A enterprise marketing data management system and method based on big data

Abstract

The present invention belongs to the technical field of enterprise marketing data management, and specifically relates to an enterprise marketing data management system and method based on big data. The invention collects the marketing data of the enterprise, sorts and segments it according to the collection time, selects the evaluation segment and extracts the fluctuation amount and fluctuation frequency of the secondary sample data as the corresponding characteristic parameters, and obtains the evaluation parameters by combining these characteristic parameters, and then dynamically adjusts the data collection interval according to the evaluation parameters. After updating the data collection interval, the real-time marketing data in the verification period is analyzed for deviation distribution by matching the verification period to ensure the stability of the data distribution, and when the verification parameter is abnormal, an alarm signal is issued in time to adjust the data collection strategy in time, effectively realize the dynamic adjustment of the data collection interval, improve the efficiency and accuracy of data collection, and ensure the stability of the data at the same time, providing strong support for the marketing decision of the enterprise.

Description

Technical Field

[0001] The present invention belongs to the technical field of enterprise marketing data management, and specifically relates to an enterprise marketing data management system and method based on big data. Background Art

[0002] In the current big data environment, the collection, analysis and utilization of enterprise marketing data are particularly important. Through in-depth mining and analysis of marketing data, enterprises can more accurately understand market dynamics, customer needs and competitive situations, thereby formulating more reasonable marketing strategies and achieving more efficient market promotion and customer service. After the collection of marketing data is completed, it will not only focus on the collection and storage of data, but also further effectively analyze and utilize the enterprise marketing data to help enterprises give full play to the value of big data.

[0003] However, traditional enterprise marketing data management systems often have problems such as unreasonable data collection intervals and unstable data distribution during execution. For example, regardless of the quality of marketing data, the inherent intervals are maintained for collection. In abnormal marketing situations, sufficient data support cannot be obtained quickly. Conversely, under normal circumstances, overly frequent collection will lead to waste of resources and increase the complexity and difficulty of data processing and analysis. Based on this, this solution provides an enterprise marketing data management method that dynamically adjusts the data collection interval to ensure the stability and accuracy of the data. Summary of the invention

[0004] The purpose of the present invention is to provide an enterprise marketing data management system and method based on big data, which can dynamically adjust the data collection interval to ensure the stability and accuracy of the data.

[0005] The technical solution adopted by the present invention is as follows:

[0006] A method for enterprise marketing data management based on big data, comprising:

[0007] Collecting the marketing data of the enterprise, and marking the marketing data as primary sample data, wherein the primary sample data includes normal marketing data and abnormal marketing data;

[0008] The first-level sample data is sorted according to the collection time, and the occurrence node and the recovery node of the abnormal marketing data are set as discontinuity points, and then the first-level sample data is segmented according to the discontinuity points to obtain multiple sample segments;

[0009] Selecting an evaluation segment from the plurality of sample segments, and marking the marketing data in the evaluation segment as secondary sample data, and then collecting the fluctuation amount and fluctuation frequency of the secondary sample data, and marking them as the first characteristic parameter and the second characteristic parameter respectively;

[0010] Performing a combined operation on the first characteristic parameter and the second characteristic parameter, and outputting the operation result as an evaluation parameter, and updating the data collection interval according to the evaluation parameter, and performing subsequent sampling according to the updated data collection interval;

[0011] The verification period is matched according to the updated data collection interval, and the deviation distribution analysis of the marketing data within the verification period is performed to obtain the verification parameters, and then the execution status of the updated data collection interval is output according to the verification parameters, and an alarm signal is issued when the execution status is abnormal.

[0012] In a preferred solution, after the first-level sample data is sorted, difference processing is performed one by one according to the sorting results to obtain data fluctuation amount;

[0013] According to the sorting result of the primary sample data, the data fluctuation amounts are grouped to obtain multiple groups of continuous data fluctuation amounts, and recorded as screening condition parameters;

[0014] Obtaining a screening interval, and comparing the screening interval with each set of screening condition parameters;

[0015] If the screening condition parameters of the same group are all within the screening range, it indicates that the primary sample data fluctuates normally, and the primary sample data under normal fluctuation is recorded as normal marketing data;

[0016] If there are screening condition parameters that are not within the screening interval among the screening condition parameters in the same group, then the proportion of the screening condition parameters that are not within the screening interval is calculated;

[0017] An evaluation threshold is obtained, and the evaluation threshold is compared with the proportion of the screening condition parameters that are not within the screening interval. When the proportion of the screening condition parameters that are not within the screening interval is less than the evaluation threshold, the corresponding first-level sample data is recorded as normal marketing data; otherwise, it indicates that the first-level sample data fluctuates normally or abnormally, and the first-level sample data under abnormal fluctuations is recorded as abnormal marketing data.

