Fresh flower product personalized recommendation system based on big data
Through a personalized recommendation system for flower products based on big data, consumer data is collected and processed, and a personalized recommendation model is built, which solves the data sparseness and cold start problems in the social network environment, and improves the accuracy and user satisfaction of the recommendation system. .
Patent Information
- Application Number
- CN202411066636.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-05
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing personalized recommendation system for flower products faces data sparseness and cold start problems in the social network environment, making it difficult to provide accurate recommendations to new users, resulting in a decrease in recommendation quality and a decrease in user satisfaction.
A personalized recommendation system for flower products based on big data is designed, including data collection, preprocessing, processing, storage and recommendation generation modules. By collecting and preprocessing consumer data, the quality rate, browsing rate and good reviews of flower products are calculated. rate, build a purchase demand index, attention index and repurchase rate index, build a personalized recommendation model, and provide personalized recommendations.
It improves users' purchasing intention for recommended flower products, enhances the accuracy of the recommendation system and the satisfaction of new users, and solves the problems of data sparseness and cold start.
Smart Images

Figure CN119006100A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of big data analysis, and in particular to a personalized recommendation system for flower products based on big data. Background Art
[0002] The personalized recommendation system for flower products plays a key role in the current flower market. It helps users quickly find products of interest and reduce the trouble caused by information overload. This technology can not only target users' potential interests, but also dynamically respond to changes in users' interests, ensuring that the recommended content is consistent with users' real-time needs, while emphasizing the novelty of the recommendations and avoiding the interference of repeated information. In addition, the recommendation system also has the ability to mine long-tail information. It is not limited to recommending popular flower products, but should also dig deep into a large number of unpopular flower products to achieve flower business growth and diversified services.
[0003] However, the existing personalized recommendation system for flower products faces challenges in meeting the personalized service needs in the social network environment. Social networks exhibit scale-free characteristics, with only a very small number of users having a large number of social relationships and interactive behaviors, while most users have relatively limited social interactions. In such an environment, the personalized recommendation system for flower products faces the problem of data sparsity, and it is difficult to provide accurate recommendations for each user. In addition, when potential consumers join, due to the lack of relevant background or interest data, traditional recommendation systems find it difficult to provide satisfactory recommendations, which is the cold start problem in the social network environment. This problem leads to a decrease in the recommendation quality of the personalized recommendation system for flower products, and new users may feel disappointed, reducing their willingness to use the e-commerce platform again. Therefore, this paper provides a personalized recommendation system for flower products based on big data. Summary of the invention
[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a personalized recommendation system for flower products based on big data.
[0005] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a personalized recommendation system for flower products based on big data, comprising a monitoring center, wherein the monitoring center is communicatively connected with a data acquisition module, a data preprocessing module, a data processing module, a data storage module and a recommendation generation module;
[0006] The data collection module is used to collect the original data of consumers and set the collection cycle;
[0007] The data preprocessing module is used to preprocess the original data to obtain the quality rate, browsing rate and favorable comment rate of different types of flower products;
[0008] The data processing module is used to obtain the purchase demand index, attention index and repurchase rate index of different types of flower products according to the quality rate, browsing rate and praise rate of different types of flower products;
[0009] The data storage module is used to store the purchase demand index, attention index and repurchase rate index of different types of flower products;
[0010] The intelligent recommendation generation module is used to construct a personalized recommendation model for standard flower products according to the purchase demand index, attention index and repurchase rate index of different types of flower products, and to recommend flower products according to the personalized recommendation model for standard flower products.
[0011] Furthermore, the process of preprocessing the original data by the data preprocessing module includes:
[0012] The raw data includes the number of flower product purchases, the total number of flower product purchases, the number of flower product add-ons, the total number of flower product add-ons, the flower product browsing time, the total flower product browsing time, the number of flower product positive reviews, and the total number of flower product reviews;
[0013] Numbering the number of flower product purchases, the number of flower product add-ons, the flower product browsing time, and the number of flower product positive reviews;
[0014] According to the number of purchases of the numbered flower products, the number of added purchases of the flower products, the browsing time of the flower products and the number of favorable comments on the flower products, the number of purchases of different types of flower products, the number of added purchases of different types of flower products, the browsing time of different types of flower products and the number of favorable comments on different types of flower products are obtained.
