Data entry method for marketing promotion online platform

Abstract

The application discloses a data entry method for a marketing promotion online platform, and relates to the technical field of data processing. The method comprises the following steps: acquiring marketing promotion data of various business types; analyzing the marketing promotion data to obtain a data abnormality degree; adjusting the dynamic priority of each marketing promotion data based on the data abnormality degree; and entering each marketing promotion data into an online platform according to the dynamic priority through a preset entry method. The application achieves the technical effect of accurately entering marketing promotion data.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a data entry method for online marketing and promotion platforms. Background Technology

[0002] Data entry methods for online marketing platforms involve systematically and standardizedly inputting data from marketing campaigns onto the platform, ensuring data availability and quality. This data includes information on advertising placement, user behavior, product information, and market analysis. Since different online platforms and marketing campaigns require different data entry methods, choosing the appropriate method can improve data processing efficiency and thus provide strong support for decision-making.

[0003] In related technologies, data is typically imported automatically in batches using templates or automatically integrated into a unified platform based on API (Application Programming Interface). However, in practical applications, this approach can lead to errors or fail to meet business requirements, making it difficult to adapt to complex and dynamically changing scenarios. Summary of the Invention

[0004] The main purpose of this application is to provide a data entry method for online marketing and promotion platforms, which aims to solve the technical problems in related technologies where data is automatically imported in batches through templates to a unified platform, resulting in errors or non-compliance with business needs, and making it difficult to adapt to complex and dynamically changing scenarios.

[0005] To achieve the above objectives, this application provides a data entry method for an online marketing platform, the method comprising:

[0006] Obtain marketing and promotion data for various business types;

[0007] Analyze marketing and promotion data to determine the degree of data anomalies;

[0008] Adjust the dynamic priority of each marketing and promotion data based on the degree of data anomalies;

[0009] By using preset data entry methods, marketing and promotion data are entered into the online platform according to dynamic priority.

[0010] In one possible implementation of this application, before analyzing marketing data to determine the degree of data anomaly, the following steps are included:

[0011] Obtain the output format of each marketing and promotion data, and determine the data classification strategy based on the output format;

[0012] Marketing and promotion data are categorized using data classification strategies to obtain basic data and complex data;

[0013] Analyzing marketing and promotion data reveals the degree of data anomalies, including:

[0014] Analyze complex data to determine the degree of data anomalies.

[0015] In one possible implementation of this application, a data classification strategy is determined based on the output format, including:

[0016] Determine the data channels and data platforms corresponding to the marketing and promotion data for the current business type;

[0017] Based on the output format, data channel, and data platform, a data segmentation index is calculated.

[0018] Based on the data partitioning index, determine the data classification strategy.

[0019] In one possible implementation of this application, marketing promotion data is classified using a data classification strategy to obtain basic data and complex data, including:

[0020] The data is divided into indices and normalized to obtain normalization coefficients;

[0021] If the normalization coefficient is less than the preset threshold, the marketing promotion data corresponding to the normalization coefficient will be classified as basic data.

[0022] If the normalization coefficient is greater than or equal to the preset threshold, the marketing promotion data corresponding to the normalization coefficient will be classified as complex data.

[0023] In one possible implementation of this application, marketing promotion data is analyzed to determine the degree of data anomaly, including:

[0024] Anomalies in marketing promotion data are detected by using the support vector machine algorithm, and the change patterns of the corresponding time series data are detected by using a preset long short-term memory model to determine the scale of abnormal data.

[0025] Based on the scale of the abnormal data, determine the degree of data abnormality in the current marketing and promotion data.

[0026] In one possible implementation of this application, the degree of data anomaly includes the degree of formal anomaly. Based on the degree of data anomaly, the dynamic priority of each marketing promotion data is adjusted, including:

[0027] Based on the degree of formal anomaly and the statistical frequency of current marketing and promotion data, a priority screening coefficient is calculated.

[0028] The intervention priority is calculated based on the priority screening coefficient and the data timeliness value;

[0029] The speed of decision-making regarding marketing and promotion data is improved, and the dynamic priority of each marketing and promotion data point is adjusted based on intervention priority and decision-making speed.

[0030] In one possible implementation of this application, a priority screening coefficient is calculated based on the degree of formal anomaly and the statistical frequency of current marketing promotion data, including:

[0031] Obtain the number of channels corresponding to the data channels for the current marketing and promotion data;

[0032] Priority screening coefficients are calculated based on the number of channels, the degree of anomalies in the form, and the statistical frequency.

