Product marketing management system and method based on big data

Abstract

The invention provides a product marketing management system and method based on big data, and relates to marketing big data management. According to the invention, through comprehensive collection and standardization processing of marketing data and quantitative evaluation of the relevance of marketing indexes by means of a correlation coefficient analysis method, three levels of relevance of strong, medium and weak are divided, and potential rules and problems among marketing indexes are revealed for enterprises; secondly, first-level indexes with strong association are further screened out and visually presented, so that an enterprise can focus on core indexes to carry out key management; meanwhile, hidden associated data such as user comment emotional tendency and market trend inflection points are mined by utilizing natural language processing and a long-short term memory network technology, and deeper market insight is provided for marketing strategy adjustment; finally, based on implicit associated data, evaluation indexes are screened out, a periodic change report is generated, an enterprise is assisted to visually grasp key index dynamic and implicit factor influences, and good support is provided for optimizing a decision process.

Description

Technical Field

[0001] The present invention relates to the technical field of marketing big data management, and particularly to a product marketing management system and method based on big data. Background Art

[0002] In today's digital and information-based business wave, enterprises are facing an increasingly complex and dynamic market environment. In order to stand out in the fierce market competition and achieve sustainable business growth, enterprises urgently need to build an efficient and accurate product marketing management system. However, in the current process of product marketing management by enterprises, a series of technical problems that need to be solved urgently are generally faced: When enterprises conduct product marketing management, they usually focus on multiple marketing indicators, such as sales volume, conversion rate, customer acquisition cost, and market share, etc. However, due to the lack of scientific and effective analysis methods, it is difficult for enterprises to comprehensively and deeply understand the internal correlations between these indicators. The vague understanding of the correlations between indicators makes enterprises often lack systematicness when formulating marketing strategies, and it is difficult to accurately grasp the mutual influence relationships between various indicators, which may lead to waste of marketing resources and unsatisfactory marketing effects.

[0003] Among numerous marketing indicators, enterprises need to clarify which indicators have the most critical impact on the business, so as to concentrate resources and efforts on key management and optimization. However, due to the lack of effective indicator screening and evaluation methods, enterprises often have difficulty accurately identifying the truly critical primary indicators. This may lead to the lack of clear goals and directions in the marketing management process of enterprises, and it is difficult to achieve the optimal allocation of resources.

[0004] When enterprises formulate marketing strategies, they often can only make decisions based on limited information. The existing marketing data management is difficult to provide enterprises with data support for accurately grasping market dynamics and user needs, which may lead to the lag and blindness of marketing strategies.

[0005] Therefore, it is necessary to provide a product marketing management system and method based on big data to solve the above technical problems. Summary of the Invention

[0006] To solve the above technical problems, the present invention provides a product marketing management system and method based on big data to solve the problems of insufficient analysis of the correlations between marketing indicators, difficulty in identifying key indicators, and inaccurate and incomplete marketing data management in the process of product marketing management.

[0007] A product marketing management system based on big data provided by the present invention includes: The management system includes: A data collection and processing module, which is used to establish a periodic data collection mechanism by integrating multi-source systems, periodically collect product marketing data for preprocessing, and form a structured data set; An association analysis module, which is used to extract various marketing indicators from the structured data set and calculate the correlation coefficients between the marketing indicators by using the correlation coefficient analysis method; A grading module, which is used to divide the index correlation into three association levels: strong level, medium level, and weak level based on the size of the correlation coefficient according to multiple preset threshold ranges, and obtain the marketing index association level classification result; A data identification module, which is used to screen out the marketing indicators with a strong association level based on the marketing index association level classification result and identify them as first-level indicators; An analysis and processing module is used to analyze and identify implicit association data from the structured data set through natural language processing technology and long short-term memory network data processing technology, where the implicit association data includes: user comment sentiment tendency and market trend inflection points; A report generation module, which is used to screen out the first-level indicators associated with the product's implicit data from the first-level indicators based on the obtained implicit association data, identify them as evaluation indicators, and generate a periodic change report at the same time.

