Tea brand loyalty evaluation method and system based on consumer interaction data

Abstract

The invention relates to the technical field of consumer sales management, and discloses a consumer interaction data-based tea brand loyalty evaluation method and system, and the system comprises a tea brand loyalty measurement parameter collection management module, a tea brand loyalty parameter measurement management module, and a tea brand sales strategy management module. Tea brand sales strategy intelligent scientific analysis is carried out by combining consumer tea brand loyalty analysis parameters with numerical value matching and tea brand sales strategy information stored based on big data according to set time, so that the tea brand sales strategy is scientifically formulated based on the consumer tea brand loyalty parameters, and tea brand management intelligence and flexibility are improved; according to the tea brand sales strategy identification parameters, in combination with the shopping management platform, tea brand sales strategy updating management operation is autonomously and efficiently executed, the tea brand market sales strategy is accurately and dynamically updated along with market demands, and the tea brand management quality and competitiveness are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of consumer marketing management, and in particular to a tea brand loyalty evaluation method and system based on consumer interaction data. Background Art

[0002] Brand loyalty is a metric used to measure brand loyalty. It is formed when consumers repeatedly purchase and use a brand over a long period of time, developing a certain level of trust, commitment, emotional connection, and even emotional dependence on the brand. Customers with high brand loyalty are less price-sensitive, willing to pay a premium for quality, recognize the brand's value, view it as a friend and partner, and are willing to contribute to the brand. Brand loyalty management can effectively enhance consumer brand recognition and improve product quality. In the tea product sales sector, tea brands need to manually and scientifically calculate consumer loyalty based on interaction data between consumers and tea brands regarding tea products, thereby providing data support for tea brands to continuously improve and optimize their tea products. However, existing tea brand loyalty evaluation methods cannot efficiently and scientifically calculate tea brand loyalty measurement indicator parameters based on consumer shopping history, nor can they intelligently and accurately measure tea brand loyalty parameters, let alone scientifically implement tea brand sales strategy update management.

[0003] The Chinese invention patent application with publication number CN111784387A discloses a consumer brand loyalty analysis method based on multi-dimensional big data. By collecting facial images and combining the POS system with the facial database, the number of valid store visitors, the number of valid store members, the number of new purchases and the number of repeat purchases are obtained, and the brand loyalty analysis results are obtained by combining the big data regression analysis method. The above technical solution mainly considers the repurchase rate when statistically analyzing brand loyalty parameters and cannot scientifically evaluate brand loyalty information. At the same time, it is inappropriate to collect consumer facial information without consent. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In order to solve the problems that the above-mentioned existing tea brand loyalty evaluation operations cannot achieve efficient and scientific statistics of tea brand loyalty measurement index parameters based on consumer historical shopping information, nor can they intelligently and accurately measure tea brand loyalty parameters, and cannot scientifically execute tea brand sales strategy update management operations, the above-mentioned accurate collection of tea brand loyalty evaluation setting time length information, accurate collection of set time consumer historical shopping information, accurate search of consumer tea product repurchase frequency information, efficient search of consumer tea product purchase decision time information, accurate search of consumer tea product purchase satisfaction information, autonomous search of consumer tea product purchase recommendation willingness value information, scientific search of consumer tea product purchase complaint frequency information, scientific statistics of consumer tea product repurchase frequency parameters, efficient measurement and scientific statistics of consumer tea product purchase complaint rate parameters, online statistics of consumer tea brand loyalty measurement index mean parameters, intelligent analysis of set time consumer tea brand loyalty parameters, accurate update of tea brand sales strategy parameters, and scientific execution of tea brand sales strategy adjustment management operations are achieved.

[0006] (2) Technical solution

[0007] The present invention is implemented through the following technical solution: a tea brand loyalty evaluation method based on consumer interaction data, the method comprising the following steps:

[0008] S1. Collecting tea brand loyalty evaluation data for a set time interval;

[0009] S2. Searching and processing the consumer's historical shopping information within the set time period based on the tea brand loyalty evaluation set time interval data and the consumer's historical shopping data to generate the consumer's historical shopping data within the set time period;

[0010] S3. Search and process the shopping behavior information and shopping feedback information of tea brand consumers based on the consumer shopping history data over a set period of time, and generate data on the number of tea product repurchases by consumers, the length of time it takes for consumers to make tea product purchase decisions, consumer tea product purchase satisfaction data, consumer tea product purchase recommendation willingness data, and the number of tea product purchase complaints by consumers;

[0011] S4. Based on the tea brand loyalty evaluation set time interval data, the consumer tea product repurchase frequency data, and the consumer tea product purchase complaint frequency data, perform numerical measurement processing on the consumer tea product repurchase frequency and the consumer tea product purchase complaint rate, and generate consumer tea product repurchase frequency data and consumer tea product purchase complaint rate data;

[0012] S5. Perform numerical measurement processing on the mean values of the measurement index parameters of the consumer tea brand loyalty based on the parameters of the consumer tea product repurchase frequency, purchase decision time, purchase satisfaction, purchase recommendation intention, and purchase complaint rate, to generate the mean value of the consumer tea brand loyalty measurement index;

[0013] S6. Performing numerical analysis of consumer tea brand loyalty within a set time period based on the average value of the consumer tea brand loyalty measurement index and the consumer tea brand loyalty data to generate consumer tea brand loyalty analysis data within the set time period;

[0014] S7. Perform tea brand sales strategy identification processing based on the tea brand loyalty analysis data of consumers at the set time and the tea brand sales strategy data, generate tea brand sales strategy identification data and execute tea brand sales strategy update management operations.

[0015] Preferably, the steps for collecting tea brand loyalty evaluation set time interval data are as follows:

[0016] S11. Collect the set time interval parameters of the target tea brand loyalty evaluation operation online through the shopping management platform information entry dialog box, and generate tea brand loyalty evaluation set time interval data O = [o1, o2], where o1 represents the tea brand loyalty evaluation set start time point data, and o2 represents the tea brand loyalty evaluation set end time point data, where the units of o1 and o2 are both composed of years, months, and days, and the shopping management platform includes any one of an e-commerce ERP management system, a B2C e-commerce system, and a WeChat shopping management system.

[0017] Preferably, the steps of searching and processing the consumer's historical shopping information within the set time period based on the tea brand loyalty evaluation set time interval data and the consumer's historical shopping data to generate the consumer's historical shopping data within the set time period are as follows:

[0018] S21. Establish a consumer history shopping data set A = (a1,…,a m ,…,a η ), m=1,2,3,…,eta; where a mrepresents the consumer shopping history data corresponding to the mth consumer, n represents the maximum number of consumers, and the consumer shopping history data includes the consumer's purchase time and date information, purchase record information, purchase decision time length information, purchase feedback satisfaction information, purchase feedback recommendation willingness value information, and purchase feedback complaint number information of the consumer's historical purchase of tea products; wherein the purchase time and date information is composed of year, month, and day, the purchase decision time length information represents the total time length information from the consumer opening the shopping management platform interface to completing the tea product order, and the unit of the purchase decision time length information is seconds, the purchase feedback satisfaction information represents a positive integer in the range of [1, 10], and the purchase feedback recommendation willingness value information represents a positive integer in the range of [1, 10];

[0019] S22, using the KD tree nearest neighbor search algorithm, compare the tea brand loyalty evaluation setting start time point data o1 to the tea brand loyalty evaluation setting end time point data o2 in the tea brand loyalty evaluation setting time interval data O with the consumer history shopping data a in the consumer history shopping data set A. m According to the time and date value matching, search for all the consumer historical shopping data a within the set time interval corresponding to the tea brand loyalty evaluation setting start time point data o1 to the tea brand loyalty evaluation setting end time point data o2 m , and construct a set of consumer shopping history data sets A'=(a' m1 ,…,a' m2 ), 1≤m1≤m≤m2≤η; where a' m1 represents the historical shopping data of the consumer at the set time corresponding to the m1th consumer within the set time, a′ m2 Represents the historical shopping data of the consumer at the set time corresponding to the m2th consumer within the set time.