[0033] In one possible implementation of this application, marketing and promotion data are entered into an online platform according to dynamic priority using a preset data entry method, including:

[0034] If the marketing and promotion data is determined to be the basic data, then the promotion data will be entered into the online platform through an automatic data entry method;

[0035] If the marketing and promotion data is determined to be complex, then a manual decision-making method will be used to enter each marketing and promotion data into the online platform according to a dynamic priority order.

[0036] In one possible implementation of this application, after entering the marketing and promotion data into the online platform according to dynamic priority, the method further includes:

[0037] Compare the intervention priority corresponding to each marketing promotion data with the preset priority threshold;

[0038] If the comparison results show that the intervention priority is greater than the preset priority threshold, the current marketing promotion data entry results will be reviewed a second time.

[0039] If the review results show that the data entry is accurate, no data entry modification will be made; otherwise, the data entry decision will be re-executed.

[0040] In one possible implementation of this application, after entering marketing and promotion data into the online platform according to dynamic priority using a preset data entry method, the method further includes:

[0041] Obtain feedback data corresponding to human decision-making methods;

[0042] Based on feedback data and decision results, the automated data entry method is optimized and trained until no human intervention is required.

[0043] This application provides a data entry method for online marketing platforms. Compared with related technologies that automatically import data in batches through templates to a unified platform, which can lead to errors or situations that do not meet business needs and are difficult to adapt to complex and dynamically changing scenarios, this application obtains marketing data from various business types and analyzes the data to determine the degree of data anomalies. The dynamic priority of the marketing data is then adjusted based on these anomaly levels. Subsequently, various types of data are entered into the online platform step by step according to the dynamic priority. Compared with the automatic batch import method, the method of adjusting the dynamic priority can adapt to complex and dynamically changing scenarios, thereby achieving the technical effect of accurate data entry for marketing promotion. Attached Figure Description

[0044] Figure 1 This is a flowchart illustrating the first embodiment of the data entry method for an online marketing platform according to this application.

[0045] Figure 2 This is a schematic diagram illustrating the process division involved in the data entry method for online marketing and promotion platforms in this application;

[0046] Figure 3 This is a flowchart illustrating the second embodiment of the data entry method for online marketing platforms according to this application;

[0047] Figure 4 This is a flowchart illustrating the third embodiment of the data entry method for online marketing platforms involved in this application;

[0048] Figure 5 This is a schematic diagram illustrating the overall execution process of the data entry method for online marketing and promotion platforms in this application.

[0049] Figure 6 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of this application. Detailed Implementation

[0050] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0051] This application provides a data entry method for an online marketing platform. In the first embodiment of this application's data entry method for an online marketing platform, refer to... Figure 1 The methods include:

[0052] Step S10: Obtain marketing and promotion data for various business types;

[0053] Step S20: Analyze the marketing and promotion data to determine the degree of data anomalies;

[0054] Step S30: Adjust the dynamic priority of each marketing and promotion data based on the degree of data anomaly;

[0055] Step S40: Using a preset data entry method, the marketing and promotion data are entered into the online platform according to dynamic priority.

[0056] This embodiment aims to accurately input marketing and promotion data when receiving data of different business types.

[0057] In this embodiment, the research and development background is as follows:

[0058] In related technologies, online platforms for marketing and promotion require data entry. This is typically achieved through automatic batch import of templates or by automatically integrating the required data into a unified platform via API interfaces. However, while fully automated data entry can improve efficiency, it can also lead to errors or situations that don't meet business needs. Automated data entry struggles to adapt to complex and dynamically changing scenarios. Therefore, manual intervention is necessary. However, complete manual intervention can result in inefficiency and high costs.

[0059] In this embodiment, the application scenario is as follows:

[0060] By analyzing the complexity and anomalies of marketing and promotion data, intelligent data recommendations are made. This is combined with the intervention patterns of the manual review mechanism to adjust dynamic priorities, thereby ensuring accurate entry of marketing and promotion data through a combination of automatic data entry and manual review.

[0061] The specific steps are as follows:

[0062] Step S10: Obtain marketing and promotion data for various business types;

[0063] As an example, the data entry method for an online marketing platform can be applied to a data entry device for an online marketing platform. The data entry device for an online marketing platform belongs to a data entry system for an online marketing platform, and the data entry system for an online marketing platform belongs to a data entry equipment for an online marketing platform.

[0064] As an example, the data entry method for marketing and promotion online platforms can also be applied to online platforms, including common types of marketing and promotion online platforms such as search engine marketing, social media marketing, video platform advertising, and e-commerce platform advertising.

