determined based on the first weight coefficient of the final regression model and the second weight coefficient of the final neural network model.

[0015] Furthermore, before calculating the index parameters of the multi-dimensional activity communication index of the target activity using the activity effect index formula and determining the target activity effect index value, the method further includes:

[0016] Obtain the activity content release volume, dissemination volume and multiple dissemination carriers of the target activity;

[0017] Calculating initial indicator parameters of at least one dimension of activity propagation indicator according to the propagation sound volume and the content of the propagation carrier;

[0018] The initial indicator parameters and the release volume of the activity content are standardized to obtain indicator parameters of the multi-dimensional activity communication indicators of the target activity.

[0019] Furthermore, the calculation of the initial index parameter of at least one dimension of the activity propagation index based on the propagation sound volume and the content of the propagation carrier includes:

[0020] Using an activity reach calculation model to predict the propagation volume and the content of the propagation carrier to obtain the activity reach; and / or,

[0021] Using an emotion recognition model to identify the content of the communication carrier, determine a first number of positive emotion category communication carriers and a second number of negative emotion category communication carriers, and calculate the activity communication health rate based on the first number and the second number; and / or,

[0022] The target keyword matching the preset keyword of the target activity is searched from the communication carrier, and the quantity of key information is determined according to the occurrence frequency of the target keyword.

[0023] Furthermore, before obtaining the data of the activity content release volume, the dissemination volume and multiple dissemination carriers of the target activity, the method further includes:

[0024] Obtaining a target activity communication subject of the target activity, and identifying a globally associated account associated with the target activity communication subject from a pre-built full-platform activity communication subject account knowledge base;

[0025] Target promotional carriers matching the theme keywords of the target activity are extracted from the promotional carriers released by the globally associated account, and the release quantity is determined based on the number of the target promotional carriers.

[0026] Furthermore, the activity communication index also includes the activity interaction volume. After extracting the target promotional carrier matching the theme keyword of the target activity from the promotional carriers released by the global associated account, the method further includes:

[0027] Extracting a plurality of initial keywords from the target promotion carrier;

[0028] Using the initial keyword and the initial logical operation relationship to verify the search result, if the result of the search result verification is passed, determining the initial keyword as a preset keyword, and determining the initial logical operation relationship as a preset logical operation relationship;

[0029] Collecting the content and interactive data of the communication carrier according to the preset logical operation relationship and the preset keywords;

[0030] The activity interaction volume is obtained by calculating the global interaction data of the propagation carriers, and the activity propagation volume is obtained by calculating the total number of the propagation carriers.

[0031] Furthermore, after calculating the index parameters of the multi-dimensional activity communication index of the target activity using the activity effect index formula and determining the target activity effect index value, the method further includes:

[0032] Obtaining the resource input amount of the target activity;

[0033] Calculating the resource input-output ratio of the target activity according to the resource input amount and the target activity effect index value;

[0034] An evaluation report of the target activity is generated based on the resource input-output ratio, the indicator parameters of the multi-dimensional activity communication indicator and the target weight coefficient.

[0035] According to another aspect of the present invention, there is provided a device for determining an activity propagation effect, comprising:

[0036] An acquisition module, used to acquire indicator parameters of multi-dimensional activity communication indicators of sample activities, wherein the activity communication indicators include multiple ones of the activity content release volume, activity communication volume, activity reach, activity interaction volume, activity communication health rate and key information quantity;

[0037] A calculation module, used to determine a regression model with the activity propagation index as an independent variable and the activity effect index as a dependent variable, calculate the activity effect index value corresponding to the indicator parameter through the regression model, and obtain a first weight coefficient of the regression model;

[0038] A training module, used to determine a neural network model with the activity propagation index as input and the activity effect index as output, and train a second weight coefficient of the neural network model using the index parameter and the activity effect index value;

[0039] A generating module, configured to perform a fusion calculation on the first weight coefficient and the second weight coefficient to determine a target weight coefficient, and to generate an activity effect index formula based on the target weight coefficient, the activity propagation index and the activity effect index;

[0040] The determination module is used to calculate the indicator parameters of the multi-dimensional activity communication indicator of the target activity by using the activity effect index formula to determine the target activity effect index value.

[0041] Furthermore, the generation module includes:

[0042] a first updating unit, configured to update the first weight coefficient according to an average of the second weight coefficient and the first weight coefficient;

[0043] A first calculation unit recalculates the activity effect index value of the indicator parameter through the regression model with the weight coefficient updated;

[0044] A training unit trains the second weight coefficient corresponding to the neural network model without relying on the indicator parameter and the recalculated activity effect index value until the second weight coefficient or the activity effect index value meets the convergence condition, and then determines the target weight coefficient based on the first weight coefficient of the final regression model and the second weight coefficient of the final neural network model.

[0045] Furthermore, the device also includes:

[0046] The acquisition module is further used to acquire the activity content publishing volume, dissemination volume and multiple dissemination carriers of the target activity;

[0047] The calculation module is further used to calculate the initial index parameter of at least one dimension of activity propagation index according to the propagation sound volume and the content of the propagation carrier;

[0048] The standardization processing module is used to perform standardization processing on the initial indicator parameters and the release volume of the activity content to obtain the indicator parameters of the multi-dimensional activity communication indicators of the target activity.