[0020] Preferably, the steps of searching and processing the shopping behavior information and shopping feedback information of tea brand consumers based on the historical shopping data of consumers over a set time period, and respectively generating data on the number of consumers' tea product repurchases, the length of time it takes for consumers to make tea product purchase decisions, the satisfaction data on consumers' tea product purchases, the willingness value of consumers to recommend tea products for purchase, and the number of consumers' tea product purchase complaints are as follows:

[0021] S31, using the BERT language model algorithm to analyze the set time consumer historical shopping data a' in the set time consumer historical shopping data set A'. m1 to a′ m2The shopping behavior information and shopping feedback information of tea brand consumers within a set time are searched and processed according to the keywords of the number of tea product purchases, the purchase decision time, the purchase feedback satisfaction, the purchase feedback recommendation willingness value and the purchase feedback complaint number, and the consumer tea product repurchase frequency data set B = (b m1 ,…,b m2 ), where b m1 represents the number of times the m1th consumer repurchases tea products, b m2 b represents the number of times the consumer repurchases tea products corresponding to the m2th consumer; m1 and b m2 The unit of is times;

[0022] The data set C of the length of time for consumers to make tea product purchase decisions is (c m1 ,…,c m2 ), where c m1 represents the length of time for the tea product purchase decision of the m1th consumer, c m2 represents the length of time for the tea product purchase decision of the m2th consumer; c m1 and c m2 The unit of is seconds;

[0023] Consumer tea product purchase satisfaction data set D = (d m1 ,…,d m2 ), where d m1 represents the consumer tea product purchase satisfaction data corresponding to the m1th consumer, d m2 represents the consumer tea product purchase satisfaction data corresponding to the m2th consumer; d m1 and d m2 All are positive integers in [1,10];

[0024] Consumers' tea product purchase recommendation willingness value data set E = (e m1 ,…,e m2 ), where e m1 represents the consumer tea product purchase recommendation value data corresponding to the m1th consumer, e m2 represents the consumer tea product purchase recommendation value data corresponding to the m2th consumer; e m1 and e m2 All are positive integers in [1,10];

[0025] The data set of consumers’ complaints about tea product purchases is F = (f m1 ,…,f m2 ), where f m1represents the number of complaints about tea products purchased by the m1th consumer, f m2 represents the number of complaints about tea products purchased by the m2th consumer, f m1 and f m2 The unit is times.

[0026] Preferably, the steps of performing numerical measurement processing on the consumer tea product repurchase frequency and the consumer tea product purchase complaint rate based on the tea brand loyalty evaluation set time interval data, the consumer tea product repurchase frequency data, and the consumer tea product purchase complaint frequency data, and generating the consumer tea product repurchase frequency data and the consumer tea product purchase complaint rate data are as follows:

[0027] S41. Performing numerical measurement processing on the total time length of the tea brand loyalty evaluation setting time interval from the tea brand loyalty evaluation setting start time point data o1 to the tea brand loyalty evaluation setting end time point data o2 in the tea brand loyalty evaluation setting time interval data O, and generating tea brand loyalty evaluation setting time length data R, where R = o2 - o1 + 1, and the unit of R is day;

[0028] S42, respectively, the consumer tea product repurchase frequency data b in the consumer tea product repurchase frequency data set B m1 to b m2 The tea brand loyalty evaluation set time length data R is processed by the number of repurchases and the set time length value, and the consumer tea product repurchase frequency data set B'=(b' m1 ,…,b' m2 ), where b' m1 represents the repurchase frequency data of tea products corresponding to the m1th consumer, b' m2 represents the repurchase frequency data of tea products corresponding to the m2th consumer, b' m1 and b' m2 The unit is times per day;

[0029] The number of consumer complaints about tea product purchases data f in the number of consumer complaints about tea product purchases data set F is m1 to f m2 The number of complaints and the set time length value are processed as a quotient with the tea brand loyalty evaluation set time length data R, and the consumer tea product purchase complaint rate data set F'=(f' m1 ,…,f' m2 ), where f' m1 represents the repurchase frequency data of tea products corresponding to the m1th consumer, f' m2represents the repurchase frequency data of tea products corresponding to the m2th consumer, f' m1 and f' m2 The unit is times per day.

[0030] Preferably, the steps of performing numerical measurement processing on the mean of the measurement index parameters of the consumer tea brand loyalty based on the repurchase frequency parameter of the consumer tea product, the purchase decision time parameter, the purchase satisfaction parameter, the purchase recommendation intention value, and the purchase complaint rate parameter to generate the mean of the consumer tea brand loyalty measurement index are as follows:

[0031] S51, based on the consumer tea product repurchase frequency data b' in the consumer tea product repurchase frequency data set B. m1 to b' m2 , the consumer tea product purchase decision time length data c in the consumer tea product purchase decision time length data set C m1 to c m2 , the consumer tea product purchase satisfaction data d in the consumer tea product purchase satisfaction data set D m1 to d m2 , the consumer tea product purchase recommendation willingness value data set E in the consumer tea product purchase recommendation willingness value data set e m1 to e m2 , the consumer tea product purchase complaint rate data f' in the consumer tea product purchase complaint rate data set F m1 to f' m2 The parameters of the repurchase frequency measurement index, purchase decision time measurement index, purchase satisfaction measurement index, purchase recommendation intention value measurement index and purchase complaint rate measurement index of consumer tea brand loyalty are measured and processed respectively, and the mean value set of consumer tea brand loyalty measurement index is generated. in It represents the mean value of the repurchase frequency measurement index of consumers’ tea brand loyalty. The unit is times per day; It represents the mean value of the measurement index of the length of time for consumers’ tea brand loyalty purchase decision. The unit is seconds; represents the mean value of the consumer tea brand loyalty purchase satisfaction measurement index, It represents the mean value of the measurement index of consumers’ tea brand loyalty and purchase recommendation intention. It represents the mean value of the measurement index of consumers’ tea brand loyalty and purchase recommendation intention. The unit is times per day.

[0032] Preferably, the numerical analysis of consumer tea brand loyalty within a set time period is performed based on the average value of the consumer tea brand loyalty measurement index and the consumer tea brand loyalty data, and the operating steps for generating the consumer tea brand loyalty analysis data within the set time period are as follows:

[0033] S61. Establish consumer tea brand loyalty data set H = (h1,…,h n ,…,h ι ), n=1,2,3,…,ι; where h n represents the consumer tea brand loyalty data corresponding to the nth loyalty measurement indicator combination information type, ι represents the maximum number of loyalty measurement indicator combination information types, and the loyalty measurement indicator combination information type represents a data combination type generated by combining the consumer tea brand loyalty repurchase frequency measurement indicator parameters, purchase decision time measurement indicator parameters, purchase satisfaction measurement indicator parameters, purchase recommendation willingness value measurement indicator parameters, and purchase recommendation willingness value measurement indicator parameters for analyzing tea brand loyalty parameters; the consumer tea brand loyalty data represents a parameter of the consumer's recognition of the tea brand, and h n The value of h is a random number in [0,1]. n The value of is positively correlated with consumers’ recognition of tea brands;

[0034] S62: The average value of the consumer tea brand loyalty repurchase frequency measurement index in the consumer tea brand loyalty measurement index mean value set P is calculated. Mean measurement index of the length of time for consumers' tea brand loyalty purchase decision The mean of the consumer tea brand loyalty purchase satisfaction measurement index Mean value of the consumer tea brand loyalty purchase recommendation intention measurement index and the mean value of the consumer tea brand loyalty purchase recommendation intention measurement index The consumer tea brand loyalty data h in the consumer tea brand loyalty data set H n Compare the parameters of the loyalty measurement index and search for the average value of the consumer tea brand loyalty repurchase frequency measurement index Mean measurement index of the length of time for consumers' tea brand loyalty purchase decision The mean of the consumer tea brand loyalty purchase satisfaction measurement index Mean value of the consumer tea brand loyalty purchase recommendation intention measurement index and the mean value of the consumer tea brand loyalty purchase recommendation intention measurement index The corresponding consumer tea brand loyalty data h n, and generate the set time consumer tea brand loyalty analysis data h after data identification fenxi , execute to generate the set time consumer tea brand loyalty analysis data h fenxi The specific steps are as follows:

[0035] S621, initialization, update the maximum number of iterations of the algorithm;

[0036] S622, grazing behavior, the foal grazes in the herd. To perform grazing behavior, the stallion is set as the center of the grazing area, and the brand loyalty parameter matching horse individuals are searched around the center, that is, with the stallion as the search center, the search space H of the consumer tea brand loyalty data set is searched for the mean of the consumer tea brand loyalty repurchase frequency measurement index. Mean measurement index of the length of time for consumers' tea brand loyalty purchase decision The mean of the consumer tea brand loyalty purchase satisfaction measurement index Mean value of the consumer tea brand loyalty purchase recommendation intention measurement index and the mean value of the consumer tea brand loyalty purchase recommendation intention measurement index Matched consumer tea brand loyalty data h n , grazing behavior simulation, simulates the behavior of matching brand loyalty parameters with individual horses moving at different radii and searching for the leading wild horse. The grazing behavior simulation formula is as follows:

[0037] in Stallion represents the simulated position of the jth brand loyalty parameter matching horse group individual in the i-dimensional space, that is, the simulated position of the jth brand loyalty parameter matching horse group individual in the consumer tea brand loyalty data set H search space with spatial dimension ι. j represents the position of the leading wild horse in the herd in the search space of the consumer tea brand loyalty data set H, represents the current position of the j-th brand loyalty parameter matching horse group individual in the i-dimensional space, that is, the current position of the j-th brand loyalty parameter matching horse group individual in the consumer tea brand loyalty data set H search space with spatial dimension ι, Π represents the adaptive coefficient, γ represents a random number in the interval [-2, 2], Where ▽ represents a vector consisting of 0 and 1, are all random vectors uniformly distributed in the interval [0,1], γ2 is a random value in the interval [0,1], and ζ is a random vector that satisfies the condition , ψ is an adaptive parameter that decreases from 1 to 0 as the number of iterations increases. Where t represents the number of iterations, and T represents the maximum number of iterations;

[0038] S623. Wild horses perform mating behavior in the search space of the consumer tea brand loyalty data set H. The mating behavior simulation formula is as follows: in, Represents a herd of horses The brand loyalty parameter matches the individual position of the horse group individual σ in the search space of the consumer tea brand loyalty data set H after leaving the group and re-entering the horse group. Crossover represents the position random function. Represents a herd of horses The brand loyalty parameter matching horse group individual σ after leaving the group re-enters the individual position of the horse group τ in the search space of the consumer tea brand loyalty data set H, Represents a herd of horses The brand loyalty parameter matches the individual position of the horse group υ after leaving the group and re-entering the horse group τ in the search space of the consumer tea brand loyalty data set H;

[0039] S624, return to the team leader, and the leading wild horse leads the members to a better habitat, that is, search for the mean of the repurchase frequency measurement index of the consumer tea brand loyalty in the search space of the consumer tea brand loyalty data set H Mean measurement index of the length of time for consumers' tea brand loyalty purchase decision The mean of the consumer tea brand loyalty purchase satisfaction measurement index Mean value of the consumer tea brand loyalty purchase recommendation intention measurement index and the mean value of the consumer tea brand loyalty purchase recommendation intention measurement index Matched consumer tea brand loyalty data h n ;

[0040] S625, selection of the leading wild horse, randomly selecting the leading wild horse to maintain the randomness of the algorithm;

[0041] S626: When the algorithm meets the maximum number of iterations, the average value of the consumer tea brand loyalty repurchase frequency measurement index is output. Mean measurement index of the length of time for consumers' tea brand loyalty purchase decision The mean of the consumer tea brand loyalty purchase satisfaction measurement index Mean value of the consumer tea brand loyalty purchase recommendation intention measurement index and the mean value of the consumer tea brand loyalty purchase recommendation intention measurement index Matched consumer tea brand loyalty data h n ; Otherwise, continue executing S622 to S624 until the maximum number of iterations is met.

[0042] S627, the consumer tea brand loyalty data h output in step S626 n After data identification, the set time consumer tea brand loyalty analysis data h is generated fenxi , where h fenxi The value of is a random number in [0,1].

[0043] Preferably, the steps of performing tea brand sales strategy identification processing based on the tea brand loyalty analysis data of consumers at the set time and the tea brand sales strategy data, generating tea brand sales strategy identification data and executing the tea brand sales strategy update management task are as follows:

[0044] S71. Establish tea brand sales strategy data set G = (g1,…,g k ,…,g λ ), k=1,2,3,…,λ; where g k represents the tea brand sales strategy data corresponding to the k-th tea brand loyalty value type, λ represents the maximum number of tea brand loyalty value types, and the tea brand sales strategy data represents the optimal tea product sales plan information set based on the consumer's tea brand loyalty parameter standard;

[0045] S72, the set time consumer tea brand loyalty analysis data h fenxi and the tea brand sales strategy data g in the tea brand sales strategy data set G k Perform tea brand loyalty value comparison to identify the tea brand loyalty analysis data h of consumers at the set time fenxi The corresponding tea brand sales strategy data g k , and generate tea brand sales strategy identification data through data identification

[0046] S73, the shopping management platform identifies the tea brand sales strategy data The information is pushed to the tea brand's offline sales store management terminal through the mobile communication network, and the tea brand sales strategy update management task is executed.

[0047] A tea brand loyalty evaluation system based on consumer interaction data, used to implement the tea brand loyalty evaluation method based on consumer interaction data, the system comprising a tea brand loyalty measurement parameter collection management module, a tea brand loyalty parameter measurement management module, and a tea brand sales strategy management module;

[0048] The tea brand loyalty measurement parameter collection and management module includes a tea brand loyalty evaluation set time range collection unit, a consumer history shopping information storage unit, a set time consumer history shopping information search unit, a consumer tea product repurchase frequency search unit, a consumer tea product purchase decision time search unit, a consumer tea product purchase satisfaction information search unit, a consumer tea product purchase recommendation willingness value information search unit, and a consumer tea product purchase complaint frequency search unit;

[0049] The tea brand loyalty evaluation setting time range collection unit collects tea brand loyalty evaluation setting time interval data through the shopping management platform; the consumer history shopping information storage unit is used to store consumer history shopping data; the set time consumer history shopping information search unit searches for consumer history shopping information within the set time based on the tea brand loyalty evaluation setting time interval data and consumer history shopping data to generate set time consumer history shopping data; the consumer tea product repurchase frequency search unit searches for shopping behavior information and shopping feedback information of tea brand consumers based on set time consumer history shopping data to generate consumer tea product repurchase frequency data; the consumer tea product purchase decision time search unit searches for tea brand consumers based on set time consumer history shopping data The shopping behavior information and shopping feedback information of consumers are searched and processed to generate data on the length of time for consumers to make tea product purchase decisions; the consumer tea product purchase satisfaction information search unit searches and processes the shopping behavior information and shopping feedback information of tea brand consumers based on the consumer historical shopping data of set time to generate data on consumer tea product purchase satisfaction; the consumer tea product purchase recommendation willingness value information search unit searches and processes the shopping behavior information and shopping feedback information of tea brand consumers based on the consumer historical shopping data of set time to generate data on the consumer tea product purchase recommendation willingness value; the consumer tea product purchase complaint frequency search unit searches and processes the shopping behavior information and shopping feedback information of tea brand consumers based on the consumer historical shopping data of set time to generate data on the number of consumer tea product purchase complaints;

[0050] The tea brand loyalty parameter measurement and management module includes a consumer tea product repurchase frequency measurement unit, a consumer tea product purchase complaint rate measurement unit, a consumer tea brand loyalty measurement index mean measurement unit, a consumer tea brand loyalty storage unit, and a set time consumer tea brand loyalty analysis unit;