[0065] As an example, marketing promotion data can be the marketing data needed for promotion on a platform. It can be obtained from different platforms, such as clicks, interactions, and audience profiles. The business types of marketing promotion data also differ for different platforms.

[0066] As an example, customer data is acquired in the following ways on different platforms:

[0067] Search engine platforms: Acquire data such as ad impressions, clicks, conversion rates, and ROI.

[0068] Social media platforms: Acquire data such as clicks, interactions, reach, audience profiles, and behaviors.

[0069] Video platform advertising: Acquire data such as viewership, view rate, viewing duration, and discovery channels.

[0070] E-commerce platform advertising: Acquire data such as click volume, cost per click, number of transactions, return on investment, product add-to-cart rate and collection rate, and number of store visitors.

[0071] Data statistics can be collected by using all the promotional and marketing data or the actual marketing effects of advertisements that the client can provide. Marketing promotion data also has corresponding attribute characteristics in marketing activities, such as life cycle, duration of effect and commercial value. These attribute characteristics are also collected in the process of acquiring marketing promotion data.

[0072] Step S20: Analyze the marketing and promotion data to determine the degree of data anomalies.

[0073] As an example, the degree of data anomaly can be the degree of anomaly in marketing and promotion data. For example, in various types of marketing and promotion data, there may be format errors, trend conflicts, and marginal case data. These are all types of abnormal data. The more errors there are in the data or the greater the deviation from normal data, the higher the degree of data anomaly.

[0074] As an example, since both automatic and manual data entry methods are difficult to adapt to complex scenarios and are inefficient, this solution analyzes marketing data to determine whether it is basic or complex data. Basic data can be directly entered automatically, while complex data may have various anomalies. The data is then prioritized based on these anomalies, and a human decision-making mechanism is incorporated into the data entry process. This combines automatic and manual data entry methods to achieve accurate data entry.

[0075] Before step S20, which involves analyzing marketing and promotion data to determine the degree of data anomalies, the process also includes:

[0076] Deduplication and standardization processing are performed on marketing and promotion data.

[0077] As an example, deduplication and standardization are preprocessing steps for data, also known as data cleaning. After obtaining relevant marketing and promotion data, these data are preprocessed by performing operations such as deduplication (deleting duplicate data records) and standardization (unifying data formats, such as date formats and phone number formats), and necessary measures are taken to protect data security and prevent data leakage and misuse.

[0078] The data entry method for online marketing platforms includes steps S11 to S12 before step S20:

[0079] Step S11: Obtain the output format of each marketing promotion data, and determine the data classification strategy based on the output format.

[0080] As an example, the output format can be based on different data platform and channel sources. It can be a platform-channel output format. The channels and platforms from which different data are obtained will vary, and the same type of data may originate from multiple different channels. Therefore, there is a certain tree-like relationship between data, channels, and platforms, as detailed below. Figure 2 As shown, taking data 2 as an example, data 2 exists in both channel a and channel c in platform 1, and also in channel b in platform 2. Therefore, for a certain type of data, it may exist in different channels or platforms. Thus, it is necessary to classify it. After determining the data source and classifying it, a data foundation is provided for subsequent data entry, which improves the accuracy of data entry.

[0081] As an example, a data classification strategy can be a classification method calculated based on the output format of the data and the input cost. Basic data is usually the basic operational statistics of the platform and channels, which reflects the marketing situation of the corresponding platform and channel itself. It can be operated directly through automatic entry. Complex data can be integrated and entered across platforms and channels. Complex data involves the analysis of user behavior, the calculation of advertising return on investment, and the prediction of long-term trends. The input cost needs to be taken into account. Complex data may be distributed across different channels on different platforms. Therefore, we can analyze the distribution breadth of a certain type of data across different platforms and channels to determine its classification. Therefore, for different data, we need to determine its data classification index, and then determine whether the current data is classified as complex data or basic data based on the data classification index.

[0082] Step S12: Classify the marketing and promotion data using a data classification strategy to obtain basic data and complex data.

[0083] As an example, basic data is usually the basic operational statistics of the platform and channels, which reflects the marketing situation of the corresponding platform and channel itself, and can be operated directly through automatic data entry.

[0084] As an example, complex data can be integrated and entered across platforms and channels, meaning that complex data may be distributed across different platforms and channels. This application mainly focuses on processing the entry of complex data, while classification is intended to reduce the workload of subsequent manual data review and improve data entry efficiency.