[0049] Furthermore, the calculation module includes:

[0050] A second calculation unit is used to predict and process the propagation sound volume and the content of the propagation carrier by using an activity reach calculation model to obtain an activity reach; and / or,

[0051] an identification unit, configured to identify the content of the communication carrier by using an emotion recognition model, determine a first number of communication carriers in a positive emotion category and a second number of communication carriers in a negative emotion category, and calculate an activity communication health rate based on the first number and the second number; and / or,

[0052] The determination unit is used to search the target keyword matching the preset keyword of the target activity from the communication carrier, and determine the quantity of key information according to the occurrence frequency of the target keyword.

[0053] Furthermore, the device also includes:

[0054] The acquisition module is further used to acquire the target activity propagation subject of the target activity, and identify the global associated account associated with the target activity propagation subject from a pre-built full-platform activity propagation subject account knowledge base;

[0055] The first extraction module is used to extract target promotional carriers that match the theme keywords of the target activity from the promotional carriers released by the global associated account, and determine the release quantity based on the number of the target promotional carriers.

[0056] Furthermore, the device also includes:

[0057] A second extraction module is used to extract a plurality of initial keywords from the target promotion carrier;

[0058] A verification module, used to verify the search results using the initial keyword and the initial logical operation relationship, and if the result of the search result verification is passed, the initial keyword is determined as a preset keyword, and the initial logical operation relationship is determined as a preset logical operation relationship;

[0059] A collection module, used to collect the content and interactive data of the communication carrier according to the preset logical operation relationship and the preset keywords;

[0060] The calculation module is also used to obtain the activity interaction volume by calculating the interaction data of the global communication carriers, and to obtain the activity communication volume by calculating the total number of the communication carriers.

[0061] Furthermore, the device also includes:

[0062] The acquisition module is further used to acquire the resource input amount of the target activity;

[0063] The calculation module is further used to calculate the resource input-output ratio of the target activity according to the resource input amount and the target activity effect index value;

[0064] A report generation module is used to generate an evaluation report for the target activity based on the resource input-output ratio, the indicator parameters of the multi-dimensional activity communication indicators and the target weight coefficient.

[0065] According to another aspect of the present invention, a storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the above-mentioned method for determining the activity propagation effect.

[0066] According to another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0067] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned method for determining the activity propagation effect.

[0068] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0069] The present invention provides a method and device for determining the effect of activity communication, a storage medium, and a terminal. An embodiment of the present invention obtains the indicator parameters of multi-dimensional activity communication indicators of sample activities, wherein the activity communication indicators include multiple types of activity content release volume, activity communication volume, activity reach, activity interaction volume, activity communication health rate, and key information quantity; determines a regression model with the activity communication indicator as an independent variable and an activity effect index as a dependent variable, calculates the activity effect index value corresponding to the indicator parameter through the regression model, and obtains the first weight coefficient of the regression model; determines a neural network model with the activity communication indicator as input and the activity effect index as output, and trains the second weight coefficient of the neural network model using the indicator parameter and the activity effect index value; performs a fusion calculation on the first weight coefficient and the second weight coefficient to determine the target weight coefficient, and generates an activity effect index formula based on the target weight coefficient, the activity communication indicator, and the activity effect index; uses the activity effect index formula to calculate the indicator parameters of the multi-dimensional activity communication indicator of the target activity to determine the target activity effect index value. It greatly increases the evaluation dimensions of activity effects and, at the same time, ensures the accuracy of the impact of each evaluation dimension on the effect, thereby greatly improving the accuracy of activity effect evaluation.

[0070] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0072] Figure 1 A flow chart of a method for determining an activity propagation effect provided by an embodiment of the present invention is shown;

[0073] Figure 2 A flow chart of another method for determining activity propagation effect provided by an embodiment of the present invention is shown;

[0074] Figure 3 A block diagram showing a device for determining an activity propagation effect provided by an embodiment of the present invention;

[0075] Figure 4 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0076] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0077] In order to solve the problem of low quantification accuracy of the effect of communication activities, the embodiment of the present invention provides a method for determining the effect of activity communication, such as Figure 1 As shown, the method includes:

[0078] 101. Obtain indicator parameters of multi-dimensional activity propagation indicators of sample activities.

[0079] In the embodiment of the present invention, the activity communication index includes multiple ones of the activity content release volume, activity communication volume, activity reach, activity interaction volume, activity communication health rate and key information quantity. The sample activities are historical communication activities with the same brand as the target activity or the same or similar market segment as the brand. The target activity is the activity for which the communication effect needs to be evaluated.

[0080] 102. Determine a regression model with the activity communication index as the independent variable and the activity effect index as the dependent variable, calculate the activity effect index value corresponding to the indicator parameter through the regression model, and obtain the first weight coefficient of the regression model.

[0081] In the embodiment of the present invention, the regression model is constructed with the number of dimensions of the activity communication index as the number of characteristic variables, wherein the characteristic weight of each characteristic variable is the corresponding weight coefficient. The preset initial weight coefficient is input into the regression model as the initial value to obtain the activity effect index value corresponding to the indicator parameter, and the first weight coefficient is the characteristic weight in the regression model solution process. Among them, the initial weight coefficient can be determined based on the empirical value, or it can be determined based on the weight coefficient of the historical activities of other brands, which is not specifically limited in the embodiment of the present invention.