[0008] A product marketing management method based on big data, the management method includes the following steps: Establish a periodic data collection mechanism by integrating multi-source systems, periodically collect product marketing data for preprocessing, and form a structured data set; Extract various marketing indicators from the structured data set and calculate the correlation coefficients between the marketing indicators by using the correlation coefficient analysis method; Based on the size of the correlation coefficient, divide the index correlation into three association levels: strong level, medium level, and weak level according to multiple preset threshold ranges, and obtain the marketing index association level classification result; Based on the marketing index association level classification result, screen out the marketing indicators with a strong association level and identify them as first-level indicators; Analyze and identify implicit association data from the structured data set through natural language processing technology and long short-term memory network data processing technology, where the implicit association data includes: user comment sentiment tendency and market trend inflection points; Based on the obtained implicit association data, screen out the first-level indicators associated with the product's implicit data from the first-level indicators, identify them as evaluation indicators, and generate a periodic change report at the same time.

[0009] Preferably, the specific steps of establishing a periodic data collection mechanism by integrating multi-source systems, periodically collecting product marketing data for preprocessing, and forming a structured data set include: Integrate multi-source data systems inside and outside the enterprise, including CRM systems, e-commerce platforms, social media platforms or market research databases, and establish a unified data interface for data synchronization; According to the preset collection period, regularly collect product marketing data, including sales data, user behavior data, and market promotion data; Remove duplicate values, missing values, and outliers from the collected product marketing data, and perform data standardization processing to convert it into a structured data set.

[0010] Preferably, the specific steps of extracting various marketing indicators from the structured data set and calculating the correlation coefficients between the marketing indicators using the correlation coefficient analysis method include: Extract marketing indicators from the structured data set, where the marketing indicators include sales volume, conversion rate, customer acquisition cost, and market share; Use the Pearson correlation coefficient or Spearman rank correlation coefficient in the correlation coefficient analysis method to calculate the correlation between each marketing indicator to obtain the correlation coefficient, and construct a correlation matrix.

[0011] Preferably, the specific steps of dividing the index correlation into three correlation levels of strong, medium, and weak levels based on the size of the correlation coefficient according to multiple preset threshold ranges to obtain the marketing index correlation level division result include: According to industry standards, preset the threshold range of the correlation coefficient; Traverse the correlation matrix, divide the index correlation into three levels of strong, medium, and weak levels according to the preset threshold, generate a marketing index correlation level division result table, and mark the correlation level of each index.

[0012] Preferably, the specific steps of screening out the marketing indicators with a strong correlation level and marking them as first-level indicators based on the marketing index correlation level division result include: Based on the correlation level division result, use a data processing tool to screen out the marketing indicators with a strong correlation level; Mark the screened strongly correlated indicators as first-level indicators in the data set, and generate a priority list through a visualization tool.

[0013] Preferably, the specific steps of analyzing and identifying implicit correlation data from the structured data set through natural language processing technology and long short-term memory network data processing technology include: Apply natural language processing technology to analyze user comments and social media texts in user behavior data, and extract the sentiment tendency of user comments; Using long short-term memory networks, we analyze the time series data in sales data to identify market trend turning points. The time series data includes historical sales data and market research data, and the market trend turning points include the time points when sales suddenly increase or decrease.

[0014] The extracted sentiment tendencies and market trend inflection points of user comments are integrated into a data set as a new field or an independent data table.

[0015] Preferably, the specific steps of screening out the primary indicators associated with the product's implicit data from the primary indicators based on the acquired implicit associated data, identifying them as evaluation indicators and generating a periodic change report at the same time include: Conduct correlation analysis on the implicit correlation data integrated with the sentiment tendency of user comments and the turning point of market trends and the primary indicators, and select the primary indicators associated with the implicit correlation data according to the preset correlation threshold; The screened primary indicators are identified as evaluation indicators, and a periodic change report is generated, wherein the periodic change report includes a trend chart of the evaluation indicators and an analysis table of the impact of implicit associated data on the evaluation indicators.

[0016] Compared with the related art, the product marketing management system and method based on big data provided by the present invention has the following beneficial effects: The present invention comprehensively collects and standardizes marketing data, and uses correlation coefficient analysis methods to quantitatively evaluate the correlation of marketing indicators, dividing them into three levels of strong, medium and weak correlations, thereby revealing the potential rules and problems between marketing indicators for enterprises; secondly, it further screens out the first-level indicators with strong correlations and presents them visually, so that enterprises can focus on core indicators for key management; at the same time, it uses natural language processing and long short-term memory network technology to mine implicit correlation data such as user comment sentiment tendencies and market trend inflection points, providing deeper market insights for marketing strategy adjustments; finally, based on the implicit correlation data, it screens out evaluation indicators and generates periodic change reports, helping enterprises to intuitively grasp the dynamics of key indicators and the impact of implicit factors, thereby comprehensively improving the accuracy and response speed of marketing management, and providing good support for optimizing decision-making processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a product marketing management method based on big data of the present invention; Figure 2 This is a system block diagram of a product marketing management system based on big data of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