[0085] Analyzing marketing and promotion data reveals the degree of data anomalies, including:

[0086] Analyze complex data to determine the degree of data anomalies.

[0087] As an example, the basic data mainly consists of marketing statistics from the platform, which is relatively simple and rarely shows anomalies. Therefore, the degree of data anomaly is only obtained by analyzing complex data, which can reduce the amount of calculation.

[0088] The steps for determining the data classification strategy based on the output format include:

[0089] Determine the data channels and data platforms corresponding to the marketing and promotion data for the current business type.

[0090] As an example, data channels can be channel 1 / 2 / 3. For a data platform, there may be different data output channels, and a type of data can exist on different data platforms or data channels.

[0091] As an example, a data platform, also known as an online marketing platform, offers different functions and services to help businesses or individuals reach their target audience, increase brand awareness, and promote the sales of products or services.

[0092] The data segmentation index is calculated based on the output format, data channel, and data platform.

[0093] As an example, the output format of data itself can measure its complexity. Generally, the lower the output format obtained directly from data statistics, the higher the output format of decision support data obtained after data analysis and mining. Therefore, the calculation method for the data partitioning index can be:

[0094]

[0095] Where Z a Divide the data of a certain data type 'a' into exponents. (C) a SC represents the number of channels with data type 'a' across all platforms. a p represents the total number of channels across all platforms. a Let p0 be the number of platforms with data type a, and p0 be the total number of platforms. a The output form coefficients for data type a (first-order data is I) a Take 1, n times data I a Take n).

[0096] As an example, the output form coefficient is used to quantify the complexity of data. For example, data such as clicks and views, which can be directly generated or statistically obtained, has an output form coefficient of 1. For data such as return on investment or audience profiles, which require multiple calculations in combination with other data, the output form coefficient is determined by the number of times the data is calculated. For example, if return on investment is calculated 5 times, then the complexity of this type of data is higher than that of clicks, and its output form coefficient is 5.

[0097] As an example, This represents the widespread presence of data type 'a' across different platforms and their corresponding channels. A larger value indicates that this data type is displayed and promoted through more platforms and channels, thus making the corresponding data more complex. This is also considered in conjunction with the data's output format, I. a The more complex the output format, the better. a The larger the value, the larger the corresponding dividing index.

[0098] Based on the data partitioning index, determine the data classification strategy.

[0099] As an example, after calculating the data segmentation index, the data segmentation index is compared with a preset threshold. Based on the comparison result, it is determined whether various types of data are classified as basic data or complex data, thus determining the data classification strategy.

[0100] The marketing and promotion data is categorized using a data classification strategy to obtain basic data and complex data, including:

[0101] The data is divided into indices and then normalized to obtain normalized coefficients.

[0102] As an example, the normalization process involves normalizing the data partition index to the range of (0,1) to obtain the normalization coefficient. The normalization coefficient can be 0.2, 0.4, etc., and there is no specific limitation.

[0103] If the normalization coefficient is less than the preset threshold, the marketing promotion data corresponding to the normalization coefficient will be classified as basic data.

[0104] As an example, the preset threshold can be 0.5, 0.7, or other data. Here, we take a preset threshold of 0.7 as an example. When the normalization coefficient is less than 0.7, the marketing promotion data corresponding to the current normalization coefficient is classified as basic data.

[0105] If the normalization coefficient is greater than or equal to the preset threshold, the marketing promotion data corresponding to the normalization coefficient will be classified as complex data.

[0106] As an example, similarly, when the normalization coefficient is greater than or equal to 0.7, the marketing data corresponding to the normalization coefficient is classified as complex data. Basic data can be entered into the online marketing platform through automated data entry methods, while complex data will be accurately entered later through manual intervention and analysis.

[0107] Step S30: Adjust the dynamic priority of each marketing and promotion data based on the degree of data anomaly.

[0108] As an example, current automated systems typically operate based on fixed processes and rules. However, when faced with anomalies or unusual situations in complex data—such as the degree of data anomaly in marketing and promotion data—where flexible judgment is required, automation struggles to adapt to complex and dynamically changing scenarios. Therefore, human decision-making needs to be introduced, with a priority given to determining the level of human intervention. The remaining normal data continues to be entered through the automated system. Here, we first determine the priority of the target data requiring human intervention within the complex dataset.