[0018] In a preferred embodiment, the step of segmenting the primary sample data according to the discontinuity points to obtain a plurality of sample segments includes:

[0019] Obtaining the time intervals between each of the recovery nodes and the occurrence nodes adjacent to the recovery node and located after the recovery node, and recording them as the reference duration period;

[0020] Obtaining the duration of each of the reference duration periods and calibrating them as screening condition parameters;

[0021] Obtaining a screening threshold, and comparing the screening threshold with a screening condition parameter;

[0022] If the screening condition parameter is greater than the screening threshold, it indicates that the primary sample data under the reference duration period is long-term time-varying data, and the reference duration period is recorded as a sample segment;

[0023] If the screening condition parameter is less than or equal to the screening threshold, it indicates that the primary sample data under the reference duration period is short-term data, and the short-term data and its corresponding reference duration period are simultaneously screened out.

[0024] In a preferred embodiment, the step of selecting an evaluation segment from the plurality of sample segments and marking the marketing data in the evaluation segment as secondary sample data includes:

[0025] Obtain all the sample segments and sort them according to their duration;

[0026] Obtain the sample segment with the shortest duration, and mark the duration as the standard duration;

[0027] Based on the standard duration, the end node of each sample segment is backtracked and offset, and the offset result is recorded as an evaluation segment, and the marketing data in the evaluation segment is recorded as secondary sample data.

[0028] In a preferred solution, the step of collecting the fluctuation amount and the fluctuation frequency of the secondary sample data and marking them as the first characteristic parameter and the second characteristic parameter respectively includes:

[0029] Acquire all the secondary sample data, perform difference processing on two adjacent secondary sample data, obtain the fluctuation amount of the secondary sample data, and record it as the first characteristic parameter;

[0030] According to the collection order of the secondary sample data, the collection time corresponding to each of the first characteristic parameters is obtained, and the collection time interval between two adjacent first characteristic parameters is calculated to obtain the fluctuation interval;

[0031] According to the fluctuation interval, the fluctuation frequency of the secondary sample data is calculated, and the fluctuation frequency is recorded as a second characteristic parameter.

[0032] In a preferred embodiment, the step of combining the first characteristic parameter and the second characteristic parameter and outputting the result of the operation as an evaluation parameter includes:

[0033] Obtaining the first characteristic parameter, and calculating an average value of the first characteristic parameter to obtain an average fluctuation amount;

[0034] Obtaining a maximum value of the first characteristic parameters;

[0035] An evaluation function is obtained, and the average fluctuation amount, the second characteristic parameter and the maximum value of the first characteristic parameter are input into the evaluation function, and the output result is calibrated as the evaluation parameter.

[0036] In a preferred embodiment, the step of updating the data collection interval according to the evaluation parameter comprises:

[0037] Obtaining the evaluation parameter, and comparing the evaluation parameter with a preset evaluation threshold;

[0038] If the evaluation parameter is greater than a preset evaluation threshold, it indicates that the current data collection interval is large, and the data collection interval is shortened;

[0039] If the evaluation parameter is less than or equal to the preset evaluation threshold, it indicates that the current data collection interval is moderate, and the current data collection interval is maintained.

[0040] In a preferred embodiment, the step of matching the verification period according to the updated data collection interval and performing deviation distribution analysis on the marketing data within the verification period to obtain the verification parameters includes:

[0041] Performing a difference processing on the evaluation parameter and a preset evaluation threshold to obtain a deviation parameter;

[0042] Acquire a scoring interval, compare the deviation parameter with the scoring interval, and output an evaluation score of the deviation parameter according to the comparison result, wherein a plurality of scoring intervals and evaluation scores are provided;

[0043] Obtaining an updated data collection interval, and matching a verification period according to the data collection interval, wherein a plurality of verification periods are provided, and correspond one to one with the evaluation scores;

[0044] Acquire the real-time marketing data within the verification period, and after the verification period ends, calculate the average value and median of the real-time marketing data;

[0045] Calculating the deviation of the marketing data based on the mean and the median, and recording the deviation as a verification parameter;

[0046] A verification threshold is obtained, and the verification threshold is compared with a verification parameter. When the verification parameter is less than or equal to the verification threshold, it indicates that the distribution of real-time marketing data within the verification period is stable, and the execution status of the updated data collection interval is recorded as a normal state. Otherwise, it indicates that the distribution of real-time marketing data within the verification period is unstable, and the execution status of the updated data collection interval is recorded as an abnormal state.

[0047] The present invention also provides a big data-based enterprise marketing data management system, which is applied to the above-mentioned big data-based enterprise marketing data management method, comprising:

[0048] A data collection module, wherein the data collection module is used to collect the marketing data of the enterprise and mark the marketing data as primary sample data, wherein the primary sample data includes normal marketing data and abnormal marketing data;

[0049] A sample segmentation module, the sample segmentation module is used to sort the primary sample data according to the collection time, and set the occurrence node and the recovery node of the abnormal marketing data as discontinuity points, and then segment the primary sample data according to the discontinuity points to obtain multiple sample segments;

[0050] A feature extraction module, the feature extraction module is used to select an evaluation segment from the plurality of sample segments, and mark the marketing data in the evaluation segment as secondary sample data, and then collect the fluctuation amount and fluctuation frequency of the secondary sample data, and mark them as a first feature parameter and a second feature parameter respectively;

[0051] An updating module, the updating module being used to perform a combined operation on the first characteristic parameter and the second characteristic parameter, and output the operation result as an evaluation parameter, and update the data collection interval according to the evaluation parameter, and perform subsequent sampling according to the updated data collection interval;

[0052] A verification module is used to match the verification period according to the updated data collection interval, and perform deviation distribution analysis on the marketing data within the verification period to obtain verification parameters, and then output the execution status of the updated data collection interval according to the verification parameters, and issue an alarm signal when the execution status is abnormal.