[0015] Furthermore, the process of the data preprocessing module obtaining different types of flower product quality rates, flower product browsing rates, and flower product favorable comment rates includes:
[0016] According to the purchase frequency of different types of flower products, the total purchase frequency of flower products, the number of additional purchases of different types of flower products, and the total additional purchase frequency of flower products, the quality rate of different types of flower products is obtained;
[0017] According to the browsing time of different types of flower products and the total browsing time of flower products, the browsing rate of different types of flower products is obtained;
[0018] According to the number of favorable comments of different types of flower products and the total number of favorable comments of flower products, the favorable comment rate of different types of flower products is obtained.
[0019] Furthermore, the data processing module obtains the purchase demand index of different types of flower products according to the quality rates and browsing rates of different types of flower products.
[0020] Furthermore, the data processing module obtains attention indexes of different types of flower products according to browsing rates and favorable ratings of different types of flower products.
[0021] Furthermore, the data processing module obtains repurchase rate indexes of different types of flower products according to the quality rates of different types of flower products and the favorable ratings of flower products.
[0022] Furthermore, the process of storing the purchase demand index, attention index and repurchase rate index of different types of flower products in the data storage module includes:
[0023] The data storage module is provided with a plurality of current flower product classification storage databases and historical flower product classification storage databases;
[0024] Sending the purchase demand index, attention index and repurchase rate index of different types of flower products in the current collection period to the current flower product classification storage database for storage;
[0025] The purchase demand index, attention index and repurchase rate index of different types of flower products in the current collection cycle stored in several current flower product classification storage databases of the previous current collection cycle are sent to the historical flower product classification storage database for storage, and the purchase demand index, attention index and repurchase rate index sent to the historical flower product classification storage database are marked as the historical purchase demand index, the historical attention index and the historical repurchase rate index.
[0026] Furthermore, the process of the recommendation generation module constructing the corresponding standard flower product personalized recommendation model includes:
[0027] The historical purchase demand index, historical attention index and historical repurchase rate index of several groups of different types of flower products are used as training sets and test sets, and the training sets and test sets are input into a personalized recommendation model for standard flower products, and the personalized recommendation model for standard flower products is trained to obtain a personalized recommendation model for standard flower products after training.
[0028] Furthermore, the process of the recommendation generation module recommending flower products includes:
[0029] According to the purchase demand index, attention index, repurchase rate index of different types of flower products in the current collection period and the personalized recommendation model of standard flower products, the flower product recommendation matching value is obtained;
[0030] According to the purchase demand index, attention index, repurchase rate index and calculation formula of the recommended matching value of flower products of different types, the recommended matching value of the actual flower products of the corresponding types is obtained;
[0031] Preset recommended matching value thresholds for different types of flower products;
[0032] If the actual recommended matching value of the corresponding type of flower product is greater than or equal to the recommended matching value threshold of the corresponding type of flower product, then the flower product of that type is recommended;
[0033] If the recommended matching value of the corresponding type of flower product is less than the recommended matching value threshold, the flower product of this type will not be recommended.
[0034] Compared with the prior art, the beneficial effects of the present invention are as follows: by collecting raw data of consumers and preprocessing the raw data, the quality rate, browsing rate and favorable comment rate of different types of flower products are obtained; according to the quality rate, browsing rate and favorable comment rate of different types of flower products, the purchase demand index, attention index and repurchase rate index of different types of flower products are obtained; according to the purchase demand index, attention index and repurchase rate index of different types of flower products, a personalized recommendation model for flower products is constructed, and according to the personalized recommendation model for flower products, corresponding flower products are recommended to consumers, thereby improving consumers' purchase intention for the recommended corresponding flower products. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0036] Figure 1 This is a schematic diagram of a personalized recommendation system for flower products based on big data. DETAILED DESCRIPTION
[0037] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.
[0038] like Figure 1As shown, a personalized recommendation system for flower products based on big data includes a monitoring center, wherein the monitoring center is communicatively connected with a data acquisition module, a data preprocessing module, a data processing module, a data storage module and a recommendation generation module;
[0039] The data collection module is used to collect raw data of consumers;
[0040] The data preprocessing module is used to preprocess the original data to obtain the quality rate θ, the browsing rate μ and the praise rate δ of different types of flower products;
[0041] The data processing module is used to obtain the purchase demand index σ, attention index τ and repurchase rate index of different types of flower products according to the quality rate θ, browsing rate μ and praise rate δ of different types of flower products.