[0082] 103. Determine a neural network model with the activity propagation index as input and the activity effect index as output, and use the index parameter and the activity effect index value to train a second weight coefficient of the neural network model.

[0083] In an embodiment of the present invention, the neural network model includes network layers corresponding to activity propagation indicators of each dimension, and an output layer for outputting activity effect index prediction results. By extracting the connection weight matrix between each network layer, the weight coefficients of activity propagation indicators of different dimensions, that is, the second weight coefficient, can be obtained.

[0084] 104. Perform a fusion calculation on the first weight coefficient and the second weight coefficient to determine a target weight coefficient, and generate an activity effect index formula based on the target weight coefficient, the activity propagation index and the activity effect index.

[0085] In the embodiment of the present invention, the activity propagation indicators of each dimension are all key factors affecting the activity propagation effect, but for different activities, the degree of influence of each factor is different, that is, the weight is different. Therefore, in order to make an accurate assessment of the activity effect based on the activity propagation indicators of different dimensions, it is necessary to determine the weight coefficients of the activity propagation indicators of each dimension. Since the analytic hierarchy process is a ranking result obtained by comparing layers, it cannot fully reflect the specific order of magnitude difference, and the principal component analysis uses the contribution rate to determine the weight after extracting the principal component, which will reduce the information of the feature. In order to more accurately determine the feature weights of the activity propagation indicators of each dimension, we use a combination of neural network model and regression model, and train the two models synchronously, and jointly determine the target weight coefficient based on the weights of each round of training of the two models.

[0086] 105. Using the activity effect index formula, calculate the indicator parameters of the multi-dimensional activity communication index of the target activity to determine the target activity effect index value.

[0087] In the embodiment of the present invention, the target activity effect index value is used to characterize the activity propagation effect of the target activity. After determining the target weight coefficient, calculate each activity propagation index x i And the corresponding target weight coefficient ω i The sum of the products of and is used to obtain the target activity effect index value y of the target activity. The calculation formula is:

[0088] y=ω 1 x 1 +ω 2 x 2 +…+ω i x i (1);

[0089] Where: x 1 、x 2 …x i ∈(0,100), i is the number of dimensions of the activity effect parameter, y∈(0,100).

[0090] It should be noted that by calculating the target activity effect index value of the target activity based on activity propagation indicators in multiple dimensions, a complete evaluation of the target activity effect can be achieved. At the same time, based on the joint training of neural network models and regression models, the weight coefficients of activity propagation indicators in each dimension are determined, which can accurately capture the impact of key factors in each dimension on the activity effect. This not only improves the accuracy of activity effect evaluation, but also provides a basis for adjustment factors for subsequent activity strategy adjustments and improvements and upgrades.

[0091] In an embodiment of the present invention, for further explanation and limitation, as Figure 2 As shown, step 104 performs a fusion calculation on the first weight coefficient and the second weight coefficient to determine a target weight coefficient, including:

[0092] 201. Update the first weight coefficient according to an average of the second weight coefficient and the first weight coefficient.

[0093] 202. Recalculate the activity effect index value of the indicator parameter through the regression model with updated weight coefficient.

[0094] 203. Train the second weight coefficient corresponding to the neural network model according to the indicator parameter and the recalculated activity effect index value until the second weight coefficient or the activity effect index value meets the convergence condition, and determine the target weight coefficient based on the first weight coefficient of the final regression model and the second weight coefficient of the final neural network model.

[0095] In the embodiment of the present invention, the joint training process of the neural network model and the regression model specifically includes the following steps: 1) taking the activity propagation index of each dimension as the independent variable and the initial activity effect index value as the dependent variable, the least square method is used to optimize the feature weight of the regression model to obtain a first weight coefficient; 2) taking the multi-dimensional activity propagation index as the input of the neural network model and the initial activity effect index value as the output value of the neural network model for training to obtain a second weight coefficient; 3) obtaining a connection weight matrix from the initial neural network model, and using the cross entropy of the connection weight matrix and the second weight coefficient as the first loss function, obtaining the feature weight from the regression model as the first weight coefficient. number, based on the cross entropy of the first weight coefficient and the initial feature weight as the second loss function; 4) based on the connection weight matrix and the feature weight calculation to obtain the activity effect index, and based on the cross entropy of the activity effect index and the initial activity effect index as the third damage function; 5) determine whether the first loss function, the second loss function and the third damage function all meet the corresponding convergence conditions. If they meet, stop training. If they do not meet, update the initial connection weight matrix based on the connection weight matrix obtained in this round, update the initial feature weight based on the feature weight obtained in this round, and update the initial activity effect index based on the activity effect index obtained in this round, and repeat 1) to 4). Among them, the convergence condition is that the loss function results of two adjacent rounds of iterations are less than a preset threshold, such as 5%, and can also be customized according to actual application requirements. The embodiment of the present invention does not make specific limitations. After determining the final neural network model and the final regression model, the first weight coefficient or the second weight coefficient can be used as the target weight coefficient, or the average of the first weight coefficient and the second weight coefficient can be used as the target weight coefficient. The embodiment of the present invention does not make specific limitations.

