Multimedia information pushing method and device, equipment and medium

Abstract

The embodiment of the invention discloses a multimedia information pushing method and device, equipment and a medium. The method comprises the steps of obtaining current cargo associated data and a current user identifier of a to-be-pushed user, inputting the current cargo associated data into a trained real-time evaluation model, and outputting a current cargo evaluation result; processing the current cargo evaluation result according to a preset cargo classification dimension to obtain current cargo special data of the user to be pushed; determining whether a target user identifier which is the same as the current user identifier exists in preset historical user identifiers, and if the target user identifier exists, taking historical goods special data corresponding to the target user identifier as first reference goods special data according to a preset historical goods corresponding relation; and according to the first reference cargo special data and the current cargo special data, determining target multimedia information currently pushed to the to-be-pushed user. And the timeliness and the accuracy of the pushed target multimedia information are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data processing technology, and in particular to a multimedia information push method, device, equipment and medium. Background Art

[0002] With the rapid development of the Internet and the popularization of smart mobile terminals, multimedia information has become an important channel for merchants to promote their products. However, the traditional way of pushing multimedia information cannot meet the personalized needs of users and often brings a bad experience to users. Therefore, it is very important to improve the accuracy and timeliness of pushed multimedia information, so as to meet the personalized needs of users and improve the user experience. Summary of the invention

[0003] The present invention provides a multimedia information push method, device, equipment and medium to improve the accuracy and timeliness of pushed multimedia information, thereby meeting the personalized needs of users and improving the user experience.

[0004] According to one aspect of the present invention, a multimedia information push method is provided, including:

[0005] Acquire the current goods-related data and the current user ID of the user to be pushed, input the current goods-related data into the trained real-time evaluation model, and output the current goods evaluation result; wherein the user to be pushed is a user who is in the process of selecting goods;

[0006] According to the preset cargo classification dimensions, the current cargo evaluation results are processed to obtain the current cargo-specific data of the user to be pushed; wherein the cargo classification dimensions include cargo area dimensions, user behavior dimensions, multimedia information dimensions, and non-multimedia information dimensions;

[0007] Determine whether there is a target user identifier that is the same as the current user identifier in the preset historical user identifiers. If there is a target user identifier, use the historical goods special data corresponding to the target user identifier as the first reference goods special data according to the preset historical goods corresponding relationship;

[0008] According to the first reference cargo-specific data and the current cargo-specific data, target multimedia information to be currently pushed to the user to be pushed is determined.

[0009] According to another aspect of the present invention, there is provided a multimedia information push device, comprising:

[0010] An evaluation result determination module is used to obtain the current goods-related data and the current user identifier of the user to be pushed, and input the current goods-related data into the trained real-time evaluation model to output the current goods evaluation result; wherein the user to be pushed is a user who is in the process of selecting goods;

[0011] The current data determination module is used to process the current cargo evaluation result according to the preset cargo classification dimension to obtain the current cargo-specific data of the user to be pushed; wherein the cargo classification dimension includes cargo area dimension, user behavior dimension, multimedia information dimension and non-multimedia information dimension;

[0012] A reference data determination module is used to determine whether there is a target user identifier that is the same as the current user identifier in the preset historical user identifiers. If there is a target user identifier, the historical goods special data corresponding to the target user identifier is used as the first reference goods special data according to the preset historical goods corresponding relationship.

[0013] The target information pushing module is used to determine the target multimedia information to be currently pushed to the user to be pushed based on the first reference cargo-specific data and the current cargo-specific data.

[0014] According to another aspect of the present invention, there is provided an electronic device, comprising:

[0015] one or more processors;

[0016] A memory for storing one or more programs;

[0017] When one or more programs are executed by one or more processors, the one or more processors can execute any one of the multimedia information pushing methods provided by the embodiments of the present invention.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions, and the computer instructions are used to enable a processor to implement any multimedia information push method provided by the embodiments of the present invention when executed.

[0019] An embodiment of the present invention provides a multimedia information push solution, which obtains current goods-related data and current user identification of a user to be pushed, and inputs the current goods-related data into a trained real-time evaluation model to output a current goods evaluation result; wherein the user to be pushed is a user who is in the process of selecting goods; according to a preset goods classification dimension, the current goods evaluation result is processed to obtain current goods-specific data of the user to be pushed; wherein the goods classification dimension includes a goods area dimension, a user behavior dimension, a multimedia information dimension and a non-multimedia information dimension; determines whether there is a target user identification that is the same as the current user identification in the preset historical user identification, and if there is a target user identification, then according to the preset historical goods correspondence, the historical goods-specific data corresponding to the target user identification is used as the first reference goods-specific data; according to the first reference goods-specific data and the current goods-specific data, determines the target multimedia information currently being pushed to the user to be pushed. The above scheme determines the current cargo-specific data by introducing a real-time evaluation model and cargo classification dimensions, and pushes target multimedia information to users who are in the process of selecting cargo based on the first reference cargo-specific data and the current cargo-specific data, thereby improving the timeliness and accuracy of the pushed target multimedia information, thereby meeting the user's personalized needs for multimedia information and improving the user experience of the users who are to be pushed.

[0020] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a flowchart of a multimedia information push method provided by Embodiment 1 of the present invention;

[0023] Figure 2 is a flowchart of a multimedia information push method provided by Embodiment 2 of the present invention;

[0024] Figure 3 It is a structural diagram of a multimedia information push device provided by Embodiment 4 of the present invention;

[0025] Figure 4It is a structural diagram of an electronic device for implementing a multimedia information push method provided in Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0026] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0027] ESL is an electronic label system used to display prices and information of goods in a retail environment. It can be updated remotely, improve efficiency and reduce reliance on traditional paper labels. AIoT is the abbreviation of the intelligent Internet of Things, which refers to the interconnection and collaboration of smart devices through the network to provide more intelligent services and solutions. Cloud services can be computing and data storage services provided through the Internet. The cloud services in the present invention are used to support the processor and storage data of generative artificial intelligence.

[0028] Level 1 goods refer to goods that are most consistent with the user's purchasing intention and behavior through data analysis of probability models (i.e. historical evaluation models) and regression models (i.e. real-time evaluation models). Level 2 goods refer to goods that are more consistent with the user's purchasing intention and behavior through data analysis of probability models and regression models. Level 3 goods refer to goods that are inconsistent with the user's purchasing intention and behavior through data analysis of probability models and regression models.