[0053] And, an electronic device, the electronic device comprising:

[0054] at least one processor;

[0055] and a memory communicatively coupled to the at least one processor;

[0056] Among them, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned enterprise marketing data management method based on big data.

[0057] The technical effects achieved by the present invention are:

[0058] The present invention collects the marketing data of the enterprise, sorts and segments the data according to the collection time, selects the evaluation segment and extracts the fluctuation amount and fluctuation frequency of the secondary sample data as the corresponding characteristic parameters, and obtains the evaluation parameters by combining these characteristic parameters, and then dynamically adjusts the data collection interval according to the evaluation parameters. After updating the data collection interval, the deviation distribution analysis is performed on the real-time marketing data in the verification period by matching the verification period to ensure the stability of the data distribution, and when the verification parameter is abnormal, an alarm signal is issued in time to adjust the data collection strategy in time, effectively realize the dynamic adjustment of the data collection interval, improve the efficiency and accuracy of data collection, and ensure the stability of the data at the same time, providing strong support for the marketing decision-making of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a flow chart of the method of the present invention;

[0060] Figure 2 It is a system module diagram of the present invention;

[0061] Figure 3 It is a structural diagram of an electronic device of the present invention. DETAILED DESCRIPTION

[0062] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0063] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0064] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" that appears in different places in this specification does not refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0065] See also Figure 1 As shown, the present invention provides an enterprise marketing data management method based on big data, comprising:

[0066] S1. Collect the marketing data of the enterprise and mark the marketing data as primary sample data, where the primary sample data includes normal marketing data and abnormal marketing data;

[0067] S2. Sort the primary sample data according to the collection time, set the occurrence node and recovery node of the abnormal marketing data as discontinuity points, and then segment the primary sample data according to the discontinuity points to obtain multiple sample segments;

[0068] S3, selecting an evaluation segment from the multiple sample segments, and marking the marketing data in the evaluation segment as secondary sample data, and then collecting the fluctuation amount and fluctuation frequency of the secondary sample data, and marking them as the first characteristic parameter and the second characteristic parameter respectively;

[0069] S4, performing a combined operation on the first characteristic parameter and the second characteristic parameter, and outputting the operation result as an evaluation parameter, and updating the data collection interval according to the evaluation parameter, and performing subsequent sampling according to the updated data collection interval;

[0070] S5. Match the verification period according to the updated data collection interval, and perform deviation distribution analysis on the marketing data within the verification period to obtain verification parameters. Then output the execution status of the updated data collection interval according to the verification parameters, and issue an alarm signal when the execution status is abnormal.

[0071] As described in the above steps S1-S5, with the continuous increase of enterprise marketing data, traditional enterprise marketing data management may not be able to adapt to the real-time changes of data, resulting in inaccurate data collection or waste of resources. Therefore, this embodiment provides an enterprise marketing data management method based on big data to solve the above problems. In this embodiment, the marketing data of the enterprise is first collected. These marketing data can come from various channels, such as the enterprise's internal database, third-party data providers, etc. The collected marketing data needs to be pre-processed, including data cleaning, deduplication, format conversion, etc., to ensure the accuracy and consistency of the data. Next, the pre-processed marketing data is calibrated as primary sample data. The primary sample data includes normal marketing data and abnormal marketing data. Normal marketing data reflects the normal marketing activities of the enterprise, while abnormal marketing data reveals possible abnormal behaviors or events. Then, the primary sample data is sorted, and the occurrence nodes and recovery nodes of the abnormal marketing data are set as discontinuity points to better identify and handle abnormal situations. After that, The first-level sample data is then segmented according to the discontinuity points to obtain multiple sample segments. Among the multiple sample segments, it is necessary to further select evaluation segments, and the marketing data in the evaluation segments are marked as second-level sample data. Then, the fluctuation amount and fluctuation frequency of the second-level sample data are collected to reflect the dynamic changes of the second-level sample data. Finally, the fluctuation amount and fluctuation frequency of the second-level sample data are used as the first characteristic parameter and the second characteristic parameter respectively, and the first characteristic parameter and the second characteristic parameter are combined and operated to obtain a comprehensive evaluation parameter. According to the evaluation parameter, the data collection interval can be adjusted to more accurately capture the changes in marketing data. After adjusting the data collection interval, it is necessary to match the verification period, and analyze the marketing data in the verification period to obtain the verification parameters. The verification parameters can help evaluate the rationality of the data collection interval. According to the verification parameters, the execution status of the updated data collection interval can be output. When the execution status is abnormal, an alarm signal will be issued to remind relevant personnel to deal with it in time to ensure the stability and security of the company's marketing data management.