[0096] In one embodiment of the present invention, for further explanation and limitation, the step uses the activity effect index formula to calculate the indicator parameters of the multi-dimensional activity communication index of the target activity, and before determining the target activity effect index value, the method further includes:

[0097] Obtain the activity content release volume, dissemination volume and multiple dissemination carriers of the target activity;

[0098] Calculating initial indicator parameters of at least one dimension of activity propagation indicator according to the propagation sound volume and the content of the propagation carrier;

[0099] The initial indicator parameters and the release volume of the activity content are standardized to obtain indicator parameters of the multi-dimensional activity communication indicators of the target activity.

[0100] In the embodiment of the present invention, the amount of activity content released is the number of publicity carriers released through cooperative media in public channels to enhance brand awareness and influence. The publicity carriers may be short videos, articles, and soft-text advertisements, etc. The communication data includes the communication volume and the data of each communication carrier. The communication carrier includes both the publicity carrier and the works derived from the publicity carrier and the target activity, such as evaluation articles and report articles on the target activity by non-cooperative media or experts, and forwarding content of the publicity carrier, etc. The communication volume is the number of all communication carriers related to the target activity.

[0101] In an embodiment of the present invention, the activity communication index may include an index for evaluating the activity communication capability from one dimension, and may also include an index for evaluating the activity communication capability from multiple dimensions. For example, the activity communication index may include one of the activity communication volume, activity reach, and activity interaction volume, or may include any two combinations or all of them. The dimensions of the activity communication index can be customized according to the specific application scenario, and the embodiment of the present invention does not make specific limitations. The release volume is a parameter for evaluating the activity content release capability from the dimension of the number of promotional works released. By combining the activity communication indicators of multiple dimensions of release volume, it is possible to achieve a comprehensive evaluation of the target activity, greatly increase the comprehensiveness and accuracy of the evaluation, and thus improve the accuracy of the evaluation of the activity effect.

[0102] It should be noted that, since different activity communication indicators and activity content releases represent different evaluation dimensions, there are also large differences in the data magnitude. For example, the activity content release magnitude ranges from tens to hundreds of articles, the activity communication magnitude ranges from hundreds to thousands of articles, the reach ranges from millions to tens of millions, and the interaction ranges from thousands to millions. The activity communication health rate is a percentage between 0 and 100, and the activity effect cannot be directly evaluated based on these parameters. Therefore, it is necessary to standardize the activity communication indicators and releases. Specifically, normal quantile standardization is adopted to map each parameter to the corresponding parameter normal distribution, perform standardization, and obtain the corresponding standardized result, that is, the corresponding activity communication indicator. Among them, the parameter normal distribution is constructed based on the parameters corresponding to each historical activity within a preset time period. For example, the release volume data column is constructed based on the release volume data of all activities within 3 months, and the parameter normal distribution of the release volume data column is calculated. The reach volume data column is constructed based on the reach of all activities within 6 months, and the parameter normal distribution of the reach volume data column is calculated. In addition, for parameters with large differences in magnitude in different categories of activities, subcategories can be divided according to different forms of activities (such as online activities, offline activities), areas of activity involvement (such as beauty, automobiles, and technology), etc., and parameter normal distributions can be constructed for each subcategory to achieve fine standardization, thereby improving the accuracy of standardization. The following uses the release volume as an example to illustrate the calculation process of standardization. The mean and standard deviation of the release volume series are extracted from the release volume parameter normal distribution corresponding to the target activity category, and the standard score of each data value in the release volume series is calculated according to the following formula:

[0103]

[0104] After calculating the standard score of each data value, the standardized result of the target standard score of the release volume of the current target activity is calculated according to the following formula, that is, the release volume activity communication index:

[0105]

[0106] By mapping parameters of different magnitudes to the corresponding pre-constructed normal distribution, it can handle data of arbitrary distribution without being affected by skewness, and can achieve accurate standardization of each parameter.

[0107] In an embodiment of the present invention, in one embodiment of the present invention, for further explanation and limitation, the step of calculating the initial indicator parameter of at least one dimension of the activity propagation indicator according to the propagation sound volume and the content of the propagation carrier includes:

[0108] Using an activity reach calculation model to predict the propagation volume and the content of the propagation carrier to obtain the activity reach; and / or,

[0109] Using an emotion recognition model to identify the content of the communication carrier, determine a first number of positive emotion category communication carriers and a second number of negative emotion category communication carriers, and calculate the activity communication health rate based on the first number and the second number; and / or,

[0110] The target keyword matching the preset keyword of the target activity is searched from the communication carrier, and the quantity of key information is determined according to the occurrence frequency of the target keyword.

[0111] In an embodiment of the present invention, the activity communication index includes at least one of the activity reach, the activity communication health rate and the number of key information. Among them, the activity reach can be determined by a reach calculation model, specifically based on the reach calculation model for each communication carrier, and the reach of the global communication carrier is summed to obtain the activity reach of the target activity. Among them, the reach calculation model can be a model used in any existing reach acquisition method, and the embodiment of the present invention does not make specific limitations. The activity communication health rate can be determined by performing emotion recognition on the communication carriers one by one through an emotion recognition model constructed based on a natural language processing (NLP: Natural Language Processing) model, and determining whether the emotion type of each communication carrier is positive, neutral or negative. Then count the number of communication carriers with positive and neutral emotion types, and take (positive number of articles + neutral number of articles) / total number of all communication carriers * 100% as the activity communication health rate. Among them, for video-type communication carriers, the emotion type recognition can be performed on the text recognition results of the video. The number of key information can be determined based on the key information given in the communication plan, by determining several preset keywords, such as technology, fashion, etc., and then counting the total frequency of all preset keywords in all articles of the global communication carrier, that is, the volume of activity communication, that is, the number of key information.