[0042] The data storage module is used to store the purchase demand index σ, attention index τ and repurchase rate index of different types of flower products. To store;
[0043] The intelligent recommendation generation module is used to generate recommendations based on the purchase demand index σ, attention index τ and repurchase rate index of different types of flower products. Build a personalized recommendation model for flower products, and recommend flower products based on the personalized recommendation model for flower products;
[0044] It should be further explained that, in the specific implementation process, the specific process of the data collection module collecting the original data of consumers includes:
[0045] The data collection module is provided with a data collection unit and a collection period is set to collect raw data of consumers in different collection periods;
[0046] The data acquisition module uploads the raw data collected in different acquisition cycles to the data preprocessing module;
[0047] It should be further explained that, in the specific implementation process, the specific process of the data preprocessing module preprocessing the raw data collected in different collection cycles includes:
[0048] It should be further explained that the raw data includes the number of flower product purchases, the total number of flower product purchases, the number of flower product add-ons, the total number of flower product add-ons, the browsing time of flower products, the total browsing time of flower products, the number of flower product positive reviews, and the total number of flower product reviews;
[0049] The purchase times of the flower products are numbered, recorded as i=1, 2, 3, ..., n, where n is a natural number and greater than 1, and the purchase times of different types of flower products A are obtained according to the numbered purchase times of the flower products. i ;
[0050] The number of times the flower products are purchased is numbered, recorded as w=1, 2, 3, ..., m, where m is a natural number and greater than 1. The number of times the flower products are purchased after the numbering is obtained. w ;
[0051] The browsing time of the flower products is numbered, recorded as x=1, 2, 3, ..., k, where k is a natural number and greater than 1, and the browsing time C of different types of flower products is obtained according to the numbered browsing time of the flower products. X ;
[0052] The number of favorable comments of the flower products is numbered, recorded as v=1, 2, 3, ..., o, where o is a natural number and greater than 1. The number of favorable comments of different types of flower products D is obtained according to the numbered favorable comments of the flower products. o ;
[0053] The total number of purchases of the flower product is marked as A; the total number of additional purchases of the flower product is marked as B; the total browsing time of the flower product is marked as C; the total number of evaluations of the flower product is marked as D;
[0054] It should be further explained that, in the specific implementation process, the specific process of the data preprocessing module obtaining the quality rate θ of different types of flower products includes:
[0055] According to the number of purchases of different types of flower products A i 、Number of times different types of flower products are purchased B w , the total number of flower product purchases A and the total number of flower product re-purchases B, the calculation formula for obtaining the quality rate θ of different types of flower products is:
[0056]
[0057] Among them, a1 is the weight coefficient of the ratio of the number of purchases of different types of flower products Ai to the total number of purchases of flower products A; a2 is the weight coefficient of the ratio of the number of additional purchases of different types of flower products Bw to the total number of additional purchases of flower products B;
[0058] It should be further explained that, in the specific implementation process, the specific process of the data preprocessing module obtaining the browsing rate μ of different types of flower products includes:
[0059] According to the browsing time of different types of flower products obtained, C XAs well as the total browsing time C of flower products, the calculation formula for obtaining the browsing rate μ of different types of flower products is:
[0060]
[0061] It should be further explained that, in the specific implementation process, the specific process of the data preprocessing module obtaining the favorable comment rate δ of different types of flower products includes:
[0062] According to the number of positive reviews of different types of flower products obtained D o And the total number of evaluations of the corresponding type of flower products D, the calculation formula for obtaining the praise rate δ of different types of flower products is:
[0063]
[0064] The data preprocessing module uploads the obtained quality rates θ, browsing rates μ and favorable ratings δ of different types of flower products to the data processing module.
[0065] It should be further explained that, in the specific implementation process, the specific process of the data processing module processing the uploaded different types of flower product quality rates θ, flower product browsing rates μ and flower product favorable comment rates δ includes:
[0066] It should be further explained that, in the specific implementation process, the specific process of the data processing module obtaining the purchase demand index σ of different types of flower products includes:
[0067] According to the quality rate θ of different types of flower products and the browsing rate μ of flower products, the calculation formula for obtaining the purchase demand index σ of different types of flower products is:
[0068] σ=H1×[ln(μ+1)] 2 +H2×μ+H3;
[0069] Among them, H1 and H2 are constant coefficients; H3 is the adjustment parameter of the purchase demand index σ;
[0070] It should be further explained that, in the specific implementation process, the specific process of the data processing module obtaining the attention index τ of different types of flower products includes:
[0071] According to the browsing rate μ and the favorable comment rate δ of different types of flower products, the calculation formula for obtaining the attention index τ of different types of flower products is:
[0072] τ=[K1×(μ) 2 +∫(μ) K2 dμ+K3×ln(δ+1)]+K4;
[0073] Among them, K1, K2 and K3 are constant coefficients; K4 is the adjustment parameter of the attention index τ;
[0074] It should be further explained that, in the specific implementation process, the data processing module obtains the repurchase rate index of different types of flower products. The specific process includes:
[0075] According to the quality rate θ of different types of flower products and the praise rate δ of flower products, the repurchase rate index of different types of flower products is obtained. The calculation formula is
[0076]
[0077] Among them, G1, G2 are index adjustment parameters, and G1>0, G2>0; D o is the number of positive reviews for different types of flower products, and D is the total number of reviews for the corresponding types of flower products; c a , c b is the self-compensation coefficient and c a >0,c b >0.
