Fresh flower matching and pushing system based on big data

Abstract

The invention discloses a fresh flower matching and pushing system based on big data, and relates to the technical field of big data analysis, the fresh flower matching and pushing system comprises a control center, and the control center is connected with a data acquisition module, a data preprocessing module, a data processing module, a matching analysis module and a comprehensive pushing module; acquiring fresh flower comprehensive data through a data acquisition module; performing data stream conversion and feature transformation on the fresh flower comprehensive data in a data preprocessing module to obtain a sample stream segment matrix; the data processing module is provided with a matching input end for keyword input, client keywords are obtained, data stream conversion and data tracing are carried out, and flower matching data are obtained; performing signal conversion and matching acquisition on the flower matching data in a matching analysis module to obtain a flower matching coefficient; the comprehensive recommendation module constructs a keyword library and a matching database and performs matching pushing to obtain matched fresh flower data and a pushed fresh flower set; the pushing precision and the working efficiency are improved, and the personalized service is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to a flower matching and pushing system based on big data. Background Art

[0002] Big data analysis refers to the process of collecting, storing, processing, analyzing and interpreting large-scale, high-dimensional and diverse data. By analyzing big data, we can find hidden patterns, associations, trends and insights, thereby helping to make more accurate decisions, discover new business opportunities, improve business processes, etc.

[0003] Flower matching push refers to the act of pushing relevant flower purchase or promotion information to users based on the results of flower matching by users. The purpose of flower matching push is to better meet the needs of users and guide users to make purchases through personalized recommendations and promotional strategies. However, there are some shortcomings in traditional flower matching push: the push time and frequency cannot be well controlled during push, and users may feel tired and disgusted with the push information, which may lead to user dissatisfaction with the platform and even choose to quit or stop using the product; the pushed information lacks personalization and targeting, or does not meet the user's preferences or needs, and users may feel boring or useless, which may also lead to user dissatisfaction; therefore, it is of great theoretical and practical significance to improve the deficiencies in the process of flower matching push.

[0004] How to use big data analysis technology to perform data stream conversion and feature transformation on the collected comprehensive flower data, obtain the sample flow segment matrix, process customer keywords, obtain flower matching data, perform signal conversion and matching collection on the flower matching data, obtain flower matching coefficients, build a keyword library and matching database and perform matching push, obtain matching flower data and push flower collections; this is the problem we need to solve; for this purpose, a flower matching push system based on big data is now provided. Summary of the invention

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A flower matching and pushing system based on big data comprises a control center, wherein the control center is connected with a data acquisition module, a data preprocessing module, a data processing module, a matching analysis module and a comprehensive pushing module;

[0007] The process of collecting comprehensive data of flowers by the data collection module includes:

[0008] Set up the login terminal and collection terminal;

[0009] Enter the user's identity information and login password through the login terminal, and the control center will review the user's identity information, and conduct personal login for the user who passes the review to obtain the application customer;

[0010] Through the collection end, information research on flowers is carried out to obtain comprehensive data of flowers.

[0011] Performing data stream conversion on the obtained flower comprehensive data to obtain a flower comprehensive data stream;

[0012] The obtained flower comprehensive data stream is node-splitting to obtain flower data stream segments, and the obtained flower data stream segments are split and marked to obtain monitoring code elements and identification code elements.

[0013] Performing front-end combination on the obtained flower data stream segments according to the obtained monitoring code elements and identification code elements to obtain a flower stream segment matrix;

[0014] Perform feature transformation on the obtained flower flow segment matrix to obtain a sample flow segment matrix;

[0015] Construct a sample matrix set, and upload the obtained sample flow segment matrix to the sample matrix set.

[0016] Set a matching input terminal, and use the matching input terminal for the application customer to input keywords to obtain the customer keywords;

[0017] Perform data stream conversion and node splitting on the obtained customer keywords to obtain keyword flow segments;

[0018] Pre-select the keyword segments according to the obtained sample matrix set to obtain pre-selected sample codes;

[0019] Performing status analysis on the obtained pre-selected sample code to obtain a pre-selected matching code and a pre-selected non-matching sample code;

[0020] Obtain the sample flow segment matrix corresponding to the pre-selected matching code, trace the data of the sample flow segment matrix, and obtain the flower matching data.

[0021] Performing signal conversion on the obtained flower matching data to obtain a flower matching signal;

[0022] Setting a matching decision coefficient according to the obtained flower matching signal;

[0023] Set the variation parameters to extract the matching decision coefficients in intervals to obtain the interval intervals, and make a variation response to the flower matching signal according to the variation parameters and the interval intervals to obtain the matching cardinality;

[0024] The matching decision coefficient is segmented according to the matching cardinality to obtain the matching decision segments.