[0112] In one embodiment of the present invention, for further explanation and limitation, before the step of obtaining the data of the activity content release volume, the propagation volume and multiple propagation carriers of the target activity, the method further includes:

[0113] Obtaining a target activity communication subject of the target activity, and identifying a globally associated account associated with the target activity communication subject from a pre-built full-platform activity communication subject account knowledge base;

[0114] Target promotional carriers matching the theme keywords of the target activity are extracted from the promotional carriers released by the globally associated account, and the release quantity is determined based on the number of the target promotional carriers.

[0115] In an embodiment of the present invention, the target activity communication subject is the media invited by the brand to promote its brand, which can be obtained from the list of invited media. Since the account involved in each media is not limited to one, for example, one of the target activity communication subjects is China Business News, and the promotional carrier can be released through multiple channels such as newspapers, China Business News websites, various social media platforms, and accounts of self-media platforms. Therefore, in order to ensure the comprehensiveness of the collection of the release volume, the content released by the associated accounts of each target activity communication subject in the entire platform is searched to collect all the promotional carriers, and the number of promotional carriers is calculated to obtain the release volume. Specifically, a knowledge base of the account of the activity communication subject of the entire platform can be pre-constructed. This knowledge base includes the mapping relationship between the global activity communication subject and its global associated account in the entire platform, so that after the target activity communication subject is determined, the associated account is comprehensively searched from this knowledge base. Through the construction of the knowledge base of the account of the activity communication subject of the entire platform, the manual verification time is reduced and the efficiency is greatly improved.

[0116] In one embodiment of the present invention, for further explanation and limitation, after extracting the target promotional carrier matching the theme keyword of the target activity from the promotional carriers published by the global associated account, the method further includes:

[0117] Extracting a plurality of initial keywords from the target promotion carrier;

[0118] Using the initial keyword and the initial logical operation relationship to verify the search result, if the result of the search result verification is passed, determining the initial keyword as a preset keyword, and determining the initial logical operation relationship as a preset logical operation relationship;

[0119] Collecting the content and interactive data of the communication carrier according to the preset logical operation relationship and the preset keywords;

[0120] The activity interaction volume is obtained by calculating the global interaction data of the propagation carriers, and the activity propagation volume is obtained by calculating the total number of the propagation carriers.

[0121] In the embodiment of the present invention, the initial keywords are determined by analyzing the promotional carrier (activity manuscript), and may include brand names, activity names, and other activity feature words. The initial logical operation relationship includes "and", "or", "not", "title hit", "text hit", "hit frequency", and "multi-word distance". The search results are verified for the initial keywords and the initial logical operation relationship. Specifically, the initial keywords and the initial logical operation relationship can be input into the search box of the mainstream search engine to confirm whether the returned article is a content that is strongly related to the activity subject keyword. If the correlation is strong, the verification result is determined to be passed. If the correlation is weak, the initial keywords and / or the initial logical operation relationship are adjusted and re-verified until the verification is passed. The propagation carrier data is collected based on the final preset keywords and preset logical operation relationship. The number of global propagation carriers is used as the propagation volume. By searching based on preset keywords and preset logical operation relationships, the search hit rate can be greatly improved, thereby improving the collection efficiency of propagation carriers.

[0122] It should be noted that the communication carrier data includes interaction data, and the interaction value is calculated for the interaction data of each communication carrier, such as the number of reposts, the number of comments, and the number of likes. Among them, the interaction value can be the sum of the number of reposts, the number of comments, and the number of likes. The interaction value of the global communication carrier is added up to get the interaction volume. Therefore, the activity communication index can also include the interaction volume.

[0123] In one embodiment of the present invention, for further explanation and limitation, the step uses the activity effect index formula to calculate the indicator parameters of the multi-dimensional activity communication index of the target activity, and after determining the target activity effect index value, the method further includes:

[0124] Obtaining the resource input amount of the target activity;

[0125] Calculating the resource input-output ratio of the target activity according to the resource input amount and the target activity effect index value;

[0126] An evaluation report of the target activity is generated based on the resource input-output ratio, the indicator parameters of the multi-dimensional activity communication indicator and the target weight coefficient.

[0127] In the embodiment of the present invention, the resource input-output ratio is the input-output ratio of the target activity, which can show the effect of the target activity in the revenue dimension. The target weight coefficient can show the criticality of the activity communication indicators of each dimension, and can provide a reference for the planning of the activity. The product of the activity communication indicators of each dimension and the corresponding target weight coefficient can show the contribution of each dimension to the activity effect, and can also provide an adjustment basis for the adjustment focus of subsequent activities. Generating an evaluation report for the target activity based on the above content can achieve a comprehensive and specific evaluation of the target activity.