[0072] In a preferred embodiment, after the first-level sample data is sorted, the sorting results are subjected to difference processing one by one to obtain the data fluctuation amount;

[0073] According to the sorting results of the primary sample data, the data fluctuations are grouped and processed to obtain multiple groups of continuous data fluctuations, which are recorded as screening condition parameters;

[0074] Obtain the screening interval, and compare the screening interval with each set of screening condition parameters;

[0075] If all the parameters of the same group of screening conditions are within the screening range, it indicates that the primary sample data fluctuates normally, and the primary sample data under normal fluctuations is recorded as normal marketing data;

[0076] If there are filtering condition parameters that are not within the filtering interval in the same group of filtering condition parameters, the proportion of the filtering condition parameters that are not within the filtering interval is calculated;

[0077] An evaluation threshold is obtained, and the evaluation threshold is compared with the proportion of the screening condition parameters that are not in the screening interval. When the proportion of the screening condition parameters that are not in the screening interval is less than the evaluation threshold, the corresponding first-level sample data is recorded as normal marketing data; otherwise, it indicates that the first-level sample data fluctuates abnormally, and the first-level sample data under abnormal fluctuations is recorded as abnormal marketing data.

[0078] In this implementation, after the primary sample data is output, the primary sample data is first sorted. After the sorting is completed, the data needs to be differenced one by one, so that the fluctuation amount of adjacent primary sample data can be obtained. Then, according to the sorting result of the primary sample data, the data fluctuation amount is grouped and processed, such as every two or three data fluctuation amounts are grouped. After the grouping processing, multiple groups of continuous data fluctuation amounts will be obtained. These data fluctuation amounts are recorded as screening condition parameters, and then the screening interval is obtained. The screening interval is a range set during the analysis process to determine whether the data fluctuation is normal. By comparing the screening interval with each group of screening condition parameters, it is easy to determine whether the data fluctuation is in line with expectations. During the comparison process, if it is found that the screening condition parameters in the same group are all within the screening interval, then this indicates that the primary sample data fluctuates normally. At this time, the first-level sample data under normal fluctuations will be recorded as normal marketing data. However, if there are filtering condition parameters in the same group that are not within the filtering interval, then it is necessary to calculate the proportion of these filtering condition parameters that are not within the filtering interval. This proportion is used to determine the proportion of the data volume with abnormal fluctuations in the overall data. Next, it is necessary to obtain the evaluation threshold and compare the evaluation threshold with the proportion of the filtering condition parameters that are not within the filtering interval. During the comparison process, if the proportion of the filtering condition parameters that are not within the filtering interval is less than the evaluation threshold, then the corresponding first-level sample data will be recorded as normal marketing data. Conversely, if the proportion is greater than the evaluation threshold, then this indicates that there is abnormal fluctuation in the first-level sample data, and the first-level sample data under such abnormal fluctuation needs to be recorded as abnormal marketing data.

[0079] In a preferred embodiment, the step of segmenting the primary sample data according to the discontinuity points to obtain a plurality of sample segments includes:

[0080] S201, obtaining each recovery node and the time interval between occurrence nodes adjacent to the recovery node and located after the recovery node, and recording them as a reference duration period;

[0081] S202, obtaining the duration of each reference duration period and calibrating it as a screening condition parameter;

[0082] S203, obtaining a screening threshold, and comparing the screening threshold with a screening condition parameter;

[0083] If the screening condition parameter is greater than the screening threshold, it indicates that the primary sample data under the reference duration period is long-term time-varying data, and the reference duration period is recorded as a sample segment;

[0084] If the screening condition parameter is less than or equal to the screening threshold, it indicates that the primary sample data under the reference duration period is short-term data, and the short-term data and its corresponding reference duration period are simultaneously screened out.

[0085] As described in the above steps S201-S203, after the normal marketing data and the abnormal marketing data are determined, it is first necessary to obtain the recovery node of each abnormal marketing data, and it is also necessary to determine the occurrence node of the abnormal marketing data adjacent to the recovery node and located after the recovery node, and then count the time intervals between the two, and these time intervals can be used as the benchmark duration period. At the same time, it is also necessary to obtain the duration of each benchmark duration period. This duration can be used as a screening condition parameter to determine whether the primary sample data is long-term data or short-term data. After obtaining the screening condition parameters, determine the screening threshold. The screening threshold is used to divide the data into The critical values ​​of long time and short time, comparing the screening threshold with the screening condition parameter, can get the following two situations, if the screening condition parameter is greater than the screening threshold, then this means that the primary sample data under the benchmark duration period is long-time data, at this time, the benchmark duration period is recorded as the sample segment, if the screening condition parameter is less than or equal to the screening threshold, then this means that the primary sample data under the benchmark duration period is short-time data, in this case, it is necessary to screen out the short-time data and its corresponding benchmark duration period synchronously to ensure that in the subsequent analysis process, it will not be affected by the short-time data, thereby further improving the accuracy of data processing.