[0025] Matching and combining the matching decision segments to obtain a combined decision segment;

[0026] Performing a combined response on the matching decision coefficients according to the obtained combined decision segments to obtain matching combination coefficients;

[0027] The flower matching signal is matched and collected according to the obtained matching combination coefficient to obtain the flower matching coefficient.

[0028] Perform signal conversion on the customer keywords of the application customers to obtain keyword signals;

[0029] Match and collect keyword signals according to the matching combination coefficients to obtain keyword coefficients, build a keyword library according to the keyword coefficients, perform spectrum conversion on the keyword coefficients, and obtain a keyword coefficient spectrum diagram;

[0030] Building a matching database based on flower matching coefficients, performing spectrum conversion on the flower matching coefficients, and obtaining a flower coefficient spectrum diagram;

[0031] Push and compare the keyword library according to the obtained matching database to obtain similar matching difference;

[0032] The similar matching differences are sorted to obtain a matching similarity sequence, and matching push is performed to application customers according to the obtained matching similarity sequence to obtain matching flower data and push flower collections.

[0033] Compared with the prior art, the invention has the following beneficial effects: data stream conversion and feature transformation are performed on the collected comprehensive flower data to obtain a sample flow segment matrix; a matching input terminal is set to input keywords, obtain customer keywords, and perform data stream conversion and data tracing to obtain flower matching data; by converting the collected flower data of different types and lengths into flower matching data of the same form, it is beneficial to increase matching efficiency and matching accuracy;

[0034] Perform signal conversion and matching collection on flower matching data to obtain flower matching coefficients; perform signal conversion and matching collection on customer keywords to obtain keyword coefficients, build keyword library and matching database, push and compare the keyword library according to the matching database to obtain similar matching difference, perform matching push to application customers according to similar matching difference, obtain matching flower data and push flower collections; through intelligent matching, users can get their favorite flowers, improve satisfaction with purchasing and receiving flowers, reduce manual intervention, and improve efficiency and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0037] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] like Figure 1 As shown, a flower matching and pushing system based on big data includes a control center, and the control center is connected with a data acquisition module, a data preprocessing module, a data processing module, a matching analysis module and a comprehensive pushing module;

[0039] The data collection module is used to collect comprehensive flower data of flowers, and the specific process includes:

[0040] Set up the login terminal and collection terminal;

[0041] Enter the user's identity information and login password through the login terminal, and the control center will review the user's identity information, grant login permissions to the users who pass the review, and generate a login account based on the login permissions obtained;

[0042] Enter the obtained login account and login password into the login terminal to log in personally, and mark the user who has completed the personal login as an application customer;

[0043] Conduct information research on flowers through the collection end to obtain comprehensive data of flowers;

[0044] It should be further explained that, in the specific implementation process, the information survey means investigating and recording the relevant information of flowers to obtain comprehensive data of flowers, and the comprehensive data of flowers includes flower attribute data, flower price data and flower source supply information;

[0045] Furthermore, the flower attribute data includes the flower name, type, variety, color, flower shape, stem length, flower size and flowering period; the flower price data includes the original price and discounted price of the flowers; and the flower source supply information includes the supplier name, credibility and geographical location.

[0046] The data preprocessing module is used to perform data stream conversion and feature transformation on the comprehensive data of flowers to obtain a sample flow segment matrix. The specific process includes:

[0047] Performing data stream conversion on the obtained flower comprehensive data to obtain a flower comprehensive data stream;

[0048] Furthermore, the data stream conversion means converting the flower comprehensive signal into a number of randomly combined binary code elements;

[0049] Perform node splitting on the obtained flower comprehensive data stream to obtain flower data stream segments;

[0050] Furthermore, the node splitting means splitting by 7 binary symbols as a group to obtain the flower data stream segment. If the last group is less than 7 binary symbols, the supplementary combination is performed forward until 7 binary symbols are satisfied as a group.