[0128] The present invention provides a method for determining the effect of activity communication. The embodiment of the present invention obtains the index parameters of the multi-dimensional activity communication index of the sample activity, wherein the activity communication index includes multiple of the activity content release volume, activity communication volume, activity reach, activity interaction volume, activity communication health rate and key information quantity; determines a regression model with the activity communication index as an independent variable and the activity effect index as a dependent variable, calculates the activity effect index value corresponding to the index parameter through the regression model, and obtains the first weight coefficient of the regression model; determines a neural network model with the activity communication index as input and the activity effect index as output, and trains the second weight coefficient of the neural network model using the index parameter and the activity effect index value; performs a fusion calculation on the first weight coefficient and the second weight coefficient to determine the target weight coefficient, and generates an activity effect index formula based on the target weight coefficient, the activity communication index and the activity effect index; uses the activity effect index formula to calculate the index parameters of the multi-dimensional activity communication index of the target activity and determine the target activity effect index value. The evaluation dimension of the activity effect is greatly increased, and at the same time, the accuracy of the influence of each evaluation dimension on the effect is ensured, thereby greatly improving the accuracy of the activity effect evaluation.

[0129] Furthermore, as a response to the above Figure 1 The implementation of the method shown in the figure, the embodiment of the present invention provides a device for determining the effect of activity propagation, such as Figure 3 As shown, the device comprises:

[0130] An acquisition module 31 is used to acquire indicator parameters of multi-dimensional activity communication indicators of sample activities, wherein the activity communication indicators include multiple ones of the activity content release volume, activity communication volume, activity reach, activity interaction volume, activity communication health rate and key information quantity;

[0131] A calculation module 32, used to determine a regression model with the activity propagation index as an independent variable and the activity effect index as a dependent variable, calculate the activity effect index value corresponding to the indicator parameter through the regression model, and obtain a first weight coefficient of the regression model;

[0132] A training module 33, used to determine a neural network model with the activity propagation index as input and the activity effect index as output, and train a second weight coefficient of the neural network model using the index parameter and the activity effect index value;

[0133] A generating module 34 is used to perform a fusion calculation on the first weight coefficient and the second weight coefficient to determine a target weight coefficient, and to generate an activity effect index formula based on the target weight coefficient, the activity propagation index and the activity effect index;

[0134] The determination module 35 is used to calculate the indicator parameters of the multi-dimensional activity communication indicator of the target activity by using the activity effect index formula to determine the target activity effect index value.

[0135] Furthermore, the generating module 34 includes:

[0136] a first updating unit, configured to update the first weight coefficient according to an average of the second weight coefficient and the first weight coefficient;

[0137] A first calculation unit recalculates the activity effect index value of the indicator parameter through the regression model with the updated weight coefficient;

[0138] A training unit trains the second weight coefficient corresponding to the neural network model without relying on the indicator parameter and the recalculated activity effect index value until the second weight coefficient or the activity effect index value meets the convergence condition, and then determines the target weight coefficient based on the first weight coefficient of the final regression model and the second weight coefficient of the final neural network model.

[0139] Furthermore, the device also includes:

[0140] The acquisition module 31 is further used to acquire the activity content release volume, dissemination volume and multiple dissemination carriers of the target activity;

[0141] The calculation module 32 is further used to calculate the initial index parameter of at least one dimension of activity propagation index according to the propagation sound volume and the content of the propagation carrier;

[0142] The standardization processing module is used to perform standardization processing on the initial indicator parameters and the release volume of the activity content to obtain the indicator parameters of the multi-dimensional activity communication indicators of the target activity.

[0143] Furthermore, the calculation module 32 includes:

[0144] A second calculation unit is used to predict and process the propagation sound volume and the content of the propagation carrier by using an activity reach calculation model to obtain an activity reach; and / or,

[0145] an identification unit, configured to identify the content of the communication carrier by using an emotion recognition model, determine a first number of communication carriers in a positive emotion category and a second number of communication carriers in a negative emotion category, and calculate an activity communication health rate based on the first number and the second number; and / or,

[0146] The determination unit is used to search the target keyword matching the preset keyword of the target activity from the communication carrier, and determine the quantity of key information according to the occurrence frequency of the target keyword.

[0147] Furthermore, the device also includes:

[0148] The acquisition module 31 is further used to acquire the target activity propagation subject of the target activity, and identify the global associated account associated with the target activity propagation subject from a pre-built full-platform activity propagation subject account knowledge base;

[0149] The first extraction module is used to extract target promotional carriers that match the theme keywords of the target activity from the promotional carriers released by the global associated account, and determine the release quantity based on the number of the target promotional carriers.

[0150] Furthermore, the device also includes:

[0151] A second extraction module is used to extract a plurality of initial keywords from the target promotion carrier;

[0152] A verification module, used to verify the search results using the initial keyword and the initial logical operation relationship, and if the result of the search result verification is passed, the initial keyword is determined as a preset keyword, and the initial logical operation relationship is determined as a preset logical operation relationship;

[0153] A collection module, used to collect the content and interactive data of the communication carrier according to the preset logical operation relationship and the preset keywords;

[0154] The calculation module 32 is further used to obtain the activity interaction volume by calculating the interaction data of the global communication carriers, and to obtain the activity communication volume by calculating the total number of the communication carriers.