[0051] The consumer tea product repurchase frequency measurement unit performs numerical measurement processing on the consumer tea product repurchase frequency based on the tea brand loyalty evaluation set time interval data and the consumer tea product repurchase frequency data, and generates consumer tea product repurchase frequency data; the consumer tea product purchase complaint rate measurement unit performs numerical measurement processing on the consumer tea product purchase complaint rate based on the tea brand loyalty evaluation set time interval data and the consumer tea product purchase complaint frequency data, and generates consumer tea product purchase complaint rate data; the consumer tea brand loyalty measurement index mean measurement unit performs numerical measurement processing on the consumer tea brand loyalty measurement index parameter mean based on the consumer tea product repurchase frequency parameter, purchase decision time length parameter, purchase satisfaction parameter, purchase recommendation intention value and purchase complaint rate parameter, and generates the consumer tea brand loyalty measurement index mean; the consumer tea brand loyalty storage unit is used to store consumer tea brand loyalty data; the set time consumer tea brand loyalty analysis unit performs numerical analysis processing on the consumer tea brand loyalty within the set time based on the consumer tea brand loyalty measurement index mean and the consumer tea brand loyalty data, and generates the set time consumer tea brand loyalty analysis data;

[0052] The tea brand sales strategy management module includes a tea brand sales strategy storage unit, a target tea brand sales strategy identification unit, and a tea brand sales strategy update management job execution unit;

[0053] The tea brand sales strategy storage unit is used to store tea brand sales strategy data; the target tea brand sales strategy identification unit performs tea brand sales strategy identification processing based on the set time consumer tea brand loyalty analysis data and tea brand sales strategy data to generate tea brand sales strategy identification data; the tea brand sales strategy update management job execution unit executes the tea brand sales strategy update management job based on the tea brand sales strategy identification data in combination with the shopping management platform.

[0054] (3) Beneficial effects

[0055] The present invention provides a tea brand loyalty evaluation method and system based on consumer interaction data. It has the following beneficial effects:

[0056] 1. Obtain tea brand loyalty evaluation parameters online through the shopping management platform, set time interval parameters, combine with intelligent search algorithms, and conduct efficient and accurate searches on consumers' historical shopping information within the set time period, thereby improving the efficiency of tea brand loyalty evaluation results; adopt intelligent search algorithms combined with data keyword classification to accurately retrieve information on the number of repeat purchases of tea products, length of purchase decision time, purchase satisfaction, purchase recommendation intention value, and number of purchase complaints from consumers' historical shopping information within the set time period, thereby achieving a comprehensive and scientific search of tea brand loyalty evaluation index parameters and improving the authenticity of tea brand loyalty evaluation results.

[0057] 2. By scientifically calculating the repurchase frequency information and purchase complaint rate parameters of consumer tea products within a set time based on numerical measurement, the parameters of consumer tea brand loyalty measurement indicators can be independently and efficiently calculated; based on the repurchase frequency parameters, purchase decision time length parameters, purchase satisfaction parameters, purchase recommendation intention value and purchase complaint rate parameters of consumer tea products, the mean value of consumer tea brand loyalty measurement indicator parameters can be intelligently and accurately calculated to achieve scientific and accurate statistics of consumer tea brand loyalty measurement indicators and improve the accuracy of consumer tea brand loyalty evaluation; based on the mean value of consumer tea brand loyalty measurement indicators combined with intelligent recognition algorithms and scientifically preset consumer tea brand loyalty parameters, numerical intelligent analysis of consumer tea brand loyalty within a set time can be carried out to achieve comprehensive and accurate evaluation of consumer tea brand loyalty results based on multiple measurement indicator parameters, improve the scientific nature of tea brand loyalty evaluation results, and enhance the efficiency and accuracy of tea brands in responding to consumer demand feedback.

[0058] 3. Through combining numerical matching with big data storage of tea brand sales strategy information based on tea brand loyalty analysis parameters of consumers at set time, intelligent and scientific analysis of tea brand sales strategies is conducted to scientifically formulate tea brand sales strategies based on consumer tea brand loyalty parameters, thereby improving the intelligence and flexibility of tea brand management; based on tea brand sales strategy identification parameters combined with the shopping management platform, the tea brand sales strategy update management operations are independently and efficiently executed, thereby realizing accurate and dynamic updates of tea brand market sales strategies in line with market demand, thereby improving the quality and competitiveness of tea brand management. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a module diagram of the tea brand loyalty evaluation system based on consumer interaction data provided by the present invention;

[0060] Figure 2 This is a flow chart of the tea brand loyalty evaluation method based on consumer interaction data provided by the present invention. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0062] The embodiments of the tea brand loyalty evaluation method and system based on consumer interaction data are as follows:

[0063] Example 1:

[0064] See also Figure 1 - Figure 2 , a tea brand loyalty evaluation method based on consumer interaction data, the method comprises the following steps:

[0065] S1. Collecting tea brand loyalty evaluation data for a set time interval;

[0066] S2. Search and process the consumer's historical shopping information within the set time period based on the tea brand loyalty evaluation data and the consumer's historical shopping data, and generate the consumer's historical shopping data within the set time period;

[0067] S3. Search and process the shopping behavior information and shopping feedback information of tea brand consumers based on the consumer shopping history data over a set period of time, and generate data on the number of tea product repurchases by consumers, the length of time it takes for consumers to make tea product purchase decisions, consumer tea product purchase satisfaction data, consumer tea product purchase recommendation willingness data, and the number of tea product purchase complaints by consumers;

[0068] S4. Based on the tea brand loyalty evaluation data, the tea product repurchase frequency data, and the tea product purchase complaint frequency data, perform numerical measurement processing on the tea product repurchase frequency and the tea product purchase complaint rate, and generate the tea product repurchase frequency data and the tea product purchase complaint rate data.

[0069] S5. Perform numerical measurement processing on the mean values of the measurement index parameters of the consumer tea brand loyalty based on the parameters of the consumer tea product repurchase frequency, purchase decision time, purchase satisfaction, purchase recommendation intention, and purchase complaint rate, to generate the mean value of the consumer tea brand loyalty measurement index;

[0070] S6. Performing numerical analysis and processing of consumer tea brand loyalty within a set time period based on the average of the consumer tea brand loyalty measurement index and the consumer tea brand loyalty data to generate consumer tea brand loyalty analysis data within the set time period;

[0071] S7. Perform tea brand sales strategy identification processing based on the tea brand loyalty analysis data of consumers at the set time and the tea brand sales strategy data, generate tea brand sales strategy identification data and execute tea brand sales strategy update management operations.

[0072] For further information, see Figure 1 - Figure 2 The steps for collecting tea brand loyalty evaluation data for a set time interval are as follows:

[0073] S11. Collect the set time interval parameters of the target tea brand loyalty evaluation task online through the shopping management platform information entry dialog box, and generate tea brand loyalty evaluation set time interval data O = [o1, o2], where o1 represents the tea brand loyalty evaluation set start time point data, and o2 represents the tea brand loyalty evaluation set end time point data, where the units of o1 and o2 are both composed of year, month, and day, and the shopping management platform includes any one of an e-commerce ERP management system, a B2C e-commerce system, and a WeChat shopping management system.