[0029] Closed-loop analysis refers to a comprehensive analysis process, which begins with the collection of user behavior and multimedia information data, and uses probability model analysis to predict market size trends and regression model analysis to analyze the correlation between goods sales and multimedia information. It then provides a market analysis report, a decisive factor analysis report, and a multimedia information effect evaluation analysis report. Through these analysis results, various factors are fed back into the multimedia information, realizing the function of dynamically adjusting the recommended multimedia information content.

[0030] The decisive factors can include vision, smell, hearing, touch and taste, and users' emotional connection and brand memory can be comprehensively improved through color psychology, layout design, high-quality images and videos, aroma creation, music and sound logos, interactive experience and tasting activities. The combination of different senses can effectively convey brand information and enhance the appeal and communication effect of multimedia information.

[0031] Although existing technologies provide a variety of ways to deliver multimedia information, such as product labels, electronic price tags, QR code or barcode scanning, NFC technology interaction, in-store digital touch screens, mobile applications and social media, they usually face the following technical problems and limitations: Lack of in-depth analysis: Most existing technologies fail to provide in-depth data analysis, including but not limited to price comparisons of goods, sales rankings, etc., which is crucial to determine the factors that affect users' behavior in selecting goods. Lack of factor analysis of multimedia information: It is impossible to determine which multimedia information affects users' selection of goods. Lack of closed-loop analysis: The push effect of multimedia information is poor.

[0032] In view of the above problems, the present invention proposes a multimedia information delivery effect analysis and purchase behavior analysis system based on smart devices. The system conducts in-depth analysis of market data, and analyzes it through probability models and regression models, and generates reports to feedback to merchants and multimedia information distributors. This system significantly improves the problems in the prior art, provides real-time feedback to merchants and multimedia information distributors, helps them deeply understand which factors have a deeper impact on users' purchasing intentions, and more intuitively understands the effects and benefits of multimedia information delivery.

[0033] The existing multimedia information push methods usually lack real-time data and model analysis to push multimedia information to users in a timely and accurate manner, and the user experience is low. In view of the above problems, the embodiment of the present invention proposes a multimedia information push method based on a smart device, which can not only provide factor analysis of multimedia information, but also analyze the impact of multimedia information on user's behavior of selecting goods.

[0034] Embodiment 1

[0035] Figure 1 This is a flowchart of a multimedia information push method provided in Example 1 of the present invention. This embodiment is applicable to situations where multimedia information is pushed to users. The method can be executed by a multimedia push device, which can be implemented in software and / or hardware and can be configured in an electronic device that carries the multimedia information push function.

[0036] See also Figure 1 The multimedia information push method shown includes:

[0037] S110, obtaining the current goods-related data and the current user identifier of the user to be pushed, and inputting the current goods-related data into the trained real-time evaluation model, and outputting the current goods evaluation result.

[0038] The user to be pushed is a user who is in the process of selecting goods, that is, the user to be pushed can be understood as a user who is purchasing goods. The current goods-related data refers to data associated with the goods selection of the user to be pushed. Exemplarily, the current goods-related data includes the current behavior data, current goods data and current sensory data of the user to be pushed.

[0039] The multimedia information refers to digital information used to introduce goods to users. For example, the multimedia information can be understood as digital advertisements.

[0040] The current behavior data refers to the behavior data of the user to be pushed. Exemplarily, the current behavior data may include at least one of data on the goods selected by the user to be pushed, the viewing time of the multimedia information watched by the user to be pushed, the movement lines before and after the user to be pushed, and whether the user to be pushed selects the goods displayed in the multimedia information.

[0041] The current goods data refers to the attribute data of the multimedia information that has been pushed to the user to be pushed. Exemplarily, the current goods data may include the default display duration of the multimedia information, the type of the multimedia information, and the type of goods displayed by the multimedia information. The default display duration of the multimedia information refers to the display duration of the multimedia information without interference. The type of the multimedia information refers to the type of the multimedia information. Exemplarily, the type of the multimedia information may include branding and effectiveness.

[0042] The current sensory data refers to the sensory data of the multimedia information that has been pushed to the user to be pushed. For example, the current sensory data may include the color psychology, layout and design, image and video quality, background music and sound identification, and fragrance of the scene creation of the multimedia information. The current sensory data covers the visual, auditory, olfactory, tactile, and taste elements of the multimedia information.

[0043] The embodiment of the present invention does not impose any limitation on the method of obtaining the current goods association data, which can be set by a technician based on experience. Exemplarily, the current goods association data can be obtained from the current client. Exemplarily, the current client can be at least one of a multimedia information carrier, an ESL controller, an Alot device, an IoT device, an intelligent shopping cart positioning device, and a mobile terminal. The diversity of the current client ensures the reliability of the current goods association data.

[0044] Among them, multimedia information carriers refer to ESL (Electronic Shelf Label), LCD (Liquid Crystal Display)-labels, ordinary display screens, smart shopping cart device displays and robot displays as carriers, using NFC (Near Field Communication) technology, QR code / barcode scanning functions or visual algorithm recognition technology as interactive perception means to achieve multimedia information delivery to users in various ways. These labels not only display traditional product information such as name, price and inventory status, but also multimedia information.

[0045] The ESL controller, as a signal transmission device, is responsible for managing the communication between the ESL tag and the cloud-based GENAI (Generative Artificial Intelligence) processor. It ensures real-time data updates and synchronization, including but not limited to changes in cargo information, price adjustments, and activities to promote the sale of cargo.

[0046] Among them, AIoT devices refer to terminals with edge computing and data collection capabilities, such as smart shopping carts, cameras, mobile terminals and store patrol robots. As interactive carriers, they trigger signals with ESL, obtain and send real-time data information through the intelligent interactive cloud service system, and present it on its interface or through voice broadcast, directly synchronized to the user, and can also collect desensitized data, and observe and analyze the user's behavior in selecting goods.

[0047] IoT devices refer to devices deployed and installed in stores to promote sales and interactive IoT sensors. Smart shopping cart positioning devices can be used to record the movement path and location of users' shopping carts. Mobile terminals can be used to record users' interactive behaviors, such as the length of time they watch multimedia information and the behavior of selecting goods.

[0048] It can be understood that by introducing current behavior data, current cargo data and current sensory data, the current cargo-related data is determined, thereby improving the comprehensiveness of the determined current cargo-related data; at the same time, a closed-loop analysis is achieved and the collection of decisive factors is supplemented, thereby ensuring the traceability and feedback of multimedia information delivery data, and deeply analyzing multimedia information content and user behavior, so as to provide more accurate basic data for the subsequent determination of target multimedia information.