[0155] Furthermore, the device also includes:

[0156] The acquisition module 31 is further used to acquire the resource input amount of the target activity;

[0157] The calculation module 32 is further used to calculate the resource input-output ratio of the target activity according to the resource input amount and the target activity effect index value;

[0158] A report generation module is used to generate an evaluation report for the target activity based on the resource input-output ratio, the indicator parameters of the multi-dimensional activity communication indicators and the target weight coefficient.

[0159] The present invention provides a device for determining the effect of activity communication. The embodiment of the present invention obtains the index parameters of the multi-dimensional activity communication index of the sample activity, wherein the activity communication index includes multiple of the activity content release volume, activity communication volume, activity reach, activity interaction volume, activity communication health rate and key information quantity; determines a regression model with the activity communication index as an independent variable and the activity effect index as a dependent variable, calculates the activity effect index value corresponding to the index parameter through the regression model, and obtains the first weight coefficient of the regression model; determines a neural network model with the activity communication index as input and the activity effect index as output, and trains the second weight coefficient of the neural network model using the index parameter and the activity effect index value; performs fusion calculation on the first weight coefficient and the second weight coefficient to determine the target weight coefficient, and generates the activity effect index formula based on the target weight coefficient, the activity communication index and the activity effect index; uses the activity effect index formula to calculate the index parameters of the multi-dimensional activity communication index of the target activity and determine the target activity effect index value. The evaluation dimension of the activity effect is greatly increased, and at the same time, the accuracy of the influence of each evaluation dimension on the effect is ensured, thereby greatly improving the accuracy of the activity effect evaluation. According to one embodiment of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the computer executable instruction can execute the method for determining the activity propagation effect in any of the above method embodiments.

[0160] Figure 4 A schematic diagram of the structure of a terminal provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the terminal.

[0161] like Figure 4 As shown, the terminal may include: a processor (processor) 402 , a communication interface (Communications Interface) 404 , a memory (memory) 406 , and a communication bus 408 .

[0162] The processor 402 , the communication interface 404 , and the memory 406 communicate with each other via a communication bus 408 .

[0163] The communication interface 404 is used to communicate with other devices such as clients or other servers.

[0164] The processor 402 is used to execute the program 410, and specifically can execute the relevant steps in the above-mentioned method embodiment for determining the activity propagation effect.

[0165] Specifically, the program 410 may include program codes, which include computer operation instructions.

[0166] The processor 402 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the terminal may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0167] The memory 406 is used to store the program 410. The memory 406 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0168] The program 410 may be specifically configured to enable the processor 402 to perform the following operations:

[0169] Obtaining the publication volume and dissemination data of the target activity, wherein the dissemination data includes dissemination volume and multiple dissemination carrier data;

[0170] Generate at least one dimensional activity communication index according to the communication volume and the communication carrier data, and perform standardization on the activity communication index and the publication volume to obtain a multi-dimensional activity communication index of the target activity;

[0171] Using the multi-dimensional activity propagation index and the initial activity effect index to train the neural network model and the regression model, and determine the target weight coefficient of the activity propagation index of each dimension;

[0172] A target activity effect index value is calculated based on the target weight coefficient and the multi-dimensional activity propagation index, and the target activity effect index value is used to characterize the activity propagation effect of the target activity.

[0173] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0174] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for determining the effect of activity communication, characterized in that: include: Obtain indicator parameters of multi-dimensional activity communication indicators of sample activities, wherein the activity communication indicators include multiple ones of the activity content release volume, activity communication volume, activity reach, activity interaction volume, activity communication health rate and key information quantity; Determine a regression model with the activity communication index as an independent variable and the activity effect index as a dependent variable, calculate the activity effect index value corresponding to the indicator parameter through the regression model, and obtain a first weight coefficient of the regression model; Determine a neural network model with the activity propagation index as input and the activity effect index as output, and train a second weight coefficient of the neural network model using the index parameter and the activity effect index value; Performing a fusion calculation on the first weight coefficient and the second weight coefficient to determine a target weight coefficient, and generating an activity effect index formula based on the target weight coefficient, the activity propagation index and the activity effect index; Using the activity effect index formula, the indicator parameters of the multi-dimensional activity communication index of the target activity are calculated to determine the target activity effect index value. The calculation formula of the target activity effect index value is: y=ω1x1+ω2x2+…+ω i x i ; where x i are the activity propagation indicators, x1, x2…x i ∈(0,100), i is the number of dimensions of the activity effect parameter, y∈(0,100), ω i is the target weight coefficient; The step of fusing the first weight coefficient and the second weight coefficient to determine the target weight coefficient includes: updating the first weight coefficient according to an average of the second weight coefficient and the first weight coefficient; Recalculate the activity effect index value of the indicator parameter through the regression model with updated weight coefficient; Training the second weight coefficient corresponding to the neural network model according to the indicator parameter and the recalculated activity effect index value until the second weight coefficient or the activity effect index value meets the convergence condition, and determining the target weight coefficient based on the first weight coefficient of the final regression model and the second weight coefficient of the final neural network model; The method further comprises: using the activity effect index formula to calculate the index parameters of the multi-dimensional activity communication index of the target activity, and determining the target activity effect index value, and further comprising: using the activity reach calculation model to predict the content of the communication volume and the communication carrier to obtain the activity reach; and / or, Using an emotion recognition model to identify the content of the communication carrier, determine a first number of positive emotion category communication carriers and a second number of negative emotion category communication carriers, and calculate the activity communication health rate based on the first number and the second number; and / or, The target keyword matching the preset keyword of the target activity is searched from the communication carrier, and the quantity of key information is determined according to the occurrence frequency of the target keyword.