[0074] Based on the tea brand loyalty evaluation set time interval data and consumer historical shopping data, the consumer historical shopping information within the set time is searched and processed. The steps for generating the consumer historical shopping data within the set time are as follows:

[0075] S21. Establish a consumer history shopping data set A = (a1,…,a m ,…,a η ), m=1,2,3,…,eta; where a m represents the consumer shopping history data corresponding to the mth consumer, η represents the maximum number of consumers, and the consumer shopping history data includes the consumer's historical purchase time and date information of tea products, purchase record information, purchase decision time length information, purchase feedback satisfaction information, purchase feedback recommendation willingness value information, and purchase feedback complaint number information; the purchase time and date information is composed of year, month, and day, the purchase decision time length information represents the total time length information from the consumer opening the shopping management platform interface to completing the tea product order, and the purchase decision time length information is in seconds, the purchase feedback satisfaction information represents a positive integer in the range of [1, 10], and the purchase feedback recommendation willingness value information represents a positive integer in the range of [1, 10];

[0076] S22, using the KD tree nearest neighbor search algorithm, the tea brand loyalty evaluation setting time interval data O, the tea brand loyalty evaluation setting start time point data o1 to the tea brand loyalty evaluation setting end time point data o2, and the consumer historical shopping data a in the consumer historical shopping data set A. mAccording to the time and date value matching, search for all consumer historical shopping data a within the set time interval corresponding to the tea brand loyalty evaluation setting start time point data o1 to the tea brand loyalty evaluation setting end time point data o2 m , and construct a set of consumer shopping history data sets A'=(a' m1 ,…,a' m2 ), 1≤m1≤m≤m2≤η; where a' m1 represents the historical shopping data of the consumer at the set time corresponding to the m1th consumer within the set time, a′ m2 Represents the historical shopping data of the consumer at the set time corresponding to the m2th consumer within the set time.

[0077] The steps for searching and processing the shopping behavior information and shopping feedback information of tea brand consumers based on the historical shopping data of consumers over a set period of time, and generating data on the number of consumers' tea product repurchases, the length of time it takes for consumers to make tea product purchase decisions, the satisfaction data of consumers' tea product purchases, the willingness value of consumers to recommend tea products, and the number of consumers' tea product purchase complaints are as follows:

[0078] S31, using the BERT language model algorithm to analyze the set time consumer historical shopping data a' in the set time consumer historical shopping data set A'. m1 to a′ m2 The shopping behavior information and shopping feedback information of tea brand consumers within a set time are searched and processed according to the keywords of the number of tea product purchases, the purchase decision time, the purchase feedback satisfaction, the purchase feedback recommendation willingness value and the purchase feedback complaint number, and the consumer tea product repurchase frequency data set B = (b m1 ,…,b m2 ), where b m1 represents the number of times the m1th consumer repurchases tea products, b m2 b represents the number of times the consumer repurchases tea products corresponding to the m2th consumer; m1 and b m2 The unit is times; the consumer tea product purchase decision length data set C = (c m1 ,…,c m2 ), where c m1 represents the length of time for the tea product purchase decision of the m1th consumer, c m2 represents the length of time for the tea product purchase decision of the m2th consumer; c m1 and c m2 The unit is seconds; the consumer tea product purchase satisfaction data set D = (d m1 ,…,d m2), where d m1 represents the consumer tea product purchase satisfaction data corresponding to the m1th consumer, d m2 represents the consumer tea product purchase satisfaction data corresponding to the m2th consumer; d m1 and d m2 are all positive integers within [1,10]; the consumer tea product purchase recommendation willingness value data set E=(e m1 ,…,e m2 ), where e m1 represents the consumer tea product purchase recommendation value data corresponding to the m1th consumer, e m2 represents the consumer tea product purchase recommendation value data corresponding to the m2th consumer; e m1 and e m2 are all positive integers within [1,10]; the data set of consumers’ complaints about tea product purchases is F=(f m1 ,…,f m2 ), where f m1 represents the number of complaints about tea products purchased by the m1th consumer, f m2 represents the number of complaints about tea products purchased by the m2th consumer, f m1 and f m2 The unit is times.

[0079] Through the cooperation of the tea brand loyalty evaluation set time range collection unit and the set time consumer historical shopping information search unit, the shopping management platform is used to obtain the tea brand loyalty evaluation set time interval parameters online, combined with the intelligent search algorithm and the set time consumer historical shopping information to conduct efficient and accurate search, thereby improving the efficiency of the tea brand loyalty evaluation results; the consumer tea product repurchase frequency search unit, the consumer tea product purchase decision time search unit, the consumer tea product purchase satisfaction information search unit, the consumer tea product purchase recommendation intention value information search unit, and the consumer tea product purchase complaint frequency search unit cooperate with each other, and the intelligent search algorithm is used in combination with data keyword classification to accurately retrieve the consumer tea product repurchase frequency, purchase decision time length, purchase satisfaction, purchase recommendation intention value and purchase complaint frequency information in the set time consumer historical shopping information, thereby realizing a comprehensive and scientific search of tea brand loyalty evaluation index parameters and improving the authenticity of the tea brand loyalty evaluation results.

[0080] For further information, see Figure 1 - Figure 2Based on the tea brand loyalty evaluation, the time interval data, the consumer tea product repurchase frequency data, and the consumer tea product purchase complaint frequency data are set to measure the consumer tea product repurchase frequency and purchase complaint rate, and the operation steps for generating the consumer tea product repurchase frequency data and the consumer tea product purchase complaint rate data are as follows:

[0081] S41. Perform numerical measurement processing on the total time length of the tea brand loyalty evaluation setting time interval from the tea brand loyalty evaluation setting start time point data o1 to the tea brand loyalty evaluation setting end time point data o2 in the tea brand loyalty evaluation setting time interval data O, and generate tea brand loyalty evaluation setting time length data R, where R = o2 - o1 + 1, and the unit of R is day;

[0082] S42, respectively, the consumer tea product repurchase frequency data b in the consumer tea product repurchase frequency data set B m1 to b m2 The tea brand loyalty evaluation set time length data R is processed by the number of repurchases and the set time length value, and the consumer tea product repurchase frequency data set B'=(b' m1 ,…,b' m2 ), where b' m1 represents the repurchase frequency data of tea products corresponding to the m1th consumer, b' m2 represents the repurchase frequency data of tea products corresponding to the m2th consumer, b' m1 and b' m2 The unit is times per day;

[0083] The number of consumer complaints about tea product purchases in the data set F is the number of consumer complaints about tea product purchases data f m1 to f m2 The number of complaints and the set time length data R of tea brand loyalty evaluation are processed as a quotient, and the consumer tea product purchase complaint rate data set F'=(f' m1 ,…,f' m2 ), where f' m1 represents the repurchase frequency data of tea products corresponding to the m1th consumer, f' m2 represents the repurchase frequency data of tea products corresponding to the m2th consumer, f' m1 and f' m2 The unit is times per day.

[0084] The mean value of the measurement index parameters of consumer tea brand loyalty is numerically processed based on the parameters of repurchase frequency, purchase decision time, purchase satisfaction, purchase recommendation intention, and purchase complaint rate of consumer tea products. The steps for generating the mean value of the measurement index of consumer tea brand loyalty are as follows:

[0085] S51, based on the consumer tea product repurchase frequency data b' in the consumer tea product repurchase frequency data set B'. m1 to b' m2 , the length of time for consumers to make tea product purchase decisions in the data set C m1 to c m2 , consumer tea product purchase satisfaction data d in the consumer tea product purchase satisfaction data set D m1 to d m2 , Consumers’ tea product purchase recommendation willingness values in data set E

[0086] Data m1 to e m2 , Consumer tea product purchase complaint rate data f' in the consumer tea product purchase complaint rate data set F m1 to f' m2 The parameters of the repurchase frequency measurement index, purchase decision time measurement index, purchase satisfaction measurement index, purchase recommendation intention value measurement index and purchase complaint rate measurement index of consumer tea brand loyalty are measured and processed respectively, and the mean value set of consumer tea brand loyalty measurement index is generated. in It represents the mean value of the repurchase frequency measurement index of consumers’ tea brand loyalty. The unit is times per day; It represents the mean value of the measurement index of the length of time for consumers’ tea brand loyalty purchase decision. The unit is seconds; represents the mean value of the consumer tea brand loyalty purchase satisfaction measurement index, It represents the mean value of the measurement index of consumers’ tea brand loyalty and purchase recommendation intention. It represents the mean value of the measurement index of consumers’ tea brand loyalty and purchase recommendation intention. The unit is times per day.