[0086] In a preferred embodiment, the step of selecting an evaluation segment from a plurality of sample segments and marking the marketing data in the evaluation segment as secondary sample data includes:

[0087] S301, obtaining all sample segments and sorting them according to their duration;

[0088] S302, obtaining a sample segment with the shortest duration, and marking the duration as a standard duration;

[0089] S303. Based on the standard duration, the end node of each sample segment is backtracked and offset, and the offset result is recorded as an evaluation segment, and the marketing data in the evaluation segment is recorded as secondary sample data.

[0090] As described in S301-S303 above, after the sample segments are output, it is necessary to obtain all sample segments and sort these sample segments. The basis for sorting is their duration, and the sample segments with shorter duration are extracted, and this sample segment is used as the standard duration as the benchmark for subsequent analysis. The purpose of setting it is to ensure the consistency of the amount of data in each sample segment, so as to ensure the accuracy of subsequent analysis. Based on the standard duration, the end nodes of other sample segments can be backtracked and offset, and then multiple evaluation segments with the same duration can be output, and the duration of each evaluation segment is the standard duration.

[0091] In a preferred embodiment, the steps of collecting the fluctuation amount and the fluctuation frequency of the secondary sample data and marking them as the first characteristic parameter and the second characteristic parameter respectively include:

[0092] S304, acquiring all secondary sample data, and performing difference processing on two adjacent secondary sample data to obtain the fluctuation amount of the secondary sample data, and recording it as the first characteristic parameter;

[0093] S305, according to the collection order of the secondary sample data, obtaining the collection time corresponding to each first characteristic parameter, and calculating the collection time interval between two adjacent first characteristic parameters to obtain the fluctuation interval;

[0094] S306. Calculate the fluctuation frequency of the secondary sample data based on the fluctuation interval, and record the fluctuation frequency as the second characteristic parameter.

[0095] As described in the above steps S304-S306, in the process of collecting the fluctuation amount and fluctuation frequency of the secondary sample data, it is first necessary to obtain all the secondary sample data under each evaluation segment respectively, and perform difference processing on two adjacent secondary sample data, so as to obtain the fluctuation amount of the secondary sample data. The fluctuation amount of the secondary sample data is used to measure the fluctuation of the secondary sample data. After obtaining the fluctuation amount of the secondary sample data, it is necessary to record it as the first characteristic parameter, and then obtain the collection time corresponding to each first characteristic parameter according to the collection order of the secondary sample data, and calculate the collection time interval between two adjacent first characteristic parameters, so as to obtain the fluctuation interval. After obtaining the fluctuation interval, it is necessary to calculate the fluctuation frequency of the secondary sample data according to the fluctuation interval. The calculation formula of the fluctuation frequency is: Fluctuation frequency = evaluation segment length / sum of all fluctuation intervals. Finally, the fluctuation frequency is recorded as the second characteristic parameter.

[0096] In a preferred embodiment, the step of combining the first characteristic parameter and the second characteristic parameter and outputting the calculation result as the evaluation parameter includes:

[0097] S401, obtaining a first characteristic parameter, and calculating an average value of the first characteristic parameter to obtain an average fluctuation amount;

[0098] S402, obtaining the maximum value of the first characteristic parameter;

[0099] S403, obtaining an evaluation function, and inputting the average fluctuation amount, the second characteristic parameter, and the maximum value of the first characteristic parameter into the evaluation function, and calibrating the output result as the evaluation parameter.

[0100] As described in the above steps S401-S403, in the process of outputting the evaluation parameters, the first characteristic parameter must be obtained, and its average value is calculated, and its output result is recorded as the average fluctuation, so as to better understand the fluctuation of the data. Next, the maximum value of the first characteristic parameter needs to be found, and then the evaluation function is introduced, and the average fluctuation, the second characteristic parameter and the maximum value of the first characteristic parameter are used as inputs, and these data are substituted into the evaluation function for calculation, wherein the expression of the evaluation function is: , where represents the evaluation parameters, represents the second characteristic parameter, represents the average volatility, It represents the maximum value of the first characteristic parameter (the maximum fluctuation of the secondary sample data). The evaluation parameter is used to reflect the overall fluctuation of the marketing data, and is used as a basis to evaluate the stability and effectiveness of the marketing data, providing support for the company's decision-making.

[0101] In a preferred embodiment, the step of updating the data collection interval according to the evaluation parameters includes:

[0102] Obtaining evaluation parameters, and comparing the evaluation parameters with preset evaluation thresholds;

[0103] If the evaluation parameter is greater than the preset evaluation threshold, it indicates that the current data collection interval is large, and the data collection interval is shortened;

[0104] If the evaluation parameter is less than or equal to the preset evaluation threshold, it indicates that the current data collection interval is moderate, and the current data collection interval is maintained.

[0105] In this implementation, after the evaluation parameter is output, the evaluation parameter is compared with a preset evaluation threshold to determine whether the data collection interval is reasonable. By comparing the evaluation parameter with the evaluation threshold, the relationship between the current data collection interval and the preset requirement can be obtained. If the evaluation parameter is greater than the preset evaluation threshold, it means that the current data collection interval is large and needs to be shortened. Shortening the data collection interval can increase the frequency of data collection, thereby improving data quality. If the evaluation parameter is less than or equal to the preset evaluation threshold, it indicates that the current data collection interval is moderate and no adjustment is required, thereby ensuring the rationality of the data collection interval and avoiding the negative impact of frequent adjustments.