[0049] The current user ID can be used to uniquely represent the identity of the user to be pushed. The real-time evaluation model can be used to evaluate the goods selection behavior of the user to be pushed in real time based on the current goods association data to obtain the current goods evaluation result. The real-time evaluation model can evaluate the impact of different multimedia information on the user to be pushed through multivariate regression analysis.

[0050] Exemplarily, the process of determining model parameters when building a real-time evaluation model can be selecting and building feature variables, performing feature variable selection, and creating interactive features and polynomial features. Among them, performing feature variable selection refers to screening out key features that affect user behavior. Creating interactive features and polynomial features can be used to capture nonlinear relationships between features.

[0051] For example, a real-time evaluation model can be constructed using multiple regression analysis techniques and regression algorithms, which can include linear regression, ridge regression, Lasso regression, etc., which can be selected based on data characteristics and analysis requirements.

[0052] Exemplarily, training and validating the real-time evaluation model may include: dividing the data set into a training set and a validation set, training the model with the training set, and evaluating the performance of the model with the validation set; at the same time, further optimizing the model parameters using a cross-validation method to improve the accuracy and stability of the model; and, checking the model's fit and residual distribution to ensure the rationality of the model.

[0053] For example, the performance of the real-time evaluation model can be evaluated by using performance indicators such as R-squared, mean square error (MSE), and mean absolute error (MAE); the coefficients of the real-time evaluation model can be analyzed to reveal the degree of influence of different factors on the sales volume of goods; the real-time evaluation model can be interpreted to identify the key factors that have the greatest impact on the sales volume of goods.

[0054] The current goods evaluation result refers to the result obtained by the real-time evaluation model evaluating the goods selection behavior of the user to be pushed.

[0055] Specifically, current goods-related data and current user identifiers of users to be pushed are obtained from multiple current clients; the current goods-related data are preprocessed, and the preprocessed current goods-related data are input into a trained real-time evaluation model to obtain current goods evaluation results.

[0056] The embodiment of the present invention does not impose any limitation on the preprocessing, which can be set by a technician based on experience. Exemplarily, the preprocessing may include data cleaning and sorting, data standardization and normalization, and outlier processing.

[0057] Among them, data cleaning and sorting can be used to remove invalid current cargo-related data and noise, ensuring that all current cargo-related data used for real-time evaluation models are accurate and reliable; at the same time, identify and process missing values, using techniques such as multiple interpolation to maintain the integrity of current cargo-related data. Data standardization and normalization refer to standardizing or normalizing the numerical data in the current cargo-related data to ensure that different variables are comparable in the model. Outlier processing refers to the use of a preset anomaly detection method to identify and process outliers in the current cargo-related data to avoid distortion of future trend predictions. The embodiment of the present invention does not impose any limitation on the preset anomaly detection method, which can be set by a technician based on experience.

[0058] S120. Process the current cargo evaluation result according to the preset cargo classification dimension to obtain the current cargo-specific data to be pushed to the user.

[0059] Among them, the cargo classification dimension includes cargo area dimension, user behavior dimension, multimedia information dimension and non-multimedia information dimension. The cargo area dimension can be used to analyze the market trend of cargo in a certain cargo area. Exemplarily, the cargo area can be determined based on the geographical location of the user to be pushed and the preset push range. The embodiment of the present invention does not impose any limitation on the preset push range, which can be set by technical personnel based on experience. The user behavior dimension can be used to analyze the user's behavior in selecting cargo. The multimedia information dimension can be used to analyze the degree of influence of multimedia information on users. The non-multimedia information dimension can be used to analyze the degree of influence of preset factors on users. The embodiment of the present invention does not impose any limitation on the preset factors, which can be other factors that affect the user's selection of cargo except multimedia information.

[0060] The current cargo-specific data refers to data corresponding to a certain cargo classification sub-dimension obtained from the analysis of the current cargo evaluation results. The cargo classification sub-dimension may be one of a cargo area dimension, a user behavior dimension, a multimedia information dimension, and a non-multimedia information dimension.

[0061] Exemplarily, if the goods classification sub-dimension is the goods area dimension, then the corresponding current goods special data is the market trend data; if the goods classification sub-dimension is the user behavior dimension, then the corresponding current goods special data is the user behavior data; if the goods classification sub-dimension is the multimedia information dimension, then the corresponding current goods special data is the relationship data between the multimedia information and the goods sold; if the goods classification sub-dimension is the non-multimedia information dimension, then the corresponding current goods special data is the relationship data between the preset factors and the goods sold.

[0062] Specifically, according to the preset cargo classification dimension, the current cargo evaluation result is analyzed to obtain the current cargo-specific data corresponding to each cargo classification sub-dimension in the cargo classification dimension.

[0063] S130. Determine whether there is a target user identifier that is the same as the current user identifier in the preset historical user identifiers. If there is a target user identifier, use the historical cargo special data corresponding to the target user identifier as the first reference cargo special data according to the preset historical cargo corresponding relationship.

[0064] The historical user identifier can be used to uniquely represent the identity of the historical push user. The historical push user refers to a user who has received multimedia information in the past. The target user identifier refers to the historical user identifier that is the same as the current user identifier.

[0065] Among them, the historical goods correspondence refers to the correspondence between the historical user identification and the historical goods special data. The historical goods special data refers to the data corresponding to a certain goods classification sub-dimension obtained from the analysis of the historical goods evaluation results. The first reference goods special data refers to the historical goods special data under each goods classification sub-dimension corresponding to the target user identification.

[0066] In an optional embodiment, the historical goods correspondence is determined based on the following method: obtaining the historical goods association data and historical user identification of the historical push user, and inputting the historical goods association data into the trained historical evaluation model to output the historical goods evaluation results; processing the historical goods evaluation results according to the preset goods classification dimension to obtain the historical goods special data of the historical push user; generating the historical goods correspondence including the historical user identification and the corresponding historical goods special data.

[0067] The historical goods-related data refers to the data related to the goods selected by the historical push user. Exemplarily, the historical goods-related data may include the historical behavior data, historical goods data and historical sensory data of the historical push user.

[0068] Among them, historical behavior data refers to the behavior data of historical push users. Historical goods data refers to the attribute data of multimedia information previously pushed to historical push users. Historical sensory data refers to the sensory data of multimedia information previously pushed to historical push users.

[0069] The historical goods evaluation result refers to the result obtained by evaluating the goods selection behavior of historical push users by the historical evaluation model. The historical evaluation model can be used to evaluate the goods selection behavior of historical push users based on the historical goods association data to obtain the historical goods evaluation result. The historical evaluation model can achieve a comprehensive evaluation of the delivery effect of multimedia information through closed-loop analysis, so as to predict the market size trend and optimize the content of multimedia information.