2. The method according to claim 1, characterized in that Before calculating the index parameters of the multi-dimensional activity communication index of the target activity by using the activity effect index formula and determining the target activity effect index value, the method further includes: Obtain the activity content release volume, dissemination volume and multiple dissemination carriers of the target activity; Calculating initial indicator parameters of at least one dimension of activity propagation indicator according to the propagation sound volume and the content of the propagation carrier; The initial indicator parameters and the release volume of the activity content are standardized to obtain indicator parameters of the multi-dimensional activity communication indicators of the target activity.

3. The method according to claim 2, characterized in that Before obtaining the data of the activity content release volume, the propagation volume and multiple propagation carriers of the target activity, the method further includes: Obtaining a target activity communication subject of the target activity, and identifying a globally associated account associated with the target activity communication subject from a pre-built full-platform activity communication subject account knowledge base; Target promotional carriers matching the theme keywords of the target activity are extracted from the promotional carriers released by the globally associated account, and the release quantity is determined based on the number of the target promotional carriers.

4. The method according to claim 3, characterized in that The activity communication index also includes the activity interaction volume. After extracting the target promotional carrier matching the theme keyword of the target activity from the promotional carriers released by the global associated account, the method further includes: Extracting a plurality of initial keywords from the target promotion carrier; Using the initial keyword and the initial logical operation relationship to verify the search result, if the result of the search result verification is passed, determining the initial keyword as a preset keyword, and determining the initial logical operation relationship as a preset logical operation relationship; Collecting the content and interactive data of the communication carrier according to the preset logical operation relationship and the preset keywords; The activity interaction volume is obtained by calculating the global interaction data of the propagation carriers, and the activity propagation volume is obtained by calculating the total number of the propagation carriers.

5. The method according to any one of claims 1 to 4, characterized in that After calculating the index parameters of the multi-dimensional activity communication index of the target activity by using the activity effect index formula and determining the target activity effect index value, the method further includes: Obtaining the resource input amount of the target activity; Calculating the resource input-output ratio of the target activity according to the resource input amount and the target activity effect index value; An evaluation report of the target activity is generated based on the resource input-output ratio, the indicator parameters of the multi-dimensional activity communication indicator and the target weight coefficient.

6. A device for determining the effect of activity communication, characterized in that: include: An acquisition module, used to acquire indicator parameters of multi-dimensional activity communication indicators of sample activities, wherein the activity communication indicators include multiple ones of the activity content release volume, activity communication volume, activity reach, activity interaction volume, activity communication health rate and key information quantity; A calculation module, used to determine a regression model with the activity propagation index as an independent variable and the activity effect index as a dependent variable, calculate the activity effect index value corresponding to the indicator parameter through the regression model, and obtain a first weight coefficient of the regression model; A training module, used to determine a neural network model with the activity propagation index as input and the activity effect index as output, and train a second weight coefficient of the neural network model using the index parameter and the activity effect index value; A generating module, configured to perform a fusion calculation on the first weight coefficient and the second weight coefficient to determine a target weight coefficient, and to generate an activity effect index formula based on the target weight coefficient, the activity propagation index and the activity effect index; A determination module, used to calculate the index parameters of the multi-dimensional activity communication index of the target activity by using the activity effect index formula, and determine the target activity effect index value; The calculation formula of the target activity effect index value is: y=ω1x1+ω2x2+…+ω i x i ; where x i are the activity propagation indicators, x1, x2…x i ∈(0,100), i is the number of dimensions of the activity effect parameter, y∈(0,100), ω i is the target weight coefficient; The step of fusing the first weight coefficient and the second weight coefficient to determine a target weight coefficient includes: updating the first weight coefficient according to an average of the second weight coefficient and the first weight coefficient; Recalculate the activity effect index value of the indicator parameter through the regression model with updated weight coefficient; Training the second weight coefficient corresponding to the neural network model according to the indicator parameter and the recalculated activity effect index value until the second weight coefficient or the activity effect index value meets the convergence condition, and determining the target weight coefficient based on the first weight coefficient of the final regression model and the second weight coefficient of the final neural network model; The method further comprises: using the activity effect index formula to calculate the index parameters of the multi-dimensional activity communication index of the target activity, and determining the target activity effect index value, and further comprising: using the activity reach calculation model to predict the content of the communication volume and the communication carrier to obtain the activity reach; and / or, Using an emotion recognition model to identify the content of the communication carrier, determine a first number of positive emotion category communication carriers and a second number of negative emotion category communication carriers, and calculate the activity communication health rate based on the first number and the second number; and / or, The target keyword matching the preset keyword of the target activity is searched from the communication carrier, and the quantity of key information is determined according to the occurrence frequency of the target keyword.

7. A storage medium storing at least one executable instruction, wherein the executable instruction enables a processor to execute operations corresponding to the method for determining the activity propagation effect as described in any one of claims 1-5.

8. A terminal, comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the method for determining the activity propagation effect as described in any one of claims 1-5.

Images