[0051] Split and mark the obtained flower data stream segment to obtain monitoring code elements and identification code elements;

[0052] Further, the split mark indicates marking the first 4 binary code elements of the 7 binary code elements of the flower data stream segment as monitoring code elements, and marking the last 3 binary code elements of the 7 binary code elements of the flower data stream segment as identification code elements;

[0053] Performing front-end combination on the obtained flower data stream segments according to the obtained monitoring code elements and identification code elements to obtain a flower stream segment matrix, and marking the obtained flower stream segment matrix as M;

[0054] Further, the front-end combination indicates that the flower data stream segments are grouped according to the order of node splitting, and three flower data stream segments are grouped to form a flower stream segment matrix. If the last group is less than three flower data stream segments, a supplementary combination is performed forward to obtain a flower stream segment matrix, and according to the obtained monitoring code elements, a matrix of three rows and four columns composed of monitoring code elements is recorded as a monitoring group, and according to the obtained identification code elements, a matrix of three rows and three columns composed of identification code elements is recorded as an identification group. Then, the flower stream segment matrix can be recorded as M=[J┆S], where J represents the monitoring group and S represents the identification group;

[0055] Perform feature transformation on the obtained flower flow segment matrix to obtain a sample flow segment matrix;

[0056] Further, the feature transformation represents performing modulo 2 addition on the flower flow segment matrix, and checking the identification group in the flower flow segment matrix, until the identification group modulo 2 addition becomes a unit matrix, then marking the transformed flower flow segment matrix as a sample flow segment matrix, and marking the obtained sample flow segment matrix as M1;

[0057] A sample matrix set is constructed, and the obtained sample flow segment matrix is ​​uploaded to the sample matrix set.

[0058] The data processing module is used to set a matching input end to input keywords, obtain customer keywords and perform data stream conversion and data tracing to obtain flower matching data. The specific process includes:

[0059] Set a matching input terminal, and use the matching input terminal for the application customer to input keywords to obtain the customer keywords;

[0060] Further, the keyword input indicates the keywords of flowers that the application customer needs to query, and the input keywords are included in the comprehensive data of flowers;

[0061] Performing data stream conversion on the obtained customer keywords to obtain keyword data stream;

[0062] In particular, the conversion rules for the data stream conversion of customer keywords and the data stream conversion of flower comprehensive data are the same;

[0063] Perform node splitting on the obtained keyword data stream to obtain keyword stream segments;

[0064] The obtained keyword flow segment is marked as G i , where i represents the number of the keyword flow segment, i=1, 2, 3, ..., v1, v1 is a positive integer;

[0065] Pre-select the keyword segments according to the obtained sample matrix set to obtain pre-selected sample codes;

[0066] The obtained pre-selected sample code is marked as Y, where represents the matrix transpose of M1;

[0067] Performing status analysis on the obtained pre-selected sample code to obtain a pre-selected matching code and a pre-selected non-matching sample code;

[0068] Furthermore, the pre-selected sample code includes a zero matrix and a non-zero matrix, and the sample stream segment matrix when the pre-selected sample code is a zero matrix is ​​marked as a pre-selected matching code, and the sample stream segment matrix when the pre-selected sample code is a non-zero matrix is ​​marked as a non-matching sample code;

[0069] Obtain the sample flow segment matrix corresponding to the pre-selected matching code, trace the data source of the sample flow segment matrix, and obtain the flower matching data;

[0070] It should be further explained that, in the specific implementation process, the data tracing means tracing the comprehensive flower data flow corresponding to the sample flow segment matrix corresponding to the pre-selected matching code, and obtaining the comprehensive flower data flow corresponding to the sample flow segment matrix, which is recorded as flower matching data; in particular, a keyword flow segment can obtain multiple flower matching data, so it is necessary to perform matching degree analysis on the multiple flower matching data to obtain the flower information that the application customer most wants to query, and make flower recommendations to the application customers based on the matching flower information.

[0071] The matching analysis module is used to perform signal conversion and matching collection on the flower matching data to obtain the flower matching coefficient. The specific process includes:

[0072] Performing signal conversion on the obtained flower matching data to obtain a flower matching signal;

[0073] Setting a matching decision coefficient according to the obtained flower matching signal;

[0074] The matching decision coefficient is expressed in a function form and is set according to the signal characteristics of the flower matching signal, wherein the signal characteristics include frequency range, center frequency, stopband interval and attenuation;

[0075] Setting a variation parameter, wherein the variation parameter is used to control the parameters of the scaling and translation variation of the flower matching signal in the time dimension and the frequency dimension;

[0076] Performing interval extraction on the matching decision coefficient according to the obtained variation parameter to obtain an interval interval, and marking the obtained interval interval as p;

[0077] The interval extraction means extracting the interval distance between two adjacent variable parameters in the matching decision coefficient to obtain the interval interval;

[0078] According to the obtained change parameters and intervals, the flower matching signal is changed to respond, and the matching base is obtained, and the obtained matching base is marked as B, where: c represents the variable parameter, β represents the weight factor, and f 中 represents the center frequency of the flower matching signal, ΔD 频 Indicates the frequency range of the flower matching signal;