[0070] Exemplarily, the model parameters determined when constructing the historical evaluation model may include feature variables, time series features, and interaction and polynomial features. Feature variables may be variables indicating changes in market trends, such as inventory levels, market shares, and user behavior trends. In the embodiments of the present invention, sliding window statistics and time decay factors may be introduced to capture time dynamics and trends and obtain time series features. Interaction and polynomial features may be used to simulate the interactions between different market factors and their impact on market trends.

[0071] For example, a time series model (such as ARIMA (Autoregressive Integrated Moving Average Model) or seasonal ARIMA) or a machine learning model for processing complex relationships (such as random forest or neural network) can be used.

[0072] Exemplarily, training and validating the historical valuation model may include: data set segmentation, i.e., using a time series method to segment historical cargo-related data to ensure that the training set and the test set reflect different time periods; cross-validation, i.e., applying time series cross-validation to gradually increase the length of the training data to verify the performance of the historical valuation model at different time points; and dynamic model adjustment, i.e., dynamically adjusting the parameters of the historical valuation model based on market feedback to cope with the rapidly changing market environment.

[0073] For example, performance indicators specifically for time series prediction, such as MAPE (mean absolute percentage error) and SMAPE (symmetric mean absolute percentage error), can be used as performance indicators to evaluate the performance of the historical evaluation model; the coefficients of the historical evaluation model are analyzed to reveal the degree of influence of different factors on sales; the historical evaluation model is used to predict future market trends and provide confidence intervals and prediction intervals for trend changes.

[0074] It can be understood that by introducing the historical valuation model, the historical cargo related data is processed to obtain the historical cargo valuation results, and then based on the historical cargo valuation results and the cargo classification dimensions, the historical cargo special data is obtained to construct the historical cargo correspondence relationship, thereby improving the accuracy of the constructed historical cargo correspondence relationship.

[0075] S140. Determine the target multimedia information currently being pushed to the user to be pushed based on the first reference cargo-specific data and the current cargo-specific data.

[0076] The target multimedia information refers to the multimedia information currently being pushed to the user to be pushed. Exemplarily, the number of the target multimedia information may be at least one.

[0077] Optionally, if the number of target multimedia information is at least two, the push times of each target multimedia information can be obtained respectively; the push order of the target multimedia information is determined according to the push times; and the target multimedia information is pushed to the user to be pushed according to the push order. The push times refers to the number of times a certain multimedia information is pushed to the user.

[0078] Or optionally, if the number of target multimedia information is at least two, the number of pushes, the total historical display time and the default display time of each target multimedia information can be obtained respectively; the historical average display time of each target multimedia information can be determined according to the number of pushes and the total historical display time of each target multimedia information; the information display degree of each target multimedia information can be determined according to the historical average display time and the corresponding default display time; the push order of the target multimedia information can be determined according to the information display degree; and the target multimedia information can be pushed to the users to be pushed according to the push order.

[0079] The total historical display duration refers to the total playback duration of the target multimedia information in the past. The default display duration refers to the playback setting duration of the target multimedia information. The historical average display duration refers to the average playback duration of the target multimedia information in the past. The information display degree can be used to quantify the playback degree of the target multimedia information.

[0080] An embodiment of the present invention provides a multimedia information push solution, which obtains current goods-related data and current user identification of a user to be pushed, and inputs the current goods-related data into a trained real-time evaluation model to output a current goods evaluation result; wherein the user to be pushed is a user who is in the process of selecting goods; according to a preset goods classification dimension, the current goods evaluation result is processed to obtain current goods-specific data of the user to be pushed; wherein the goods classification dimension includes a goods area dimension, a user behavior dimension, a multimedia information dimension and a non-multimedia information dimension; determines whether there is a target user identification that is the same as the current user identification in the preset historical user identification, and if there is a target user identification, then according to the preset historical goods correspondence, the historical goods-specific data corresponding to the target user identification is used as the first reference goods-specific data; according to the first reference goods-specific data and the current goods-specific data, determines the target multimedia information currently being pushed to the user to be pushed. The above scheme determines the current cargo-specific data by introducing a real-time evaluation model and cargo classification dimensions, and pushes target multimedia information to users who are in the process of selecting cargo based on the first reference cargo-specific data and the current cargo-specific data, thereby improving the timeliness and accuracy of the pushed target multimedia information, thereby meeting the user's personalized needs for multimedia information and improving the user experience of the users who are to be pushed.

[0081] Based on the above technical solution, the present invention also provides an optional embodiment. Exemplarily, if there is no target cargo identification, the historical cargo-specific data corresponding to the cargo area dimension is used as the second reference cargo-specific data; the current cargo area factor is determined based on the second reference cargo-specific data, and the target multimedia information currently pushed to the user to be pushed is determined based on the current cargo area factor and the current cargo-specific data.

[0082] The second reference goods-specific data refers to the historical goods-specific data corresponding to the goods area dimension. The current goods area factor refers to the trend factor of users purchasing goods in the area where the user to be pushed is located.

[0083] Exemplarily, based on the current cargo area factor and the current cargo special data, the target multimedia information to be pushed to the user to be pushed is determined, including: based on the current cargo area factor, filtering out the regional multimedia information from the preset candidate multimedia information; based on the current cargo special data, filtering out the special multimedia information from the preset candidate multimedia information; based on the preset regional special ratio, determining the target multimedia information according to the regional multimedia information and the special multimedia information.

[0084] Among them, candidate multimedia information refers to pre-set multimedia information. Regional multimedia information refers to candidate multimedia information that meets the current cargo regional factors. Special multimedia information refers to candidate multimedia information determined based on the current cargo special data. Exemplarily, the current cargo special data under a cargo classification sub-dimension has at least one corresponding special multimedia information.

[0085] The area-specific ratio can be used to determine the proportion of the regional multimedia information and the special multimedia information in the target multimedia information, that is, the area-specific ratio can determine the amount of regional multimedia information as the target multimedia information and the amount of special multimedia information as the target multimedia information. The embodiment of the present invention does not impose any limitation on the size of the area-specific ratio, which can be set by technicians based on experience or needs.

[0086] It should be noted that if there is overlapping multimedia information in the regional multimedia information and the special multimedia information, the overlapping multimedia information can be directly used as at least part of the target multimedia information; according to the preset regional-specific ratio, the remaining target multimedia information is screened from the regional multimedia information and the special multimedia information from which the overlapping multimedia information is respectively removed. Exemplarily, the same regional multimedia information and the special multimedia information are used as overlapping multimedia information.