[0019] Embodiment 1 As shown Figure 1 in the figure, a product marketing management method based on big data includes the following steps: S1. Establish a periodic data collection mechanism by integrating multi-source systems to periodically collect product marketing data for preprocessing and form a structured data set; S2. Extract various marketing indicators from the structured data set and calculate the correlation coefficients between the marketing indicators by using the correlation coefficient analysis method; S3. Based on the magnitude of the correlation coefficient, divide the index correlation into three correlation levels: strong level, medium level, and weak level according to a plurality of preset threshold ranges to obtain the marketing index correlation level division result; S4. Based on the marketing index correlation level division result, screen out the marketing indicators with a strong correlation level and label them as first-level indicators; S5. Through natural language processing technology and long short-term memory network data processing technology, analyze and identify implicit association data from the structured data set, where the implicit association data includes: user comment sentiment tendency and market trend inflection points; S6. Based on the obtained implicit association data, screen out the first-level indicators associated with the product implicit data from the first-level indicators, label them as evaluation indicators, and generate a periodic change report.

[0020] In the specific implementation process, the specific steps of step S1 include: S101. Integrate multi-source data systems inside and outside the enterprise, including CRM systems, e-commerce platforms, social media platforms, or market research databases, and establish a unified data interface for data synchronization.

[0021] Specifically, for different data systems, corresponding data interfaces are adopted. For example, for the CRM system, use the API interface provided by it to read data; for the e-commerce platform, according to the different platforms, use the official SDK provided to obtain data; for the social media platform, use the open social media API of it to obtain data such as user comments, likes, and shares; for the market research database, establish a connection through database connection technologies (such as JDBC, ODBC, etc.). Integrate the developed interfaces to build a unified data interface platform to realize the docking of each data system with this platform.

[0022] S102. According to the preset collection period, regularly collect product marketing data, including sales data, user behavior data, and market promotion data.

[0023] Specifically, set a reasonable data collection period. Exemplarily, for sales data, since sales situations change in real time, it is set to be collected once a day; for user behavior data, such as user browsing records, search records, etc., it is set to be collected once a week; for market promotion data, such as advertising placement effect data, the collection frequency is set according to the advertising placement cycle, such as once a week or once a month. Use a unified data interface platform to automatically collect product marketing data from each data system according to the preset collection period. It should be noted that during the collection process, ensure the integrity and accuracy of the data. Store the collected product marketing data in a specified data storage system, such as a relational database (such as MySQL, Oracle) or a distributed file system (such as Hadoop HDFS). During the storage process, classify and store the product marketing data, and organize it according to dimensions such as data type and collection time to facilitate subsequent data query and management.

[0024] S103. Remove duplicate values, missing values, and outliers from the collected product marketing data, and perform data standardization processing to convert it into a structured data set.

[0025] Specifically, remove duplicate values from the product marketing data: Use a data deduplication algorithm, such as a deduplication method based on a hash table, to perform deduplication processing on the collected data. By comparing the key fields of the data (such as order number, user ID, etc.), identify and delete duplicate data records; for numerical data, fill it with the mean, median, or mode; for categorical data, fill it with the most common category; for important missing values, it can also be filled manually or delete the records containing missing values. Finally, perform data standardization processing on the product marketing data and convert it into a structured data set.

[0026] In the specific implementation process, the specific steps of step S2 include: S201. Extract marketing metrics from the structured data set, where the marketing metrics include sales volume, conversion rate, customer acquisition cost, and market share.

[0027] Specifically, define the meaning and calculation method of each marketing metric. Sales refer to the total revenue obtained from product sales within a certain period; conversion rate refers to the ratio of the number of users who complete the target actions (such as purchases, registrations, etc.) to the number of visiting users; customer acquisition cost refers to the average cost required to acquire a new customer; market share refers to the proportion of the enterprise's product sales in the total sales of similar products in the market. Then, according to the field structure of the structured dataset, select the data fields related to the marketing metrics. For example, obtain the sales field from the sales data table, obtain the number of visiting users and the number of users who complete the target actions from the user behavior data table to calculate the conversion rate, obtain the relevant data fields of the customer acquisition cost from the market promotion data table, and obtain the relevant data from the sales data and market research data to calculate the market share. Use existing data processing tools (such as SQL query statements, the Pandas library in Python, etc.) to extract the required data fields from the structured dataset and calculate according to the definition of the marketing metrics.