[0087] The numerical analysis and processing of consumer tea brand loyalty within a set time period is performed based on the mean value of the consumer tea brand loyalty measurement index and the consumer tea brand loyalty data. The operation steps for generating the consumer tea brand loyalty analysis data within a set time period are as follows:

[0088] S61. Establish consumer tea brand loyalty data set H = (h1,…,h n ,…,h ι ), n=1,2,3,…,ι; where h n represents the consumer tea brand loyalty data corresponding to the nth loyalty measurement indicator combination information type, ι represents the maximum number of loyalty measurement indicator combination information types, and the loyalty measurement indicator combination information type represents the data combination type generated by combining the consumer tea brand loyalty repurchase frequency measurement indicator parameters, purchase decision time measurement indicator parameters, purchase satisfaction measurement indicator parameters, purchase recommendation willingness value measurement indicator parameters, and purchase recommendation willingness value measurement indicator parameters for analyzing tea brand loyalty parameters; consumer tea brand loyalty data represents the parameters of consumers' recognition of tea brands, and h n The value of h is a random number in [0,1]. n The value of is positively correlated with consumers’ recognition of tea brands;

[0089] S62. The average value of the consumer tea brand loyalty measurement index in the consumer tea brand loyalty measurement index set P is calculated as Mean value of consumer tea brand loyalty purchase decision time measurement index Mean value of consumer tea brand loyalty and purchase satisfaction measurement indicators Mean value of consumer tea brand loyalty and purchase recommendation intention measurement index and the mean value of consumer tea brand loyalty and purchase recommendation intention measurement index The consumer tea brand loyalty data h in the consumer tea brand loyalty data set H n Compare the parameters of loyalty measurement indicators and search for the mean value of consumer tea brand loyalty and repurchase frequency measurement indicators Mean value of consumer tea brand loyalty purchase decision time measurement index Mean value of consumer tea brand loyalty and purchase satisfaction measurement indicators Mean value of consumer tea brand loyalty and purchase recommendation intention measurement index and the mean value of consumer tea brand loyalty and purchase recommendation intention measurement index Corresponding consumer tea brand loyalty data h n , and generate the set time consumer tea brand loyalty analysis data h after data identification fenxi , execute and generate the set time consumer tea brand loyalty analysis data h fenxi The specific steps are as follows:

[0090] S621, initialization, update the maximum number of iterations of the algorithm;

[0091] S622, grazing behavior. The foal grazes in the herd. To perform grazing behavior, the stallion is set as the center of the grazing area. The brand loyalty parameter matches the individual horses in the herd and searches around the center. That is, with the stallion as the search center, the search space H of the consumer tea brand loyalty data set is used to search for the mean of the consumer tea brand loyalty repurchase frequency measurement index. Mean value of consumer tea brand loyalty purchase decision time measurement index Mean value of consumer tea brand loyalty and purchase satisfaction measurement indicators Mean value of consumer tea brand loyalty and purchase recommendation intention measurement index and the mean value of consumer tea brand loyalty and purchase recommendation intention measurement index Matched consumer tea brand loyalty data h n , grazing behavior simulation, simulates the behavior of matching brand loyalty parameters with individual horses moving at different radii and searching for the leading wild horse. The grazing behavior simulation formula is as follows:

[0092] in Stallion represents the simulated position of the j-th brand loyalty parameter matching horse group individual in the i-dimensional space, that is, the simulated position of the j-th brand loyalty parameter matching horse group individual in the consumer tea brand loyalty data set H search space with spatial dimension ι. j represents the position of the leading wild horse in the herd in the search space of the consumer tea brand loyalty data set H, represents the current position of the j-th brand loyalty parameter matching horse group individual in the i-dimensional space, that is, the j-th brand loyalty parameter matching horse group individual in the consumer tea brand loyalty data set H search space with spatial dimension ι, Π represents the adaptive coefficient, γ represents a random number in the interval [-2, 2], in represents a vector consisting of 0 and 1, are all random vectors uniformly distributed in the interval [0,1], γ2 is a random value in the interval [0,1], and ζ is a random vector that satisfies the condition , ψ is an adaptive parameter that decreases from 1 to 0 as the number of iterations increases. Where t represents the number of iterations, and T represents the maximum number of iterations;

[0093] S623, Wild Horse performs mating behavior in the search space H of the consumer tea brand loyalty data set. The mating behavior simulation formula is as follows: in, Represents a herd of horses The brand loyalty parameter matches the individual position of the horse group individual σ in the search space of the consumer tea brand loyalty data set H after leaving the group and re-entering the horse group. Crossover represents the position random function. Represents a herd of horses After the individual σ of the horse group leaves the brand loyalty parameter matching group, it re-enters the individual position of the horse group τ in the search space of the consumer tea brand loyalty data set H. Represents a herd of horses The brand loyalty parameter matches the individual position of the horse group υ after leaving the group and re-entering the horse group τ in the search space of the consumer tea brand loyalty data set H;

[0094] S624, return to the team leader, the leading wild horse leads the members to a better habitat, that is, search for the mean value of the repurchase frequency measurement index of consumer tea brand loyalty in the consumer tea brand loyalty data set H search space Mean value of consumer tea brand loyalty purchase decision time measurement index Mean value of consumer tea brand loyalty and purchase satisfaction measurement indicators Mean value of consumer tea brand loyalty and purchase recommendation intention measurement index and the mean value of consumer tea brand loyalty and purchase recommendation intention measurement index Matched consumer tea brand loyalty data h n ;

[0095] S625, selection of the leading wild horse, randomly selecting the leading wild horse to maintain the randomness of the algorithm;

[0096] S626. When the algorithm meets the maximum number of iterations, the output is the average value of the consumer tea brand loyalty repurchase frequency measurement index Mean value of consumer tea brand loyalty purchase decision time measurement index Mean value of consumer tea brand loyalty and purchase satisfaction measurement indicators Mean value of consumer tea brand loyalty and purchase recommendation intention measurement index and the mean value of consumer tea brand loyalty and purchase recommendation intention measurement index Matched consumer tea brand loyalty data h n ; Otherwise, continue executing S622 to S624 until the maximum number of iterations is met.

[0097] S627, the consumer tea brand loyalty data h output in step S626 n After data identification, the set time consumer tea brand loyalty analysis data h is generated fenxi , where h fenxi The value of is a random number in [0,1].

[0098] Through the cooperation between the consumer tea product repurchase frequency measurement unit and the consumer tea product purchase complaint rate measurement unit, the repurchase frequency information and purchase complaint rate parameters of consumer tea products within the set time are scientifically counted based on numerical measurement, thereby realizing independent and efficient statistics of consumer tea brand loyalty measurement index parameters; the consumer tea brand loyalty measurement index mean measurement unit, based on the consumer tea product repurchase frequency parameters, purchase decision time length parameters, purchase satisfaction parameters, purchase recommendation intention value and purchase complaint rate parameters, intelligently and accurately counts the mean of consumer tea brand loyalty measurement index parameters, thereby realizing scientific and accurate statistics of consumer tea brand loyalty measurement indicators, and improving the accuracy of consumer tea brand loyalty evaluation; the set time consumer tea brand loyalty analysis unit, based on the consumer tea brand loyalty measurement index mean combined with intelligent recognition algorithm and scientifically preset consumer tea brand loyalty parameters, conducts numerical intelligent analysis of consumer tea brand loyalty within the set time, thereby realizing comprehensive and accurate evaluation of consumer tea brand loyalty results based on multiple measurement index parameters, thereby improving the scientific nature of tea brand loyalty evaluation results, and enhancing the efficiency and accuracy of tea brands in responding to consumer demand feedback.