[0109] As an example, dynamic priority is calculated from intervention priority. During the acquisition of various marketing and promotional data, the number of channels corresponding to low-frequency data found in different searches is counted. For low-frequency marketing data, the lower the frequency, the more likely the data belongs to the data edge case category. Furthermore, the more times low-frequency data appears in different channels, the higher the cross-platform risk, and the higher the priority requiring manual intervention. The intervention priority of the current marketing and promotional data is then determined based on the degree of data anomaly. Furthermore, adjusting the dynamic priority according to the actual situation of the data ensures accurate data entry even in complex and ever-changing circumstances.

[0110] Step S40: Using a preset data entry method, the marketing and promotion data are entered into the online platform according to dynamic priority.

[0111] As an example, the preset data entry methods include automatic entry and manual decision-making. For basic data, the automatic entry method is used to enter marketing and promotion data into the online platform. For complex data, the manual decision-making method is used to enter marketing and promotion data into the online platform according to the dynamic priority of each data.

[0112] This application provides a data entry method for online marketing platforms. Compared with related technologies that automatically import data in batches through templates to a unified platform, which can lead to errors or situations that do not meet business needs and are difficult to adapt to complex and dynamically changing scenarios, this application obtains marketing data from various business types and analyzes the data to determine the degree of data anomalies. The dynamic priority of the marketing data is then adjusted based on these anomaly levels. Subsequently, various types of data are entered into the online platform step by step according to the dynamic priority. Compared with the automatic batch import method, the method of adjusting the dynamic priority can adapt to complex and dynamically changing scenarios, thereby achieving the technical effect of accurate data entry for marketing promotion.

[0113] Furthermore, referring to Figure 3 Based on the first embodiment of this application, another embodiment of this application is provided. In this embodiment, the data entry step S20 for the online marketing platform includes:

[0114] Step S21: Anomalies in marketing promotion data are detected using the support vector machine algorithm, and the change patterns of the corresponding time series data are detected using a preset long short-term memory model to determine the location of data anomalies.

[0115] As an example, in this embodiment, marketing promotion data requiring manual intervention often suffers from data format errors, missing or duplicate data, and abrupt changes in time-series data trends. Directly importing this abnormal data automatically may affect data quality and the accuracy of decision-making results. Furthermore, during promotion across multiple channels and platforms, data loss or missing information may occur. Therefore, the more severe the data loss when transmitting data to different platforms, the more abnormal the form of such data is.

[0116] As an example, the SVM (Support Vector Machine) algorithm is used to detect data anomalies, and the LSTM (Long Short-Term Memory) model is used to detect the changing patterns of time series data, identify the location of the abnormal data, and mark it. Then, the abnormal data is statistically analyzed to obtain the scale of the abnormal data.

[0117] Step S22: Determine the degree of data anomaly in the current marketing and promotion data based on the scale of the abnormal data.

[0118] As an example, the scale of abnormal data can be the number of various abnormal data in the marketing and promotion data, or it can be the number of bytes. After determining the abnormal location of the marketing and promotion data, the abnormal data can be located through the abnormal location, thereby calculating the degree of data abnormality of the currently calculated marketing and promotion data.

[0119] As an example, the degree of data anomaly can be formal anomaly, which exists for different types of target data:

[0120]

[0121] Wherein, γ a G represents the degree of formal anomalousness of a certain type of target data 'a'. a For the total size of this type of data, x a This refers to the scale of anomalous data within this type of data. This represents the average loss rate of such data transmitted to different platforms.

[0122] As an example, This represents the degree of format anomalousness of such data. The higher the value, the more anomalous data there is, and the greater the anomalousness. It is also considered in conjunction with the average loss scale of such data transmitted to different platforms. The larger the scale of data loss, the greater the degree of data anomaly.

[0123] The data anomaly level includes the degree of formal anomaly. Step S30, which adjusts the dynamic priority of various marketing and promotion data based on the degree of data anomaly, includes:

[0124] Step S31: Calculate the priority screening coefficient based on the degree of anomaly and the statistical frequency of current marketing promotion data.

[0125] As an example, some marketing data may represent certain meanings (such as channel discovery, advertising keywords, etc.). This data may not exhibit sudden trend changes, but it may contain marginal cases—anomalies where the data is meaningless. However, this data still falls under the target data to be entered, requiring manual evaluation to decide whether to enter it. Marketing data is displayed in a tree structure, while marginal cases generally appear infrequently within this structure. Therefore, the following applies:

[0126] 1. Traverse the tree structure to determine the frequency of different types of data to identify possible low-frequency data (frequency less than 0.05).

[0127] 2. During the traversal process, count the number of low-frequency data points found in different channels.