[0078] The data processing module will obtain the purchase demand index σ, attention index τ and repurchase rate index of different types of flower products. Upload to the data storage module;
[0079] It should be further explained that, in the specific implementation process, the data storage module has different types of flower product purchase demand index σ, attention index τ and repurchase rate index The specific process of storage includes:
[0080] The data storage module is provided with a plurality of current flower product classification storage databases and historical flower product classification storage databases;
[0081] It should be further explained that the current flower product classification storage database is used to store the purchase demand index σ, attention index τ and repurchase rate index of different types of flower products in the current collection period.
[0082] The database stores the purchase demand index σ, attention index τ and repurchase rate index of different types of flower products in the current collection cycle. The purchase demand index σ, attention index τ and repurchase rate index sent to the historical flower product classification storage database are stored. Marked as historical purchase demand index σ′, historical attention index τ′ and historical repurchase rate index
[0083] The data storage module stores the purchase demand index σ, attention index τ, and repurchase rate index of different types of flower products in the current collection period. Historical purchase demand index σ′, historical attention index τ′ and historical repurchase rate index Upload to the recommendation generation module;
[0084] It should be further explained that, in the specific implementation process, the recommendation generation module is used to generate the recommendation based on the historical purchase demand index σ′, historical attention index τ′ and historical repurchase rate index of different types of flower products. The specific process of building the corresponding standard flower product personalized recommendation model includes:
[0085] The historical purchase demand index σ′, historical attention index τ′ and historical repurchase rate index of several groups of different types of flower products are calculated. as a training set and a test set, and inputting the training set and the test set into a standard flower product personalized recommendation model to train the standard flower product personalized recommendation model to obtain a trained standard flower product personalized recommendation model;
[0086] According to the purchase demand index σ, attention index τ, and repurchase rate index of different types of flower products in the current collection period And the standard flower product personalized recommendation model, obtain the flower product recommendation matching value γ, where the calculation formula of the flower product recommendation matching value γ is:
[0087]
[0088] Among them, R, T and Y are weight coefficients;
[0089] According to the purchase demand index σ, attention index τ, and repurchase rate index of different types of flower products and the calculation formula of the flower product recommendation matching value γ, to obtain the actual flower product recommendation matching value γ′ of the corresponding type;
[0090] Preset recommended matching value thresholds for different types of flower products;
[0091] If the actual recommended matching value γ′ of the corresponding type of flower product is greater than or equal to the recommended matching value threshold of the corresponding type of flower product, then the flower product of this type is recommended;
[0092] If the recommended matching value γ′ of the corresponding type of flower product is less than the recommended matching value threshold, the flower product of this type will not be recommended.
[0093] By collecting and preprocessing the original data of consumers, we can obtain the quality rate θ, browsing rate μ and favorable rate δ of different types of flower products; according to the quality rate θ, browsing rate μ and favorable rate δ of different types of flower products, we can obtain the purchase demand index σ, attention index τ and repurchase rate index of different types of flower products. According to the purchase demand index, attention index and repurchase rate index of different types of flower products, a personalized recommendation model for flower products is constructed, and corresponding flower products are recommended to consumers based on the personalized recommendation model for flower products, thereby increasing consumers' purchase intention for the recommended corresponding flower products.
[0094] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A personalized recommendation system for flower products based on big data, including a monitoring center, characterized in that: The monitoring center is communicatively connected with a data acquisition module, a data preprocessing module, a data processing module, a data storage module and a recommendation generation module; The data collection module is used to collect the original data of consumers and set the collection cycle; The data preprocessing module is used to preprocess the original data to obtain the quality rate, browsing rate and favorable comment rate of different types of flower products; The data processing module is used to obtain the purchase demand index, attention index and repurchase rate index of different types of flower products according to the quality rate, browsing rate and praise rate of different types of flower products; The data storage module is used to store the purchase demand index, attention index and repurchase rate index of different types of flower products; The intelligent recommendation generation module is used to construct a personalized recommendation model for standard flower products according to the purchase demand index, attention index and repurchase rate index of different types of flower products, and to recommend flower products according to the personalized recommendation model for standard flower products.