[0028] S202. Use the Pearson correlation coefficient or Spearman rank correlation coefficient in the correlation coefficient analysis method to calculate the correlation between each marketing metric to obtain the correlation coefficient, and construct a correlation matrix.

[0029] Specifically, according to the data type and distribution characteristics of the marketing metrics, select an appropriate correlation coefficient analysis method. If the data of the marketing metric follows a normal distribution and the variables have a linear relationship, the Pearson correlation coefficient can be selected; if the data of the marketing metric does not follow a normal distribution or the variables have a non-linear relationship, the Spearman rank correlation coefficient can be selected. Then, use statistical software (such as the R language, SPSS, etc.) to calculate the correlation coefficient between each marketing metric. For each pair of marketing metrics, calculate their correlation coefficient values, and the value range of this value is between -1 and 1. A positive value indicates a positive correlation, a negative value indicates a negative correlation, the closer the absolute value is to 1, the stronger the correlation, and the closer the absolute value is to 0, the weaker the correlation. Finally, organize the calculated correlation coefficient values in the form of a matrix to form a correlation matrix, where the rows and columns of the correlation matrix correspond to each marketing metric respectively, and the elements in the matrix represent the correlation coefficient between the marketing metrics corresponding to the row and column.

[0030] In the specific implementation process, the specific steps of step S3 include: S301. Preset the threshold range of the correlation coefficient according to industry standards.

[0031] Specifically, consult relevant research reports and industry white papers in the same industry to understand the generally recognized correlation coefficient thresholds in the industry. Exemplarily, in the fast-moving consumer goods industry, a correlation coefficient greater than 0.7 is considered a strong association. In the high-tech industry, due to the relatively fast market changes, 0.6 is used as the threshold for strong association.

[0032] S302. Traverse the correlation matrix, divide the index correlations into three levels: strong, medium, and weak according to the preset threshold, generate a marketing index correlation level classification result table, and mark the correlation levels of each index.

[0033] Specifically, use a data analysis tool (such as Excel) to traverse the correlation matrix constructed in step S202. For each element in the matrix, that is, the correlation coefficient between each pair of marketing indicators, make a judgment according to the preset threshold range. According to the judgment results, divide the index correlations into three levels: strong, medium, and weak. If the correlation coefficient falls within the strong-level threshold range, mark the correlation level of this pair of indicators as strong; if it falls within the medium-level threshold range, mark it as medium; if it falls within the weak-level threshold range, mark it as weak. Then, organize the correlation level information of each marketing indicator with other indicators into a table, that is, the marketing index correlation level classification result table. The table should include fields such as index name, associated index name, and correlation level.

[0034] In the specific implementation process, the specific steps of step S4 include: S401. Based on the correlation level classification result, use a data processing tool to screen out the marketing indicators with a strong correlation level.

[0035] Specifically, based on the correlation level classification result, use the Pandas library, R language, or Excel in the data processing tool to screen out the marketing indicators with a strong correlation level.

[0036] S402. Mark the screened strongly associated indicators as first-level indicators in the dataset and generate a priority list through a visualization tool.

[0037] Specifically, add a new field (such as index level) to the screened strongly associated indicators and mark the screened strongly associated indicators as first-level indicators in this field. For example, in the above dataset, add an index level field for sales, conversion rate, and market share and assign the value of first-level indicator. Use a visualization tool (such as Tableau, Power BI, ECharts, etc.) to generate a priority list. Exemplarily, taking Tableau as an example, create a bar chart or table view, use "first-level indicator" as the dimension, and the index name as the measure.

[0038] In the specific implementation process, the specific steps of step S5 include: S501. Analyze the user comments and social media texts in the user behavior data using natural language processing technology, and extract the sentiment tendency of the user comments.

[0039] Specifically, use Chinese sentiment dictionaries (such as BosonNLP sentiment dictionary, Dalian University of Technology sentiment vocabulary ontology, etc.) in natural language processing technology to match the segmented text, count the number of positive sentiment words and negative sentiment words, and based on the quantity of sentiment words, further judge the sentiment tendency. Specifically, take the three sentiment words with the largest quantity as the sentiment tendency.

[0040] S502. Use a long short-term memory network to analyze the time series data in the sales data and identify the inflection points of the market trend. Among them, the time series data includes historical sales data and market research data, and the inflection points of the market trend include the time points of sharp increase or decrease in sales volume.