[0128] The step S31, which calculates the priority screening coefficient based on the degree of formal anomaly and the statistical frequency of current marketing promotion data, includes:

[0129] Obtain the number of channels corresponding to the data channels for the current marketing and promotion data.

[0130] As an example, the same type of marketing data can exist in multiple data channels. Counting the number of data channels where this type of marketing data is located can also reflect the frequency of the occurrence of the marketing data.

[0131] Priority screening coefficients are calculated based on the number of channels, the degree of anomalies in the form, and the statistical frequency.

[0132] As an example, statistical frequency can be the frequency of data occurrence obtained after statistical analysis of marketing and promotion data. The statistical frequency can be 0.05, 0.2, etc., and there is no specific limitation.

[0133] As an example, for low-frequency marketing data, the lower the frequency, the more likely the data is to be considered a marginal case. Furthermore, the more instances of low-frequency data appear across different channels, the higher the cross-platform risk and the higher the priority for manual intervention. Therefore, considering its anomaly profile, for a specific target data point (the currently calculated marketing promotion data), the following anomalies exist:

[0134]

[0135] Wherein, H a Q represents the priority filtering coefficient for a certain type of target data 'a'. a D represents the number of channels containing target data a in the tree structure. a γ represents the statistical frequency of the target data a. a The degree of anomalousness in the form of a certain type of target data 'a'.

[0136] Step S32: Calculate the intervention priority based on the priority screening coefficient and the data timeliness value.

[0137] As an example, a higher priority screening coefficient indicates that the corresponding data is more likely to be a marginal case, thus requiring a higher priority for manual intervention. The timeliness of the data itself also directly impacts priority determination. Therefore, for any target data, its intervention priority can be expressed as:

[0138]

[0139] Where, Y a Intervention priority for target data a. H a Tl is the target selection coefficient for target data a.a The timeliness of target data 'a', also known as the data timeliness value, is determined by the severity of any anomalies in this type of data, i.e., H. a The larger the value, the shorter the time-limited period, i.e., Tl a The smaller the value, the higher the priority of requiring human intervention. Finally, the calculation results of the intervention priority are normalized to the range of [0,1].

[0140] As an example, data with high intervention priority is first pushed to the human decision-making system, from which labeled data is selected, and then a human decision is made on whether to enter or modify it. Non-target data can be entered automatically.

[0141] Step S33: Obtain the decision speed of marketing promotion data, and adjust the dynamic priority of each marketing promotion data based on the intervention priority and decision speed.

[0142] As an example, after determining the intervention priority, the human decision-making system will execute the corresponding input decision for each data according to the size of the intervention priority. Different target data will have different decision speeds. Therefore, it is necessary to first determine the decision speed of each type of data and then calculate the corresponding dynamic priority. The dynamic priority is also the dynamic intervention priority. Since human decision-making may have differences in decision speed for different types of data (such as the decision speed of return on investment is faster than that of audience profile data), if the data volume is large, in order to improve the efficiency of human decision-making, the priority ranking is adjusted by the feedback of the decision speed of different data.

[0143] As an example, for data with faster decision-making speeds, the intervention priority for data with faster decision-making speeds needs to be increased. When calculating dynamic priorities, it is as follows:

[0144]

[0145] Where, Y′ a This refers to the dynamic intervention priority for target data 'a', also known as dynamic priority. Y a For the intervention priority of target data a, v a For the decision speed of target data a, Average decision speed for different types of target data.

[0146] So This represents the relative decision-making speed of target data 'a', and is used as the weight. This type of data has a higher decision-making speed than the average decision-making speed at the initial intervention. The larger the value, the higher its priority will be if there are subsequent decisions regarding this type of data, i.e., Y′. aThe larger the value, the lower the priority; conversely, the smaller the value, the lower the priority. Therefore, by adjusting the priority of different types of data based on historical decision-making speed feedback using the above method, the overall decision-making efficiency of the data can be improved.

[0147] In this embodiment, the priority of historical decision-making speed feedback for different types of data is adjusted to improve the overall data decision-making efficiency.

[0148] Furthermore, referring to Figure 4 Based on the first and second embodiments of this application, another embodiment of this application is provided. In this embodiment, step S40, which involves entering marketing and promotion data into the online platform according to dynamic priority using a preset entry method, includes:

[0149] Step S41: If the marketing and promotion data is determined to be the basic data, the promotion data is entered into the online platform through an automatic entry method.

[0150] As an example, when determining marketing and promotion data as the basic data, it means that this data is standardized and has clearly defined rules, and can be automatically entered into the online platform.