2. According to the big data-based personalized flower product recommendation system of claim 1, it is characterized in that: The process of preprocessing the raw data by the data preprocessing module includes: The raw data includes the number of flower product purchases, the total number of flower product purchases, the number of flower product add-ons, the total number of flower product add-ons, the flower product browsing time, the total flower product browsing time, the number of flower product positive reviews, and the total number of flower product reviews; Numbering the number of flower product purchases, the number of flower product add-ons, the flower product browsing time, and the number of flower product positive reviews; According to the number of purchases of the numbered flower products, the number of added purchases of the flower products, the browsing time of the flower products and the number of favorable comments on the flower products, the number of purchases of different types of flower products, the number of added purchases of different types of flower products, the browsing time of different types of flower products and the number of favorable comments on different types of flower products are obtained.
3. The big data-based personalized flower product recommendation system according to claim 2, characterized in that: The process of obtaining different types of flower product quality rates, flower product browsing rates and flower product praise rates by the data preprocessing module includes: According to the purchase frequency of different types of flower products, the total purchase frequency of flower products, the number of additional purchases of different types of flower products, and the total additional purchase frequency of flower products, the quality rate of different types of flower products is obtained; According to the browsing time of different types of flower products and the total browsing time of flower products, the browsing rate of different types of flower products is obtained; According to the number of favorable comments of different types of flower products and the total number of favorable comments of flower products, the favorable comment rate of different types of flower products is obtained.
4. The big data-based personalized flower product recommendation system according to claim 3, characterized in that: The data processing module obtains the purchase demand index of different types of flower products according to the quality rates and browsing rates of different types of flower products.
5. The big data-based personalized recommendation system for flower products according to claim 4, characterized in that: The data processing module obtains the attention index of different types of flower products according to the browsing rates and favorable comment rates of different types of flower products.
6. The big data-based personalized flower product recommendation system according to claim 5, characterized in that: The data processing module obtains the repurchase rate index of different types of flower products according to the quality rates of different types of flower products and the praise rates of flower products.
7. The big data-based personalized recommendation system for flower products according to claim 6, characterized in that: The process of storing the purchase demand index, attention index and repurchase rate index of different types of flower products in the data storage module includes: The data storage module is provided with a plurality of current flower product classification storage databases and historical flower product classification storage databases; Sending the purchase demand index, attention index and repurchase rate index of different types of flower products in the current collection period to the current flower product classification storage database for storage; The purchase demand index, attention index and repurchase rate index of different types of flower products stored in several current flower product classification storage databases in the previous current collection cycle are sent to the historical flower product classification storage database for storage, and the purchase demand index, attention index and repurchase rate index sent to the historical flower product classification storage database are marked as the historical purchase demand index, the historical attention index and the historical repurchase rate index.
8. The big data-based personalized flower product recommendation system according to claim 7, characterized in that: The process of constructing a personalized recommendation model for standard flower products by the recommendation generation module includes: The historical purchase demand index, historical attention index and historical repurchase rate index of several groups of different types of flower products are used as training sets and test sets, and the training sets and test sets are input into a personalized recommendation model for standard flower products, and the personalized recommendation model for standard flower products is trained to obtain a personalized recommendation model for standard flower products after training.
9. The big data-based personalized flower product recommendation system according to claim 8, characterized in that: The process of the recommendation generation module recommending flower products includes: According to the purchase demand index, attention index, repurchase rate index of different types of flower products in the current collection period and the personalized recommendation model of standard flower products, the flower product recommendation matching value is obtained; According to the purchase demand index, attention index, repurchase rate index and calculation formula of the recommended matching value of flower products of different types, the recommended matching value of the actual flower products of the corresponding types is obtained; Preset recommended matching value thresholds for different types of flower products; If the actual recommended matching value of the corresponding type of flower product is greater than or equal to the recommended matching value threshold of the corresponding type of flower product, then the flower product of that type is recommended; If the recommended matching value of the corresponding type of flower product is less than the recommended matching value threshold, the flower product of this type will not be recommended.
Citation Information
Patent Citations
E-commerce platform based on cloud computing
CN112418993A
Personalized recommendation system based on big data analysis
CN117217865A