[0106] In a preferred embodiment, the step of matching the verification period according to the updated data collection interval and performing deviation distribution analysis on the marketing data within the verification period to obtain the verification parameters includes:

[0107] S501, performing difference processing on the evaluation parameter and the preset evaluation threshold to obtain a deviation parameter;

[0108] S502, obtaining a scoring interval, comparing the deviation parameter with the scoring interval, and outputting an evaluation score of the deviation parameter according to the comparison result, wherein a plurality of scoring intervals and evaluation scores are provided;

[0109] S503, obtaining the updated data collection interval, and matching the verification period according to the data collection interval, wherein there are multiple verification periods, and they correspond one to one with the evaluation scores;

[0110] S504, obtaining real-time marketing data within the verification period, and calculating the average and median of the real-time marketing data after the verification period ends;

[0111] S505. Calculate the deviation of the marketing data based on the mean and the median, and record the deviation as a verification parameter;

[0112] S506. Obtain a verification threshold and compare the verification threshold with a verification parameter. When the verification parameter is less than or equal to the verification threshold, it indicates that the distribution of real-time marketing data within the verification period is stable, and the execution status of the updated data collection interval is recorded as a normal state. Otherwise, it indicates that the distribution of real-time marketing data within the verification period is unstable, and the execution status of the updated data collection interval is recorded as an abnormal state.

[0113] As described in the above steps S501-S506, after the updated data collection interval is output, it is necessary to further verify it to ensure that the updated data collection interval can be effectively executed. First, according to the updated data collection interval, the corresponding verification period needs to be matched. The setting of the verification period should take into account the data collection interval and the adequacy of the data volume to ensure the accuracy and effectiveness of the data. The length of the verification period is less than the length of the evaluation period. It is set according to actual needs. After the verification period ends, the evaluation parameters and the preset evaluation thresholds are firstly subtracted to obtain the deviation parameters. The deviation parameters are used to measure the gap between the marketing data and the expected data in the verification period. Then, the scoring interval is obtained, and the deviation parameters are compared with the scoring interval. According to the comparison result, the deviation parameters are matched with the evaluation scores. There are multiple scoring intervals and evaluation scores to evaluate the deviation parameters more comprehensively. On this basis, the updated data collection interval is obtained, and the verification period is matched according to the data collection interval. Correspondingly, there should be multiple verification periods, and they should correspond to the evaluation scores one by one, so as to conduct targeted evaluation of marketing data under different update data collection intervals. During the verification period, real-time marketing data needs to be obtained. After the verification period ends, the average and median of these real-time marketing data are calculated. Next, according to the average and median, the deviation of the marketing data can be calculated. The calculation formula is: Deviation = (average-median) / median*100%. After recording the deviation as the verification parameter, the verification threshold is obtained synchronously, and the verification threshold is compared with the verification parameter. When the verification parameter is less than or equal to the verification threshold, it indicates that the distribution of real-time marketing data in the verification period is stable. At this time, the execution status of the updated data collection interval is recorded as normal. On the contrary, if the verification parameter is greater than the verification threshold, it means that the distribution of real-time marketing data in the verification period is unstable. In this case, the execution status of the updated data collection interval is recorded as abnormal, and an alarm signal is issued synchronously to remind relevant personnel to handle it.

[0114] See also Figure 2 , a big data-based enterprise marketing data management system, applied to the above-mentioned big data-based enterprise marketing data management method, comprising:

[0115] A data collection module is used to collect the marketing data of the enterprise and mark the marketing data as primary sample data, wherein the primary sample data includes normal marketing data and abnormal marketing data;

[0116] The sample segmentation module is used to sort the primary sample data according to the collection time, set the occurrence node and recovery node of the abnormal marketing data as discontinuity points, and then segment the primary sample data according to the discontinuity points to obtain multiple sample segments;

[0117] A feature extraction module, the feature extraction module is used to select an evaluation segment from multiple sample segments, and mark the marketing data in the evaluation segment as secondary sample data, and then collect the fluctuation amount and fluctuation frequency of the secondary sample data, and mark them as the first feature parameter and the second feature parameter respectively;

[0118] An updating module, the updating module is used to perform a combined operation on the first characteristic parameter and the second characteristic parameter, and output the operation result as an evaluation parameter, and update the data collection interval according to the evaluation parameter, and perform subsequent sampling according to the updated data collection interval;

[0119] Verification module: The verification module is used to match the verification period according to the updated data collection interval, and perform deviation distribution analysis on the marketing data within the verification period to obtain verification parameters, and then output the execution status of the updated data collection interval according to the verification parameters, and issue an alarm signal when the execution status is abnormal.