[0087] It can be understood that in the absence of a target cargo identification, the target multimedia information is determined based on the historical cargo-specific data and the current cargo-specific data in the cargo area dimension, thereby improving the accuracy of the determined target multimedia information.

[0088] Embodiment 2

[0089] Figure 2 This is a flowchart of a multimedia information push method provided by the second embodiment of the present invention. Based on the above embodiments, this embodiment further refines the operation of "determining the target multimedia information currently pushed to the user to be pushed based on the first reference cargo-specific data and the current cargo-specific data" into "determining the current cargo push factor based on the first reference cargo-specific data and the current cargo-specific data; determining the target multimedia information from the preset candidate multimedia information based on the current cargo push factor" to improve the determination mechanism of the target multimedia information. It should be noted that for the part not described in detail in the embodiment of the present invention, reference can be made to the description of other embodiments.

[0090] See also Figure 2 The multimedia information push method shown includes:

[0091] S210, obtaining the current goods-related data and the current user identifier of the user to be pushed, and inputting the current goods-related data into the trained real-time evaluation model, and outputting the current goods evaluation result.

[0092] Among them, the users to be pushed are users who are in the process of selecting goods.

[0093] S220. Process the current cargo evaluation result according to the preset cargo classification dimension to obtain the current cargo-specific data to be pushed to the user.

[0094] Among them, the cargo classification dimensions include cargo area dimension, user behavior dimension, multimedia information dimension and non-multimedia information dimension.

[0095] S230. Determine whether there is a target user identifier that is the same as the current user identifier in the preset historical user identifiers. If there is a target user identifier, use the historical cargo special data corresponding to the target user identifier as the first reference cargo special data according to the preset historical cargo corresponding relationship.

[0096] S240. Determine current cargo push factors according to the first reference cargo specific data and the current cargo specific data.

[0097] Among them, the current goods push factor refers to the factor that affects the goods selection behavior of the user to be pushed.

[0098] In an optional embodiment, determining the current cargo push factor based on the first reference cargo special data and the current cargo special data includes: comparing the first reference cargo special data and the current cargo special data belonging to the same cargo classification sub-dimension; determining the current cargo push factor based on the comparison results corresponding to different cargo classification sub-dimensions.

[0099] Among them, the cargo classification sub-dimension refers to one of the cargo classification dimensions.

[0100] Exemplarily, for any cargo classification sub-dimension, the first reference cargo special data corresponding to the cargo classification sub-dimension and the current cargo special data are compared; if the comparison result is consistent, the current cargo push key factor is determined according to the current cargo special data under the cargo classification sub-dimension; if the comparison result is inconsistent, the current cargo push marginal factor is determined according to the current cargo special data under the cargo classification sub-dimension; the current cargo push factor is determined according to the current cargo push key factor and / or the current cargo push marginal factor.

[0101] Among them, the current key factors of goods push refer to the main factors that affect the selection of goods by the users to be pushed, and the current marginal factors of goods push refer to the secondary factors that affect the selection of goods by the users to be pushed.

[0102] It can be understood that by comparing the first reference cargo special data and the current cargo special data belonging to the same cargo classification sub-dimension, the current cargo push factor is determined, thereby comprehensively considering the historical cargo special data and the current cargo special data, and improving the accuracy of the determined current cargo push factor.

[0103] S250: Determine target multimedia information from preset candidate multimedia information according to current cargo push factors.

[0104] Exemplarily, if the current cargo push factors include the current cargo push key factors and the current cargo push edge factors, then according to the current cargo push key factors, the candidate key multimedia information is determined from the candidate multimedia information, and according to the current cargo push edge factors, the candidate edge multimedia information is determined from the candidate multimedia information; according to the preset push ratio, the number of key information of the target key multimedia information in the target multimedia information and the number of edge information of the target edge multimedia information in the target multimedia information are determined respectively; according to the number of key information, the number of edge information, the information display degree of the target key multimedia information, and the information display degree of the target edge multimedia information, the push order of the target multimedia information is determined; according to the push order, the target multimedia information is pushed to the users to be pushed.

[0105] The candidate key multimedia information refers to the candidate multimedia information corresponding to the key factors of the current cargo push. The candidate edge multimedia information refers to the candidate multimedia information corresponding to the edge factors of the current cargo push. The embodiment of the present invention does not impose any limitation on the size of the preset push ratio, which can be set by a technician based on experience or needs, or determined repeatedly through a large number of experiments.

[0106] The number of key information refers to the number of target key multimedia information in the target multimedia information, and the number of edge information refers to the number of target edge multimedia information in the target multimedia information.

[0107] The target key multimedia information refers to candidate key multimedia information for the target multimedia information, and the target edge multimedia information refers to candidate edge multimedia information for the target multimedia information.

[0108] Exemplarily, the push order of target multimedia information is determined based on the number of key information, the number of edge information, the information display degree of candidate key multimedia information, and the information display degree of candidate edge multimedia information, including: selecting target key multimedia information from the candidate key multimedia information based on the number of key information and the information display degree of candidate key multimedia information; selecting target edge multimedia information from the candidate edge multimedia information based on the number of edge information and the information display degree of candidate edge multimedia information; determining the push order of target multimedia information based on the information display degree of target key multimedia information, the information display degree of target edge multimedia information and a preset information sorting method.

[0109] Among them, the information sorting method refers to the sorting method of the target key multimedia information and the target edge multimedia information. The embodiment of the present invention does not impose any limitation on the information sorting method, which can be set by the technician according to experience or needs. Exemplarily, the information sorting method can be that the target key multimedia information is in front and the target edge multimedia information is in the back; the information sorting method can also be that the target key multimedia information and the target edge multimedia information are sorted according to a preset edge ratio. The embodiment of the present invention does not impose any limitation on the size of the preset edge ratio, which can be set by the technician according to experience. Exemplarily, if the preset edge ratio is 1:1, it indicates that the information sorting method is one target key multimedia information followed by one target edge multimedia information; if the preset edge ratio is 2:1, it indicates that the information sorting method is two target key multimedia information followed by one target edge multimedia information.

[0110] The embodiment of the present invention provides a multimedia information push scheme, which refines the target multimedia information operation of the user to be pushed according to the first reference cargo-specific data and the current cargo-specific data into determining the current cargo push factor according to the first reference cargo-specific data and the current cargo-specific data; and determining the target multimedia information from the preset candidate multimedia information according to the current cargo push factor, thereby improving the determination mechanism of the target multimedia information. The above scheme improves the accuracy of the determined current cargo push factor by determining the current cargo push factor according to the first reference cargo-specific data and the current cargo-specific data, thereby improving the accuracy of the target multimedia information determined based on the current cargo push factor in the subsequent process.