[0099] For further information, see Figure 1 - Figure 2 The steps for identifying the sales strategy of a tea brand based on the tea brand loyalty analysis data of consumers at a set time and the tea brand sales strategy data, generating the tea brand sales strategy identification data, and executing the tea brand sales strategy update management task are as follows:

[0100] S71. Establish tea brand sales strategy data set G = (g1,…,g k ,…,g λ ), k=1,2,3,…,λ; where g k represents the tea brand sales strategy data corresponding to the k-th tea brand loyalty value type, λ represents the maximum number of tea brand loyalty value types, and the tea brand sales strategy data represents the optimal tea product sales plan information set based on the consumer's tea brand loyalty parameter standard;

[0101] S72, set the time consumer tea brand loyalty analysis data h fenxi and tea brand sales strategy data g in tea brand sales strategy data set G k Compare the tea brand loyalty values and identify the tea brand loyalty analysis data of consumers at a set time h fenxi Corresponding tea brand sales strategy data g k , and generate tea brand sales strategy identification data through data identification

[0102] S73. The shopping management platform identifies tea brand sales strategy data The information is pushed to the tea brand's offline sales store management terminal through the mobile communication network, and the tea brand sales strategy update management task is executed.

[0103] Through the cooperation between the tea brand sales strategy storage unit and the target tea brand sales strategy identification unit, the tea brand sales strategy is intelligently and scientifically analyzed according to the tea brand loyalty analysis parameters of consumers at the set time combined with numerical matching and tea brand sales strategy information stored based on big data, so as to scientifically formulate tea brand sales strategies based on consumer tea brand loyalty parameters and enhance the intelligence and flexibility of tea brand management; the tea brand sales strategy update management operation execution unit independently and efficiently executes the tea brand sales strategy update management operation according to the tea brand sales strategy identification parameters combined with the shopping management platform, so as to realize the accurate and dynamic update of tea brand market sales strategies in line with market demand and enhance the quality and competitiveness of tea brand management.

[0104] Example 2:

[0105] See also Figure 1 - Figure 2 , a tea brand loyalty evaluation system based on consumer interaction data, used to implement a tea brand loyalty evaluation method based on consumer interaction data, the system includes a tea brand loyalty measurement parameter collection management module, a tea brand loyalty parameter measurement management module, and a tea brand sales strategy management module;

[0106] The tea brand loyalty measurement parameter collection and management module includes a tea brand loyalty evaluation setting time range collection unit, a consumer history shopping information storage unit, a set time consumer history shopping information search unit, a consumer tea product repurchase frequency search unit, a consumer tea product purchase decision time search unit, a consumer tea product purchase satisfaction information search unit, a consumer tea product purchase recommendation willingness value information search unit, and a consumer tea product purchase complaint frequency search unit;

[0107] The tea brand loyalty evaluation setting time range collection unit collects the tea brand loyalty evaluation setting time interval data through the shopping management platform; the consumer history shopping information storage unit is used to store the consumer history shopping data; the set time consumer history shopping information search unit searches and processes the consumer history shopping information within the set time based on the tea brand loyalty evaluation setting time interval data and the consumer history shopping data, and generates the set time consumer history shopping data; the consumer tea product repurchase frequency search unit searches and processes the shopping behavior information and shopping feedback information of tea brand consumers based on the set time consumer history shopping data, and generates the consumer tea product repurchase frequency data; the consumer tea product purchase decision time search unit searches and processes the tea brand consumer history shopping data based on the set time consumer history shopping data The shopping behavior information and shopping feedback information of consumers are searched and processed to generate the length of time for consumers to make tea product purchase decisions; the consumer tea product purchase satisfaction information search unit searches and processes the shopping behavior information and shopping feedback information of tea brand consumers according to the consumer historical shopping data of set time to generate the consumer tea product purchase satisfaction data; the consumer tea product purchase recommendation willingness value information search unit searches and processes the shopping behavior information and shopping feedback information of tea brand consumers according to the consumer historical shopping data of set time to generate the consumer tea product purchase recommendation willingness value data; the consumer tea product purchase complaint number search unit searches and processes the shopping behavior information and shopping feedback information of tea brand consumers according to the consumer historical shopping data of set time to generate the consumer tea product purchase complaint number data;

[0108] The tea brand loyalty parameter measurement management module includes a consumer tea product repurchase frequency measurement unit, a consumer tea product purchase complaint rate measurement unit, a consumer tea brand loyalty measurement index mean measurement unit, a consumer tea brand loyalty storage unit, and a set time consumer tea brand loyalty analysis unit;

[0109] A consumer tea product repurchase frequency measurement unit performs numerical measurement processing on the consumer tea product repurchase frequency based on the tea brand loyalty evaluation set time interval data and the consumer tea product repurchase frequency data, and generates consumer tea product repurchase frequency data; a consumer tea product purchase complaint rate measurement unit performs numerical measurement processing on the consumer tea product purchase complaint rate based on the tea brand loyalty evaluation set time interval data and the consumer tea product purchase complaint frequency data, and generates consumer tea product purchase complaint rate data; a consumer tea brand loyalty measurement index mean measurement unit performs numerical measurement processing on the consumer tea brand loyalty measurement index parameter mean based on the consumer tea product repurchase frequency parameter, purchase decision time length parameter, purchase satisfaction parameter, purchase recommendation intention value and purchase complaint rate parameter, and generates the consumer tea brand loyalty measurement index mean; a consumer tea brand loyalty storage unit is used to store consumer tea brand loyalty data; a set time consumer tea brand loyalty analysis unit performs numerical analysis processing on the consumer tea brand loyalty within a set time based on the consumer tea brand loyalty measurement index mean and the consumer tea brand loyalty data, and generates the set time consumer tea brand loyalty analysis data;

[0110] The tea brand sales strategy management module includes a tea brand sales strategy storage unit, a target tea brand sales strategy identification unit, and a tea brand sales strategy update management operation execution unit;

[0111] The tea brand sales strategy storage unit is used to store tea brand sales strategy data; the target tea brand sales strategy identification unit identifies and processes the tea brand's sales strategy based on the tea brand loyalty analysis data of consumers at the set time and the tea brand sales strategy data, and generates tea brand sales strategy identification data; the tea brand sales strategy update management job execution unit executes the tea brand sales strategy update management job based on the tea brand sales strategy identification data and the shopping management platform.

[0112] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A tea brand loyalty evaluation method based on consumer interaction data, characterized in that: The method comprises the following steps: S1. Collecting tea brand loyalty evaluation data for a set time interval; S2. Searching and processing the consumer's historical shopping information within the set time period based on the tea brand loyalty evaluation set time interval data and the consumer's historical shopping data to generate the consumer's historical shopping data within the set time period; S3. Search and process the shopping behavior information and shopping feedback information of tea brand consumers based on the consumer shopping history data over a set period of time, and generate data on the number of tea product repurchases by consumers, the length of time it takes for consumers to make tea product purchase decisions, consumer tea product purchase satisfaction data, consumer tea product purchase recommendation willingness data, and the number of tea product purchase complaints by consumers; S4. Based on the tea brand loyalty evaluation set time interval data, the consumer tea product repurchase frequency data, and the consumer tea product purchase complaint frequency data, perform numerical measurement processing on the consumer tea product repurchase frequency and the consumer tea product purchase complaint rate, and generate consumer tea product repurchase frequency data and consumer tea product purchase complaint rate data; S5. Perform numerical measurement processing on the mean values of the measurement index parameters of the consumer tea brand loyalty based on the parameters of the consumer tea product repurchase frequency, purchase decision time, purchase satisfaction, purchase recommendation intention, and purchase complaint rate, to generate the mean value of the consumer tea brand loyalty measurement index; S6. Performing numerical analysis of consumer tea brand loyalty within a set time period based on the average value of the consumer tea brand loyalty measurement index and the consumer tea brand loyalty data to generate consumer tea brand loyalty analysis data within the set time period; S7. Perform tea brand sales strategy identification processing based on the tea brand loyalty analysis data of consumers at the set time and the tea brand sales strategy data, generate tea brand sales strategy identification data and execute tea brand sales strategy update management operations.

2. The tea brand loyalty evaluation method based on consumer interaction data according to claim 1, characterized in that: Said S1 comprises the following steps: S11. Collect the time interval parameters of the target tea brand loyalty evaluation task online through the information entry dialog box of the shopping management platform, and generate the tea brand loyalty evaluation setting time interval data O = [o1, o2], where o1 represents the tea brand loyalty evaluation setting start time point data, and o2 represents the tea brand loyalty evaluation setting end time point data, where the units of o1 and o2 are both composed of year, month, and day.