[0151] Step S42: If the marketing promotion data is determined to be complex data, then the marketing promotion data will be entered into the online platform according to the dynamic priority order through manual decision-making.

[0152] As an example, when marketing and promotion data is complex, human decision-making methods are needed to help the automated system make more accurate judgments. In this case, automated data entry and human decision-making data entry can be balanced during data entry. Then, based on the dynamic priority order and human decision-making methods, each marketing and promotion data can be entered into the online platform.

[0153] After step S42, which involves entering marketing and promotion data into the online platform according to dynamic priority, the process also includes:

[0154] Compare the intervention priority corresponding to each marketing promotion data with the preset priority threshold.

[0155] As an example, the preset priority threshold can be 0.7, 0.8, etc., and there is no specific limitation.

[0156] As an example, after calculating the intervention priority, the method of manual decision-making can be determined according to the priority of manual intervention required by the target data. However, some data scenarios are very complex, involving knowledge from multiple fields. Manual decision-making is often affected by the personal experience, limitations, or negligence of employees, resulting in inconsistent fairness and rationality of the decision results. If some important data is not rationally decided, it may affect the overall marketing effect. Therefore, it is necessary to conduct a secondary review of marketing data with higher intervention priority. Thus, it is necessary to detect and compare the intervention priority to ensure that the data entry results will not be affected when the decision is unreasonable.

[0157] If the comparison results show that the intervention priority is greater than the preset priority threshold, the current marketing promotion data entry results will be subject to a second review.

[0158] As an example, taking a preset priority threshold of 0.88 as an example, when the intervention priority is greater than or equal to 0.88, the current marketing promotion data entry results will be reviewed a second time. The entry results here can also be decision results. In order to avoid inaccurate entry, a second review is required to ensure the accuracy and rationality of the data entry.

[0159] If the review results show that the data entry is accurate, no data entry modification will be made; otherwise, the data entry decision will be re-executed.

[0160] As an example, after a second review, the review results will be displayed. If the review results show that the current data entry is accurate, there is no need to operate on the current data again, and there is no need to make any changes. Otherwise, the manual data entry decision needs to be re-executed.

[0161] The process, following step S40 which involves inputting marketing and promotion data into the online platform using a preset input method and dynamic priority, also includes:

[0162] Obtain feedback data corresponding to human decision-making methods.

[0163] As an example, current human decision-making methods are intended to make data entry more accurate, but not to be used indefinitely. After multiple data entries, feedback data from human intervention can be continuously collected to optimize intelligent priority intervention methods.

[0164] Based on feedback data and decision results, the automated data entry method is optimized and trained until no human intervention is required.

[0165] As an example, the decision result can be the input result after human decision-making. The automated input method can be trained by feedback data and decision results, thereby reducing the frequency of human intervention, and gradually achieving complete and accurate automation until no human intervention strategy is used.

[0166] In this embodiment, the automated data entry method is trained by using feedback data and decision results corresponding to the manual decision-making method, thereby reducing the frequency of manual intervention and achieving a fully and accurately automated data entry method.

[0167] Specifically, the overall implementation flowchart of this application is as follows: Figure 5 As shown, by Figure 5 Therefore, firstly:

[0168] 1. Preprocess different types of marketing and promotion data obtained from corresponding customers and platforms;

[0169] 2. Determine the corresponding classification strategy based on the output format and cost of marketing promotion data;

[0170] 3. Calculate the degree of anomaly based on the anomalies in complex data;

[0171] 4. Determine the intervention priority based on the degree of data anomalies and the frequency of marginal cases;

[0172] 5. Establish corresponding human decision-making mechanisms and adjust dynamic priorities based on human intervention feedback;

[0173] 6. Combine automatic data entry with manual decision-making to accurately input marketing and promotion data into the online platform.

[0174] After the data entry is completed, the automatic data entry method is continuously optimized based on feedback data from human decision-making until fully and accurately automated data entry is achieved.

[0175] Reference Figure 6 , Figure 6 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of this application.

[0176] like Figure 6 As shown, the data entry device for the online marketing platform may include: a processor 1001, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to enable communication between the processor 1001 and the memory 1005.

[0177] Optionally, the data entry device for the online marketing platform may also include a user interface, a network interface, a camera, RF (Radio Frequency) circuitry, sensors, a WiFi module, etc. The user interface may include a display screen and an input submodule such as a keyboard; optional user interfaces may also include standard wired or wireless interfaces. The network interface may include standard wired or wireless interfaces (such as a Wi-Fi interface).