[0111] Embodiment 3

[0112] Based on the above embodiments, the embodiment of the present invention provides a multimedia information push system that takes the smart device carried by the shopping cart as the core, combines mobile terminal applications and QR code technology, and realizes the intelligent function of the shopping cart. Multimedia information is pushed through the screen of the smart device to provide users with timely and accurate multimedia information, thereby improving the push effect of multimedia information.

[0113] Among them, the multimedia information push system may include data sources and corresponding hardware carriers, data collection equipment, data processing servers and data analysis servers.

[0114] Among them, the data sources and corresponding hardware carriers may include multimedia information carriers (such as ESL / LCD labels and televisions), ESL controllers, AIoT devices, smart shopping cart positioning devices, mobile terminals and cash registers.

[0115] The data collection device includes a data collection module. The data collection module can be installed on a hardware carrier such as a shopping cart to collect multimedia information data and user behavior data. This includes the viewing time of multimedia information, the type of multimedia information, the type of goods, the user's behavior path, etc. In addition, multi-sensory elements of multimedia information, such as color psychology, sound, fragrance, etc., are also collected to provide a more in-depth effect analysis of multimedia information.

[0116] Among them, the data processing server may include a probability model (i.e., a historical evaluation model) and a regression model (i.e., a real-time evaluation model). The probability model can integrate the duration of multimedia information, the type of multimedia information, the type of goods, and other data, and use the probability model to conduct a closed-loop analysis of the effect of multimedia information and optimize the delivery strategy of multimedia information. The model uses a machine learning algorithm to identify the most effective elements and combinations of multimedia information to achieve dynamic adjustment and real-time response. The regression model can use multiple regression analysis technology to take factors such as the duration of multimedia information, the type of multimedia information, and the type of goods as independent variables, and the number of goods sold as the dependent variable to analyze the specific impact of the elements of multimedia information on the number of goods sold, provide real-time data analysis capabilities, and ensure that the push strategy of multimedia information can be adjusted in time to adapt to market changes and improve the user experience.

[0117] The data analysis server includes a data analysis model and an analysis result module. The data analysis module may include a user behavior data analysis module, a regional behavior analysis module, a multimedia information analysis module, and other analysis modules.

[0118] Among them, the user behavior data analysis module can be used to analyze the user's habits and preferences for selecting goods and provide personalized multimedia information push suggestions. Specifically, multiple security measures are used to protect user accounts, and the user's browsing habits, goods selection history and preferences are accurately recorded through background data collection. The multi-database strategy is used to improve query efficiency and ensure efficient data processing and analysis; if the user authorizes, the user behavior data can be collected and analyzed to provide accurate data support for the user analysis module. This module uses machine learning algorithms and real-time data analysis to generate user behavior analysis reports to enhance merchants' understanding of user purchasing behavior. By efficiently managing and analyzing various types of user behavior data, more accurate goods-specific data is provided. This module includes the following but is not limited to the following analysis methods: First, through the analysis method of user statistical data, the user behavior analysis module can identify the characteristics and needs of different user groups and customize multimedia information push strategies. The analysis of data enables the intelligent engine to understand the user's goods selection preferences, price sensitivity and goods selection time, helping merchants optimize product portfolios and pricing strategies. The analysis of psychological data can create user portraits and enhance the emotional connection between merchants and users. The use of user feedback data can timely discover problems and improve products and services to improve user satisfaction.

[0119] Among them, the regional behavior analysis module can provide region-specific market insights by analyzing local and regional user selection data. Data storage and management adopts a hybrid database architecture, using a database to store unstructured user behavior and preference data, and using a relational database to store structured user basic information and product selection records to ensure efficient data query, and through advanced encryption technology and data protection measures, ensure the privacy and security of user data. The local goods data collection module expands the data source, that is, the relevant data of goods purchases from different merchants in the region. IoT devices, POS systems and user movement trajectory data are used to obtain more detailed information on goods selection behavior, and public economic indicators, social media activities and weather data in the region are integrated to form a comprehensive data pool. In the data cleaning and preprocessing steps, the data is enhanced and corrected through a unified data format and machine learning technology to ensure data consistency and comparability. Next, based on multi-dimensional data such as shopping behavior and social media interaction, users are segmented, the characteristics and preferences of different groups are identified, and the goods selection behavior patterns of each group in each region are analyzed, including but not limited to goods category preferences, shopping frequency, time preferences and promotion reactions.

[0120] Among them, the multimedia information analysis module can evaluate different aspects of multimedia information, such as the effect of duration, type and content, and optimize the push strategy of multimedia information. Big data technology is used to efficiently collect and process the data of multimedia information to ensure the integrity and accuracy of the data. A distributed database system is used to store and manage the data of multimedia information so as to quickly query and analyze the variables of various types of multimedia information. This module is mainly used to analyze multiple aspects of multimedia information data, especially to explore the relationship between elements such as duration, type, content and the number of goods sold. This module reveals the impact of various multimedia information attributes on user response by performing detailed data collection and analysis on the variables of multimedia information. For example, analyze how video multimedia information of different lengths affects the viewing completion rate, or evaluate the effect of different multimedia information formats in stimulating user purchasing behavior. In this way, the multimedia information analysis module provides key data on which elements of multimedia information influence users to select goods.

[0121] Among them, other analysis modules can be used to integrate and analyze data other than multimedia information that affect users' selection of goods, in order to determine other factors that affect the sale of goods. Implement data warehouse technology to integrate various marketing data to ensure data uniformity and consistency. Use cloud storage solutions to safely store large amounts of historical goods-related data and support large-scale concurrent access and analysis. Provide corresponding data by analyzing comprehensive data related to the sale of goods, such as the relationship between activity type, goods profit margin, quantity of goods sold, and inventory. Other analysis modules can be used to determine other factors other than multimedia information that affect users' selection of goods, so that merchants can better adjust multimedia information push and goods display.

[0122] Among them, the analysis result module can obtain market analysis results, multimedia information analysis results and key factor analysis results. Market analysis results can be used to provide macro-perspective insights into market trends, user preferences and industry dynamics. Analyze goods sales data, market growth rate and user behavior patterns to predict future market trends. Multimedia information analysis results can be used to specifically evaluate the direct impact of the duration, type and content of multimedia information on goods sales, optimize the push strategy of multimedia information; analyze the performance indicators of different multimedia information, such as viewing rate, click-through rate, conversion rate, etc.; deeply analyze which content is most attractive to users and how different creativity affects the overall effect of multimedia information; evaluate how different types (branded multimedia information, direct response multimedia information) and different lengths of multimedia information affect user reactions and behaviors. Key factor analysis results can be used to deeply explore how factors such as visual design, emotional elements, and interactive experience of multimedia information affect user behavior in selecting goods. Evaluate the specific impact of elements such as color psychology, layout and design, music and sound identification.