[0120] As mentioned above, the system includes a data acquisition module, a sample segmentation module, a feature extraction module, an update module and a verification module. The data acquisition module is responsible for collecting the marketing data of the enterprise and marking these data as primary sample data. The primary sample data includes normal marketing data and abnormal marketing data. The sample segmentation module processes the primary sample data. First, the primary sample data is sorted according to the acquisition time, and then the occurrence node and recovery node of the abnormal marketing data are set as discontinuity points. On this basis, the primary sample data is segmented according to the discontinuity points to obtain multiple sample segments. The feature extraction module selects an evaluation segment from the multiple sample segments and marks the marketing data in the evaluation segment as secondary sample data. In addition, this module is also responsible for collecting the fluctuation amount and fluctuation frequency of the secondary sample data, and using them as the first characteristic parameter and the second characteristic parameter respectively. The update module is responsible for combining the first characteristic parameter and the second characteristic parameter, and outputting the calculation result as the evaluation parameter. According to the evaluation parameter, the system can automatically adjust the data collection interval to improve the efficiency of data collection. Finally, the verification module matches the verification period according to the updated data collection interval, and performs deviation distribution analysis on the marketing data within the verification period to obtain the verification parameters. According to the verification parameters, the system can output the execution status of the updated data collection interval. When the execution status is abnormal, an alarm signal will be issued to remind relevant personnel to deal with it in time.

[0121] See also Figure 3 , an electronic device, the electronic device comprising:

[0122] at least one processor;

[0123] and a memory communicatively coupled to the at least one processor;

[0124] Among them, the memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the above-mentioned enterprise marketing data management method based on big data.

[0125] The processor in the electronic device may be a central processing unit (CPU), a graphics processing unit (GPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities. The processor may directly or indirectly call a computer program stored in a memory to implement corresponding functions. The memory may include a high-speed random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. The memory may also be at least one storage device located away from the aforementioned processor. When the computer program stored in the memory is called and executed, the above-mentioned enterprise marketing data management method based on big data can be implemented. In addition, the electronic device may also include components such as input and output devices, a network adapter, and an operator. The input and output devices may include a display screen, a keyboard, a mouse, a touch screen, etc., for receiving instructions or data input by a user and displaying processing results. The network adapter is used to connect the electronic device to an external network to realize data transmission and communication. The operator may be a microprocessor, a microcontroller or other components with computing capabilities, for performing various arithmetic and logical operations to support the normal operation of the electronic device. Through the coordinated work of these components, the electronic device can effectively execute the above-mentioned enterprise marketing data management method based on big data.

[0126] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0127] The above is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the art unless otherwise specified and limited.

Claims

1. A method for enterprise marketing data management based on big data, characterized by: include: Collecting the marketing data of the enterprise, and marking the marketing data as primary sample data, wherein the primary sample data includes normal marketing data and abnormal marketing data; The first-level sample data is sorted according to the collection time, and the occurrence node and the recovery node of the abnormal marketing data are set as discontinuity points, and then the first-level sample data is segmented according to the discontinuity points to obtain multiple sample segments; Selecting an evaluation segment from the plurality of sample segments, and marking the marketing data in the evaluation segment as secondary sample data, and then collecting the fluctuation amount and fluctuation frequency of the secondary sample data, and marking them as the first characteristic parameter and the second characteristic parameter respectively; Performing a combined operation on the first characteristic parameter and the second characteristic parameter, and outputting the operation result as an evaluation parameter, and updating the data collection interval according to the evaluation parameter, and performing subsequent sampling according to the updated data collection interval; Matching the verification period according to the updated data collection interval, and performing deviation distribution analysis on the marketing data within the verification period to obtain verification parameters, and then outputting the execution status of the updated data collection interval according to the verification parameters, and issuing an alarm signal when the execution status is abnormal; The step of combining the first characteristic parameter and the second characteristic parameter and outputting the result of the operation as an evaluation parameter includes: Obtaining the first characteristic parameter, and calculating an average value of the first characteristic parameter to obtain an average fluctuation amount; Obtaining a maximum value of the first characteristic parameters; Obtaining an evaluation function, and inputting the average fluctuation amount, the second characteristic parameter, and the maximum value of the first characteristic parameter into the evaluation function, and calibrating the output result thereof as the evaluation parameter; The step of updating the data collection interval according to the evaluation parameters comprises: Obtaining the evaluation parameter, and comparing the evaluation parameter with a preset evaluation threshold; If the evaluation parameter is greater than a preset evaluation threshold, it indicates that the current data collection interval is large, and the data collection interval is shortened; If the evaluation parameter is less than or equal to the preset evaluation threshold, it indicates that the current data collection interval is moderate, and the current data collection interval is maintained; The step of matching the verification period according to the updated data collection interval and performing deviation distribution analysis on the marketing data within the verification period to obtain the verification parameters includes: Performing a difference processing on the evaluation parameter and a preset evaluation threshold to obtain a deviation parameter; Acquire a scoring interval, compare the deviation parameter with the scoring interval, and output an evaluation score of the deviation parameter according to the comparison result, wherein a plurality of scoring intervals and evaluation scores are provided; Obtaining an updated data collection interval, and matching a verification period according to the data collection interval, wherein a plurality of verification periods are provided, and correspond one to one with the evaluation scores; Acquire the real-time marketing data within the verification period, and after the verification period ends, calculate the average value and median of the real-time marketing data; Calculating the deviation of the marketing data based on the mean and the median, and recording the deviation as a verification parameter; A verification threshold is obtained, and the verification threshold is compared with a verification parameter. When the verification parameter is less than or equal to the verification threshold, it indicates that the distribution of real-time marketing data within the verification period is stable, and the execution status of the updated data collection interval is recorded as a normal state. Otherwise, it indicates that the distribution of real-time marketing data within the verification period is unstable, and the execution status of the updated data collection interval is recorded as an abnormal state.