[0041] Specifically, obtain the historical sales data and market research data, including timestamps and corresponding sales volumes or market research index values, and use them as a training set to train the long short-term memory network model. Then, use the long short-term memory network to analyze the time series data in the sales data and identify the inflection points of the market trend.

[0042] S503. Integrate the extracted sentiment tendency of the user comments and the inflection points of the market trend into a data set, which is used as a new field or an independent data table.

[0043] In the specific implementation process, the specific steps of step S6 include: S601. Conduct a correlation analysis on the implicit association data integrated with the sentiment tendency of the user comments and the inflection points of the market trend and the first-level indicators, and screen out the first-level indicators associated with the implicit association data according to the preset association threshold.

[0044] Specifically, use data analysis tools (such as the pandas and scipy libraries in Python, or the R language, etc.) to calculate the correlation between the first-level indicators and the implicit association data. For each first-level indicator, calculate its correlation coefficients with the sentiment tendency of the user comments and the inflection points of the market trend respectively. According to the business requirements and data analysis objectives, preset an association threshold, which is used to judge whether the association degree between the first-level indicator and the implicit association data is significant enough, so as to determine whether to screen it as an evaluation indicator. In this embodiment, the threshold is set to 0.6. When the correlation coefficient is greater than or equal to 0.6, it is considered that there is a significant association between the first-level indicator and the implicit association data. Then, traverse the correlation coefficients of all first-level indicators, and screen out the first-level indicators with correlation coefficients greater than or equal to the preset association threshold as the indicators associated with the implicit association data.

[0045] S602: Identify the screened primary indicators as evaluation indicators, and generate a periodic change report, wherein the periodic change report includes a trend chart of the evaluation indicators and an analysis table of the impact of implicit associated data on the evaluation indicators.

[0046] Specifically, use data visualization tools (such as Tableau, Power BI, Excel, etc.) to generate trend charts of evaluation indicators from the filtered primary indicators. Trend charts can show the changes of evaluation indicators in different time periods, helping marketers to intuitively understand the trend of indicators. For example, you can draw a line chart of sales over time to observe the fluctuation trend of sales.

[0047] Embodiment 2 like Figure 2 As shown, a product marketing management system based on big data includes: The data collection and processing module is used to establish a periodic data collection mechanism by integrating multiple source systems to periodically collect product marketing data for preprocessing and form a structured data set; The correlation analysis module is used to extract various marketing indicators from the structured data set and calculate the correlation coefficients between the marketing indicators using the correlation coefficient analysis method; A level classification module is used to classify the indicator relevance into three levels of strong, medium and weak based on the size of the correlation coefficient and according to a plurality of preset threshold segments, and obtain the marketing indicator relevance level classification result; A data identification module is used to screen out marketing indicators with a strong correlation level and identify them as first-level indicators based on the marketing indicator correlation level classification results; The analysis and processing module is used to analyze and identify implicit related data from structured data sets through natural language processing technology and long short-term memory network data processing technology, wherein the implicit related data includes: user comment sentiment tendency and market trend inflection point; The report generation module is used to screen out the first-level indicators associated with the product implicit data from the first-level indicators based on the acquired implicit associated data, identify them as evaluation indicators and generate a periodic change report at the same time.

[0048] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a means for implementing the functions specified in one or more of the following processes Figure 1 one or more of the following processes and / or blocks Figure 1 or a means for implementing the functions specified in a plurality of blocks.

[0049] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program, and this program can be stored in a computer-readable storage medium. The storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM), or other optical disc memories, magnetic disk memories, tape memories, or any other medium that can be used to carry or store data and is readable by a computer.

[0050] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, commodity or device including the element.

Claims

1. A product marketing management system based on big data, characterized in that, The management system includes: A data collection and processing module, which is used to establish a periodic data collection mechanism by integrating multi-source systems, periodically collect product marketing data for preprocessing, and form a structured data set; An association analysis module, which is used to extract various marketing indicators from the structured data set and calculate the correlation coefficients between the marketing indicators by using the correlation coefficient analysis method; A level division module, which is used to divide the indicator associations into three association levels of strong level, medium level and weak level based on the size of the correlation coefficient according to a plurality of preset threshold ranges, and obtain the marketing indicator association level division result; A data identification module, which is used to screen out the marketing indicators with a strong association level based on the marketing indicator association level division result and identify them as first-level indicators; An analysis and processing module, which is used to analyze and identify implicit association data from the structured data set by using natural language processing technology and long short-term memory network data processing technology, where the implicit association data includes: user comment sentiment tendency and market trend inflection points; A report generation module, which is used to screen out the first-level indicators associated with the product implicit data from the first-level indicators based on the obtained implicit association data, identify them as evaluation indicators and generate a periodic change report at the same time.