[0123] Exemplarily, the current goods-related data is collected through the data collection modules of each device. Specifically, the multimedia information data collection modules of each multimedia information carrier collect the data of multimedia information including the type, duration, push time, form, location, etc. of the multimedia information; the user behavior data collection module records the behavior data of the user to be pushed, such as the length of time watching the video or pop-up window, whether to close the multimedia information, the movement line after watching, whether to go to the area of ​​the goods displayed in the multimedia information, and whether to purchase the corresponding goods; the transaction device data collection module collects the goods selection data, such as the selected goods information, etc.

[0124] Furthermore, the current cargo-related data and historical cargo-related data are imported into the data processing server for processing. Specifically, the historical cargo-related data are integrated through the probability model, and the push strategy of multimedia information is optimized through the machine learning algorithm to improve the accuracy of multimedia information. The impact of the elements of multimedia information on the users to be pushed is analyzed through the regression model multivariate regression, providing real-time data analysis and personalized multimedia information recommendation.

[0125] Furthermore, the processed data is imported into the analysis server for further analysis. Specifically, from the perspective of cargo classification, the historical cargo evaluation results are analyzed in multiple dimensions to obtain historical cargo-specific data, and the current cargo evaluation results are analyzed in multiple dimensions to obtain current cargo-specific data.

[0126] The technical solution provided by the embodiment of the present invention can be implemented to provide personalized multimedia information push for users who are in the process of selecting goods, thereby improving the accuracy and timeliness of multimedia information push, avoiding users' resistance, and improving users' usage experience; at the same time, in this solution, by acquiring current goods-related data and historical goods-related data from multiple terminals, the comprehensiveness of the acquired current goods-related data and historical goods-related data is improved; and in this solution, by introducing a probability model and a regression model, the current goods-related data and the historical goods-related data are processed respectively to obtain the current goods evaluation results and the historical goods evaluation results, thereby realizing the use of machine learning and multivariate regression analysis technology to perform in-depth analysis on the current goods-related data and the historical goods-related data, providing refined current goods evaluation results and historical goods evaluation results, and improving the accuracy and practicality of the current goods evaluation results and the historical goods evaluation results.

[0127] The solution of the embodiment of the present invention can enable users to select goods offline through multimedia information pushed online. Through personalized multimedia information push, a more accurate information push service is provided to users, the conversion rate of multimedia information is improved, users are prevented from having resistance, the accuracy and timeliness of multimedia information push are improved, and the user experience is improved.

[0128] The technical solution provided by the embodiment of the present invention can be applied in the scenario of expiring food and daily necessities management, and it can provide users with accurate multimedia information push while providing merchants with accurate goods management methods, that is, it can effectively manage the inventory of expiring food and daily necessities, increase the number of goods sold, reduce waste, and provide users with a better goods selection experience.

[0129] Embodiment 4

[0130] Figure 31 is a schematic diagram of the structure of a multimedia information push device provided in Embodiment 4 of the present invention. This embodiment is applicable to situations where multimedia information is pushed to users. The method can be executed by a multimedia push device, which can be implemented in software and / or hardware and can be configured in an electronic device that carries a multimedia information push function.

[0131] like Figure 3 As shown, the device includes: an evaluation result determination module 310, a current data determination module 320, a reference data determination module 330 and a target information push module 340.

[0132] The evaluation result determination module 310 is used to obtain the current goods related data and the current user identifier of the user to be pushed, and input the current goods related data into the trained real-time evaluation model to output the current goods evaluation result; wherein the user to be pushed is a user who is in the process of selecting goods;

[0133] The current data determination module 320 is used to process the current cargo evaluation result according to the preset cargo classification dimension to obtain the current cargo-specific data of the user to be pushed; wherein the cargo classification dimension includes cargo area dimension, user behavior dimension, multimedia information dimension and non-multimedia information dimension;

[0134] The reference data determination module 330 is used to determine whether there is a target user identifier that is the same as the current user identifier in the preset historical user identifiers. If there is a target user identifier, the historical goods special data corresponding to the target user identifier is used as the first reference goods special data according to the preset historical goods corresponding relationship.

[0135] The target information pushing module 340 is used to determine the target multimedia information to be currently pushed to the user to be pushed according to the first reference cargo-specific data and the current cargo-specific data.

[0136] An embodiment of the present invention provides a multimedia information push solution, which obtains current goods-related data and current user identification of a user to be pushed, and inputs the current goods-related data into a trained real-time evaluation model to output a current goods evaluation result; wherein the user to be pushed is a user who is in the process of selecting goods; according to a preset goods classification dimension, the current goods evaluation result is processed to obtain current goods-specific data of the user to be pushed; wherein the goods classification dimension includes a goods area dimension, a user behavior dimension, a multimedia information dimension and a non-multimedia information dimension; determines whether there is a target user identification that is the same as the current user identification in the preset historical user identification, and if there is a target user identification, then according to the preset historical goods correspondence, the historical goods-specific data corresponding to the target user identification is used as the first reference goods-specific data; according to the first reference goods-specific data and the current goods-specific data, determines the target multimedia information currently being pushed to the user to be pushed. The above scheme determines the current cargo-specific data by introducing a real-time evaluation model and cargo classification dimensions, and pushes target multimedia information to users who are in the process of selecting cargo based on the first reference cargo-specific data and the current cargo-specific data, thereby improving the timeliness and accuracy of the pushed target multimedia information, thereby meeting the user's personalized needs for multimedia information and improving the user experience of the users who are to be pushed.

[0137] Optionally, the target information push module 340 includes:

[0138] a cargo push factor determination unit, configured to determine a current cargo push factor according to the first reference cargo specific data and the current cargo specific data;

[0139] The target information determination unit is used to determine the target multimedia information from the preset candidate multimedia information according to the current cargo push factor.

[0140] Optionally, the cargo push factor determination unit is specifically used to:

[0141] Compare the first reference cargo-specific data and the current cargo-specific data belonging to the same cargo classification sub-dimension;

[0142] Determine the current cargo push factors based on the comparison results corresponding to different cargo classification sub-dimensions.