2. The enterprise marketing data management method based on big data according to claim 1, characterized in that: After the first-level sample data is sorted, a difference process is performed one by one according to the sorting results to obtain the data fluctuation amount; According to the sorting result of the primary sample data, the data fluctuation amounts are grouped to obtain multiple groups of continuous data fluctuation amounts, and recorded as screening condition parameters; Obtaining a screening interval, and comparing the screening interval with each set of screening condition parameters; If the screening condition parameters of the same group are all within the screening range, it indicates that the primary sample data fluctuates normally, and the primary sample data under normal fluctuation is recorded as normal marketing data; If there are screening condition parameters that are not within the screening interval among the screening condition parameters in the same group, then the proportion of the screening condition parameters that are not within the screening interval is calculated; An evaluation threshold is obtained, and the evaluation threshold is compared with the proportion of the screening condition parameters that are not within the screening interval. When the proportion of the screening condition parameters that are not within the screening interval is less than the evaluation threshold, the corresponding first-level sample data is recorded as normal marketing data; otherwise, it indicates that the first-level sample data fluctuates normally or abnormally, and the first-level sample data under abnormal fluctuations is recorded as abnormal marketing data.

3. The enterprise marketing data management method based on big data according to claim 1 is characterized by: The step of segmenting the primary sample data according to the discontinuity points to obtain a plurality of sample segments comprises: Obtaining the time intervals between each of the recovery nodes and the occurrence nodes adjacent to the recovery node and located after the recovery node, and recording them as the reference duration period; Obtaining the duration of each of the reference duration periods and calibrating them as screening condition parameters; Obtaining a screening threshold, and comparing the screening threshold with a screening condition parameter; If the screening condition parameter is greater than the screening threshold, it indicates that the primary sample data under the reference duration period is long-term time-varying data, and the reference duration period is recorded as a sample segment; If the screening condition parameter is less than or equal to the screening threshold, it indicates that the primary sample data under the reference duration period is short-term data, and the short-term data and its corresponding reference duration period are simultaneously screened out.

4. The enterprise marketing data management method based on big data according to claim 1, characterized in that: The step of selecting an evaluation segment from the plurality of sample segments and marking the marketing data in the evaluation segment as secondary sample data comprises: Obtain all the sample segments and sort them according to their duration; Obtain the sample segment with the shortest duration, and mark the duration as the standard duration; Based on the standard duration, the end node of each sample segment is backtracked and offset, and the offset result is recorded as an evaluation segment, and the marketing data in the evaluation segment is recorded as secondary sample data.

5. The enterprise marketing data management method based on big data according to claim 1, characterized in that: The step of collecting the fluctuation amount and the fluctuation frequency of the secondary sample data and marking them as the first characteristic parameter and the second characteristic parameter respectively includes: Acquire all the secondary sample data, perform difference processing on two adjacent secondary sample data, obtain the fluctuation amount of the secondary sample data, and record it as the first characteristic parameter; According to the collection order of the secondary sample data, the collection time corresponding to each of the first characteristic parameters is obtained, and the collection time interval between two adjacent first characteristic parameters is calculated to obtain the fluctuation interval; According to the fluctuation interval, the fluctuation frequency of the secondary sample data is calculated, and the fluctuation frequency is recorded as a second characteristic parameter.

6. An enterprise marketing data management system based on big data, applied to the enterprise marketing data management method based on big data according to any one of claims 1 to 5, characterized in that: include: A data collection module, wherein the data collection module is used to collect the marketing data of the enterprise and mark the marketing data as primary sample data, wherein the primary sample data includes normal marketing data and abnormal marketing data; A sample segmentation module, the sample segmentation module is used to sort the primary sample data according to the collection time, and set the occurrence node and the recovery node of the abnormal marketing data as discontinuity points, and then segment the primary sample data according to the discontinuity points to obtain multiple sample segments; A feature extraction module, the feature extraction module is used to select an evaluation segment from the plurality of sample segments, and mark the marketing data in the evaluation segment as secondary sample data, and then collect the fluctuation amount and fluctuation frequency of the secondary sample data, and mark them as a first feature parameter and a second feature parameter respectively; An updating module, the updating module being used to perform a combined operation on the first characteristic parameter and the second characteristic parameter, and output the operation result as an evaluation parameter, and update the data collection interval according to the evaluation parameter, and perform subsequent sampling according to the updated data collection interval; A verification module is used to match the verification period according to the updated data collection interval, and perform deviation distribution analysis on the marketing data within the verification period to obtain verification parameters, and then output the execution status of the updated data collection interval according to the verification parameters, and issue an alarm signal when the execution status is abnormal.

7. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; Wherein, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the enterprise marketing data management method based on big data as described in any one of claims 1 to 5.

Images