[0178] Those skilled in the art will understand that Figure 6 The data entry device structure shown in the diagram for an online marketing platform does not constitute a limitation on the data entry device for an online marketing platform. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0179] like Figure 6 As shown, the memory 1005, serving as a storage medium, may include an operating system, a network communication module, and a data entry program for an online marketing platform. The operating system is a program that manages and controls the hardware and software resources of the data entry device for the online marketing platform, supporting the operation of the data entry program and other software and / or programs. The network communication module is used to enable communication between the various components within the memory 1005, as well as communication with other hardware and software in the data entry system for the online marketing platform.

[0180] exist Figure 6 In the data entry device for the marketing and promotion online platform shown, the processor 1001 is used to execute the data entry program for the marketing and promotion online platform stored in the memory 1005 to implement the steps of the data entry method for the marketing and promotion online platform described above.

[0181] The specific implementation method of the data entry device for the marketing and promotion online platform in this application is basically the same as the above-described embodiments of the data entry method for the marketing and promotion online platform, and will not be repeated here.

[0182] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0183] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0184] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods of the various embodiments of this application.

[0185] The above are merely preferred embodiments of this application and do not limit the scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the scope of protection of this application.

[0186] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0187] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A data entry method for an online marketing and promotion platform, characterized in that, The method includes: Obtain marketing and promotion data for various business types; The marketing and promotion data is analyzed to determine the degree of data anomaly, including analyzing complex data to determine the degree of data anomaly. Based on the degree of data anomaly, the dynamic priority of each marketing promotion data is adjusted, including: obtaining the number of channels corresponding to the data channels where the current marketing promotion data exists; calculating a priority screening coefficient based on the number of channels, the degree of data anomaly, and the statistical frequency; calculating an intervention priority based on the priority screening coefficient and the data timeliness value; obtaining the decision speed of the marketing promotion data, and adjusting the dynamic priority of each marketing promotion data based on the intervention priority and the decision speed; wherein, the statistical frequency represents the frequency of data occurrence after statistical analysis of the marketing promotion data; and the data timeliness value represents the timeliness of the data. The marketing and promotion data are entered into the online platform according to the preset input method and the dynamic priority. Before analyzing the marketing promotion data to determine the degree of data anomaly, the process includes: obtaining the output format of each marketing promotion data item and determining the data channel and data platform corresponding to the marketing promotion data of the current business type; calculating a data segmentation index based on the output format, data channel, and data platform; determining a data classification strategy based on the data segmentation index; normalizing the data segmentation index to obtain a normalization coefficient; if the normalization coefficient is less than a preset threshold, classifying the marketing promotion data corresponding to the normalization coefficient as basic data; if the normalization coefficient is greater than or equal to the preset threshold, classifying the marketing promotion data corresponding to the normalization coefficient as complex data.

2. The data entry method for an online marketing platform as described in claim 1, characterized in that, The analysis of the marketing and promotion data to determine the degree of data anomalies includes: Anomalies in the marketing and promotion data are detected by using a support vector machine algorithm, and the change patterns of the corresponding time-series data are detected by using a preset long short-term memory model to determine the scale of abnormal data. Based on the scale of the abnormal data, the degree of data abnormality in the current marketing and promotion data is determined.

3. The data entry method for an online marketing platform as described in claim 1, characterized in that, The step of entering the marketing and promotion data into the online platform according to the preset entry method and the dynamic priority includes: If the marketing and promotion data is determined to be the basic data, then the promotion data is entered into the online platform using an automatic data entry method; If the marketing promotion data is determined to be complex, then through a manual decision-making method, each piece of marketing promotion data is entered into the online platform according to the dynamic priority.

4. The data entry method for an online marketing platform as described in claim 3, characterized in that, After entering the marketing and promotion data into the online platform according to the dynamic priority, the process further includes: Compare the intervention priority corresponding to each of the aforementioned marketing promotion data with the preset priority threshold; If the comparison results show that the intervention priority is greater than the preset priority threshold, the current marketing promotion data entry results will be reviewed a second time. If the review results show that the data entry is accurate, no data entry modification will be made; otherwise, the data entry decision will be re-executed.

5. The data entry method for an online marketing platform as described in claim 3, characterized in that, After entering the marketing and promotion data into the online platform according to the dynamic priority using the preset input method, the method further includes: Obtain feedback data corresponding to human decision-making methods; Based on the feedback data and decision results, the automated data entry method is optimized and trained until no manual intervention is required.

Images