[0143] Optionally, the device further comprises:

[0144] A second data determination module is used to use the historical cargo special data corresponding to the cargo area dimension as the second reference cargo special data if the target cargo identifier does not exist;

[0145] The target information push module is used to determine the current cargo area factor according to the second reference cargo special data, and determine the target multimedia information to be currently pushed to the user to be pushed according to the current cargo area factor and the current cargo special data.

[0146] Optionally, the historical goods correspondence is determined based on the following means:

[0147] A historical evaluation result determination module is used to obtain historical goods related data and historical user identification of historical push users, input the historical goods related data into the trained historical evaluation model, and output the historical goods evaluation results;

[0148] A historical data determination module, used to process the historical goods evaluation results according to a preset goods classification dimension to obtain the historical goods special data of the historical push user;

[0149] The historical correspondence determination module is used to generate a historical cargo correspondence including a historical user identifier and corresponding historical cargo special data.

[0150] Optionally, the current goods-related data includes current behavior data, current goods data and current sensory data of the user to be pushed.

[0151] The multimedia information pushing device provided in the embodiment of the present invention can execute the multimedia information pushing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing each multimedia information pushing method.

[0152] In the technical solution of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of current goods-related data, current user identification, historical goods-related data and historical user identification, etc., all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0153] Embodiment 5

[0154] Figure 4 It is a structural diagram of an electronic device for implementing a multimedia information push method provided by Embodiment 5 of the present invention. Electronic device 410 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0155] like Figure 4 As shown, the electronic device 410 includes at least one processor 411, and a memory connected to the at least one processor 411 in communication, such as a read-only memory (ROM) 412, a random access memory (RAM) 413, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 411 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 412 or the computer program loaded from the storage unit 418 to the random access memory (RAM) 413. In the RAM 413, various programs and data required for the operation of the electronic device 410 can also be stored. The processor 411, the ROM 412, and the RAM 413 are connected to each other via a bus 414. An input / output (I / O) interface 415 is also connected to the bus 414.

[0156] Multiple components in the electronic device 410 are connected to the I / O interface 415, including: an input unit 416, such as a keyboard, a mouse, etc.; an output unit 417, such as various types of displays, speakers, etc.; a storage unit 418, such as a disk, an optical disk, etc.; and a communication unit 419, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 419 allows the electronic device 410 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0157] The processor 411 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 411 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 411 executes the various methods and processes described above, such as a multimedia information push method.

[0158] In some embodiments, the multimedia information push method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 418. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 410 via the ROM 412 and / or the communication unit 419. When the computer program is loaded into the RAM 413 and executed by the processor 411, one or more steps of the multimedia information push method described above can be performed. Alternatively, in other embodiments, the processor 411 can be configured to execute the multimedia information push method in any other appropriate manner (for example, by means of firmware).

[0159] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0160] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0161] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0162] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0163] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0164] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0165] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0166] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A multimedia information push method, characterized in that: include: Acquire the current goods-related data and the current user ID of the user to be pushed, input the current goods-related data into the trained real-time evaluation model, and output the current goods evaluation result; wherein the user to be pushed is a user who is in the process of selecting goods; According to the preset cargo classification dimensions, the current cargo evaluation results are processed to obtain the current cargo-specific data of the user to be pushed; wherein the cargo classification dimensions include cargo area dimensions, user behavior dimensions, multimedia information dimensions, and non-multimedia information dimensions; Determine whether there is a target user identifier that is the same as the current user identifier in the preset historical user identifiers. If there is a target user identifier, use the historical goods special data corresponding to the target user identifier as the first reference goods special data according to the preset historical goods corresponding relationship; According to the first reference cargo-specific data and the current cargo-specific data, target multimedia information to be currently pushed to the user to be pushed is determined.

2. The method according to claim 1, characterized in that The step of determining the target multimedia information to be currently pushed to the user to be pushed according to the first reference cargo-specific data and the current cargo-specific data includes: Determining current cargo push factors according to the first reference cargo specific data and the current cargo specific data; According to the current cargo push factor, target multimedia information is determined from preset candidate multimedia information.

3. The method according to claim 2, characterized in that The determining of the current cargo push factor according to the first reference cargo specific data and the current cargo specific data includes: Compare the first reference cargo-specific data and the current cargo-specific data belonging to the same cargo classification sub-dimension; Determine the current cargo push factors based on the comparison results corresponding to different cargo classification sub-dimensions.

4. The method according to claim 1, characterized in that: The method further comprises: If the target cargo identifier does not exist, the historical cargo special data corresponding to the cargo area dimension is used as the second reference cargo special data; The current cargo area factor is determined according to the second reference cargo-specific data, and the target multimedia information currently pushed to the user to be pushed is determined according to the current cargo area factor and the current cargo-specific data.

5. The method according to claim 1, characterized in that The historical goods correspondence is determined based on the following method: Obtain historical goods-related data and historical user identifiers of historical push users, input the historical goods-related data into a trained historical evaluation model, and output historical goods evaluation results; According to the preset cargo classification dimension, the historical cargo evaluation results are processed to obtain the historical cargo-specific data of the historical push user; Generate a historical cargo correspondence including historical user identification and corresponding historical cargo special data.

6. The method according to any one of claims 1 to 5, characterized in that The current goods-related data includes the current behavior data, current goods data and current sensory data of the user to be pushed.

7. A multimedia information push device, characterized in that: include: An evaluation result determination module is used to obtain the current goods-related data and the current user identifier of the user to be pushed, and input the current goods-related data into the trained real-time evaluation model to output the current goods evaluation result; wherein the user to be pushed is a user who is in the process of selecting goods; The current data determination module is used to process the current cargo evaluation result according to the preset cargo classification dimension to obtain the current cargo-specific data of the user to be pushed; wherein the cargo classification dimension includes cargo area dimension, user behavior dimension, multimedia information dimension and non-multimedia information dimension; A reference data determination module is used to determine whether there is a target user identifier that is the same as the current user identifier in the preset historical user identifiers. If there is a target user identifier, the historical goods special data corresponding to the target user identifier is used as the first reference goods special data according to the preset historical goods corresponding relationship. The target information pushing module is used to determine the target multimedia information to be currently pushed to the user to be pushed based on the first reference cargo-specific data and the current cargo-specific data.

8. The device according to claim 7, characterized in that The target information pushing module includes: a cargo push factor determination unit, configured to determine a current cargo push factor according to the first reference cargo specific data and the current cargo specific data; The target information determination unit is used to determine the target multimedia information from the preset candidate multimedia information according to the current cargo push factor.

9. An electronic device, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement a multimedia information push method as described in any one of claims 1-6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, a multimedia information pushing method as described in any one of claims 1 to 6 is implemented.

Images