3. The tea brand loyalty evaluation method based on consumer interaction data according to claim 2, characterized in that: The S2 comprises the following steps: S21. Establish a consumer history shopping data set A = (a1,…,a m ,…,a η ), m=1,2,3,…,eta; where a m represents the historical shopping data of the mth consumer, and η represents the maximum number of consumers; S22, using the KD tree nearest neighbor search algorithm to compare o1 to o2 in O with a in A m According to the time and date value matching, search for all the consumer's historical shopping data a within the set time interval corresponding to o1 to o2 m , and construct a set of consumer shopping history data sets A'=(a' m1 ,…,a' m2 ), 1≤m1≤m≤m2≤η; where a' m1 represents the historical shopping data of the consumer at the set time corresponding to the m1th consumer within the set time, a' m2 Represents the historical shopping data of the consumer at the set time corresponding to the m2th consumer within the set time.

4. The tea brand loyalty evaluation method based on consumer interaction data according to claim 3, characterized in that: The S3 includes the following steps: S31, using the BERT language model algorithm to analyze the a' in the A' m1 to a' m2 The shopping behavior information and shopping feedback information of tea brand consumers within a set time are searched and processed according to the keywords of the number of tea product purchases, the purchase decision time, the purchase feedback satisfaction, the purchase feedback recommendation willingness value and the purchase feedback complaint number, and the consumer tea product repurchase frequency data set B = (b m1 ,…,b m2 ), where b m1 represents the number of times the m1th consumer repurchases tea products, b m2 b represents the number of times the consumer repurchases tea products corresponding to the m2th consumer; m1 and b m2 The unit is times; the consumer tea product purchase decision length data set C = (c m1 ,…,c m2 ), where c m1 represents the length of time for the tea product purchase decision of the m1th consumer, c m2 represents the length of time for the tea product purchase decision of the m2th consumer; c m1 and c m2 The unit is seconds; the consumer tea product purchase satisfaction data set D = (d m1 ,…,d m2 ), where d m1 represents the consumer tea product purchase satisfaction data corresponding to the m1th consumer, d m2 represents the consumer tea product purchase satisfaction data corresponding to the m2th consumer; d m1 and d m2 are all positive integers within [1,10]; the consumer tea product purchase recommendation willingness value data set E=(e m1 ,…,e m2 ), where e m1 represents the consumer tea product purchase recommendation value data corresponding to the m1th consumer, e m2 represents the consumer tea product purchase recommendation value data corresponding to the m2th consumer; e m1 and e m2 are all positive integers within [1,10]; the data set of consumers’ complaints about tea product purchases is F = (f m1 ,…,f m2 ), where f m1 represents the number of complaints about tea products purchased by the m1th consumer, f m2 represents the number of complaints about tea products purchased by the m2th consumer, f m1 and f m2 The unit is times.

5. The tea brand loyalty evaluation method based on consumer interaction data according to claim 4, characterized in that: The S4 comprises the following steps: S41, performing numerical measurement processing on the total time length of the tea brand loyalty evaluation set time interval from o1 to o2 in O, and generating tea brand loyalty evaluation set time length data R, where R=o2-o1+1, and the unit of R is day; S42, respectively, the b in the B m1 to b m2 The repurchase frequency of the consumer tea product is calculated by dividing the repurchase frequency of the consumer tea product by the set time length. m1 ,…,b' m2 ), where b' m1 represents the repurchase frequency data of tea products corresponding to the m1th consumer, b' m2 represents the repurchase frequency data of tea products corresponding to the m2th consumer, b' m1 and b' m2 The unit is times per day; The f in the F m1 to f m2 The number of complaints and the set time length are processed as a quotient with R, and the consumer tea product purchase complaint rate data set F'=(f' m1 ,…,f' m2 ), where f' m1 represents the repurchase frequency data of tea products corresponding to the m1th consumer, f' m2 represents the repurchase frequency data of tea products corresponding to the m2th consumer, f' m1 and f' m2 The unit is times per day.

6. The tea brand loyalty evaluation method based on consumer interaction data according to claim 5, characterized in that: The S5 comprises the following steps: S51, according to b' in B' m1 to b' m2 、c in C m1 to c m2 d in the D m1 to d m2 、The e in the E m1 to e m2 、the f' in the F' m1 to f' m2 The parameters of the repurchase frequency measurement index, purchase decision time measurement index, purchase satisfaction measurement index, purchase recommendation intention value measurement index and purchase complaint rate measurement index of consumer tea brand loyalty are measured and processed respectively, and the mean value set of consumer tea brand loyalty measurement index is generated. in It represents the mean value of the repurchase frequency measurement index of consumers’ tea brand loyalty. The unit is times per day; It represents the mean value of the measurement index of the length of time for consumers’ tea brand loyalty purchase decision. The unit is seconds; It represents the mean value of the consumer tea brand loyalty purchase satisfaction measurement index; It represents the mean value of the measurement index of consumers’ tea brand loyalty and purchase recommendation intention; It represents the mean value of the measurement index of consumers’ tea brand loyalty and purchase recommendation intention. The unit is times per day.

7. The tea brand loyalty evaluation method based on consumer interaction data according to claim 6, characterized in that: The S6 comprises the following steps: S61. Establish consumer tea brand loyalty data set H = (h1,…,h n ,…,h ι ), n=1,2,3,…,ι; where h n represents the consumer tea brand loyalty data corresponding to the nth loyalty measurement indicator combination information type, ι represents the maximum number of loyalty measurement indicator combination information types; S62, the P described described described and stated With the H in the h n Compare the loyalty measurement index parameters and search for the described described described and stated The corresponding consumer tea brand loyalty data h n , and generate the set time consumer tea brand loyalty analysis data h after data identification fenxi , execute to generate the set time consumer tea brand loyalty analysis data h fenxi The specific steps are as follows: S621, initialization, update the maximum number of iterations of the algorithm; S622, grazing behavior, the foal grazes in the herd. To perform grazing behavior, the stallion is set as the center of the grazing area, and the brand loyalty parameter matching horse individuals search around the center, that is, with the stallion as the search center, search for the horse that matches the brand loyalty parameter in the H search space. described described described and stated Match the h n , grazing behavior simulation, simulating the behavior of horse herd individuals moving in different radii and searching for the leading wild horse with matching brand loyalty parameters; S623, wild horses perform mating behavior in the H search space; S624, return to the team leader, the leading wild horse leads the members to a better habitat, that is, search for the habitat that matches the one in the H search space. described described described and stated Match the h n ; S625, selection of the leading wild horse, randomly selecting the leading wild horse to maintain the randomness of the algorithm; S626. When the algorithm meets the maximum number of iterations, the output is the same as the described described described and stated Match the h n ; Otherwise, continue executing S622 to S624 until the maximum number of iterations is met. S627, the h output in step S626 n Generate h after data identification fenxi , where h fenxi The value of is a random number in [0,1].

8. The tea brand loyalty evaluation method based on consumer interaction data according to claim 7, characterized in that: The S7 comprises the following steps: S71. Establish tea brand sales strategy data set G = (g1,…,g k ,…,g λ ), k=1,2,3,…,λ; where g k represents the tea brand sales strategy data corresponding to the k-th tea brand loyalty value type, and λ represents the maximum number of tea brand loyalty value types; S72, the h fenxi With the G in g k Compare the tea brand loyalty values and identify the h fenxi The corresponding g k , and generate tea brand sales strategy identification data through data identification S73, the shopping management platform will The information is pushed to the tea brand's offline sales store management terminal through the mobile communication network, and the tea brand sales strategy update management task is executed.

9. A tea brand loyalty evaluation system based on consumer interaction data, used to implement the tea brand loyalty evaluation method based on consumer interaction data according to any one of claims 1 to 8, characterized in that: The system includes a tea brand loyalty measurement parameter collection management module, a tea brand loyalty parameter measurement management module, and a tea brand sales strategy management module.

Images