2. A product marketing management method based on big data, applied to a product marketing management system based on big data as described in claim 1, characterized in that, The management method includes the following steps: Establish a periodic data collection mechanism by integrating multi-source systems, periodically collect product marketing data for preprocessing, and form a structured data set; Extract various marketing indicators from the structured data set and calculate the correlation coefficients between the marketing indicators by using the correlation coefficient analysis method; Based on the size of the correlation coefficient, divide the indicator associations into three association levels of strong level, medium level and weak level according to a plurality of preset threshold ranges, and obtain the marketing indicator association level division result; Based on the marketing indicator association level division result, screen out the marketing indicators with a strong association level and identify them as first-level indicators; Analyze and identify implicit association data from the structured data set by using natural language processing technology and long short-term memory network data processing technology, where the implicit association data includes: user comment sentiment tendency and market trend inflection points; Based on the obtained implicit association data, screen out the first-level indicators associated with the product implicit data from the first-level indicators, identify them as evaluation indicators and generate a periodic change report at the same time.

3. The method for product marketing management based on big data according to claim 2, wherein The specific steps of establishing a periodic data collection mechanism by integrating multi-source systems, periodically collecting product marketing data for preprocessing, and forming a structured data set include: Integrate multi-source data systems inside and outside the enterprise, including CRM systems, e-commerce platforms, social media platforms or market research databases, and establish a unified data interface for data synchronization; According to the preset collection period, regularly collect product marketing data, including sales data, user behavior data, and market promotion data; Remove duplicate values, missing values and outliers from the collected product marketing data, and perform data standardization processing to convert it into a structured data set.

4. The product marketing management method based on big data according to claim 3, characterized in that, The specific steps of extracting various marketing indicators from the structured data set and calculating the correlation coefficients between the marketing indicators by using the correlation coefficient analysis method include: Extract marketing metrics from structured datasets, where the marketing metrics include sales, conversion rate, customer acquisition cost, and market share; Use the Pearson correlation coefficient or Spearman rank correlation coefficient in the correlation analysis method to calculate the correlation between each marketing metric to obtain a correlation coefficient, and construct a correlation matrix.

5. A product marketing management method based on big data according to claim 4, characterized in that, Based on the magnitude of the correlation coefficient, divide the metric correlation into three correlation levels: strong, medium, and weak according to multiple preset threshold ranges, and the specific steps to obtain the marketing metric correlation level classification result include: According to industry standards, preset the threshold range of the correlation coefficient; Traverse the correlation matrix, divide the metric correlation into three levels: strong, medium, and weak according to the preset threshold, generate a marketing metric correlation level classification result table, and mark the correlation level of each metric.

6. The product marketing management method based on big data according to claim 5, characterized in that, Based on the marketing metric correlation level classification result, the specific steps to screen out the marketing metrics with a strong correlation level and mark them as first-level metrics include: Based on the correlation level classification result, use a data processing tool to screen out the marketing metrics with a strong correlation level; Mark the screened strongly correlated metrics as first-level metrics in the dataset, and generate a priority list through a visualization tool.

7. A product marketing management method based on big data according to claim 6, characterized in that, The specific steps to analyze and identify implicit association data from structured datasets through natural language processing technology and long short-term memory network data processing technology include: Apply natural language processing technology to analyze user comments and social media texts in user behavior data, and extract the sentiment tendency of user comments; Use a long short-term memory network to analyze time series data in sales data to identify market trend inflection points, where the time series data includes historical sales data and market research data, and the market trend inflection points include time points of sudden increase or decrease in sales volume; Integrate the sentiment tendency of user comments and market trend inflection points extracted into a dataset as a new field or an independent data table.

8. A product marketing management method based on big data according to claim 7, characterized in that, Based on the obtained implicit association data, the specific steps to screen out the first-level metrics associated with product implicit data from the first-level metrics, mark them as evaluation metrics, and generate a periodic change report include: Conduct a correlation analysis on the implicit association data integrated from the sentiment tendency of user comments and market trend inflection points and the first-level metrics, and according to the preset association threshold, screen out the first-level metrics associated with the implicit association data; Mark the screened first-level metrics as evaluation metrics, and generate a periodic change report, where the periodic change report includes a trend chart of the evaluation metrics and an impact analysis table of the implicit association data on the evaluation metrics.

Images