[0079] The matching decision coefficient is segmented according to the obtained matching cardinality to obtain matching decision segments;

[0080] It should be further explained that, in the specific implementation process, the segment division means that the matching decision coefficients are equally divided according to the number of matching bases to obtain matching decision segments of equal length;

[0081] Matching and combining the obtained matching decision segments to obtain a combined decision segment;

[0082] The combined decision segment includes an upper decision segment, a middle decision segment and a lower decision segment;

[0083] Furthermore, the matching combination is to combine the obtained matching decision segments into groups of three, and to perform combination changes on the combined matching decision segments to obtain an upper decision segment, a middle decision segment, and a lower decision segment;

[0084] The obtained upper decision segment is marked as K 上 , where K 上 =θ1*Q e , θ1 represents the proportionality coefficient, Q e represents the matching decision segment, e represents the number of the matching decision segment, e=1, 2, 3, ..., v2, v2 is a positive integer;

[0085] The obtained middle decision segment is marked as K 中 , where K 中 =θ2*Q e , θ2 represents the proportionality coefficient;

[0086] The obtained lower decision segment is marked as K 下 , where K 下 =θ3*Q e , θ3 represents the proportionality coefficient, and θ1≠θ2≠θ3;

[0087] Performing a combined response on the matching decision coefficients according to the obtained combined decision segments to obtain matching combination coefficients;

[0088] The obtained matching combination coefficient is marked as Z, where W represents the matching decision coefficient;

[0089] Matching and collecting the flower matching signals according to the obtained matching combination coefficients to obtain the flower matching coefficients;

[0090] It should be further explained that, in the specific implementation process, the matching collection process includes:

[0091] The obtained matching combination coefficient is uploaded to the flower matching signal, and the matching combination coefficient is convolved with the corresponding position of the flower matching signal to obtain the segmented flower coefficient, the obtained segmented flower coefficient is logarithmically valued to obtain the segmented flower logarithm value, the obtained segmented flower logarithm value is summed to obtain the segmented flower coefficient, the obtained segmented flower coefficient is inversely discrete cosine transformed to obtain the flower matching coefficient.

[0092] The comprehensive recommendation module is used to build a keyword library and a matching database and perform matching push, obtain matching flower data and push flower collections. The specific process includes:

[0093] Perform signal conversion on the customer keywords of the application customers to obtain keyword signals;

[0094] Match and collect keyword signals according to the obtained matching combination coefficient to obtain the keyword coefficient;

[0095] Building a keyword library according to the obtained keyword coefficients, and uploading the obtained keyword coefficients to the keyword library;

[0096] Performing spectrum conversion on the obtained keyword coefficients to obtain a keyword coefficient spectrum graph;

[0097] Marking the obtained keyword coefficients with a curve to obtain a keyword coefficient curve;

[0098] Building a matching database according to the obtained flower matching coefficients, and uploading the obtained flower matching coefficients to the matching database;

[0099] Performing spectrum conversion on the obtained flower matching coefficients to obtain a flower coefficient spectrum diagram;

[0100] Marking the obtained flower coefficient spectrum graph with a curve to obtain a flower coefficient curve;

[0101] Push and compare the keyword library according to the obtained matching database to obtain similar matching difference;

[0102] It should be further explained that, in the specific implementation process, the push comparison process includes:

[0103] In the keyword library, any keyword coefficient is selected as the comparison keyword, and the keyword coefficient curve of the comparison keyword and the flower coefficient curve in the matching database are statistically analyzed in turn to obtain the similar matching difference, and the obtained similar matching difference is marked as ΔR, where ΔR = ∫(H0(t)-H n (t))dt, H0(t) represents the keyword coefficient curve of the comparison keyword, H n (t) represents the flower coefficient curve in the matching database, n represents the number of the flower coefficient curve in the matching database, and n=1, 2, 3, ..., v3, v3 is a positive integer, dt represents the differential of t, and t represents the intersection of the curve and the horizontal axis;

[0104] In particular, “the keyword coefficient curve of the comparison keyword and the flower coefficient curves in the matching database are sequentially subjected to difference statistics” means that the keyword coefficient curve of the comparison keyword and all flower coefficient curves in the matching database are subjected to difference statistics;

[0105] The obtained similar matching differences are sorted in descending order to obtain a matching similarity sequence;

[0106] According to the obtained matching similarity sequence, matching and pushing are performed to the application customers to obtain matching flower data and push flower collection;

[0107] It should be further explained that, in the specific implementation process, the flower comprehensive data of the flower coefficient curve corresponding to the first similar matching difference in the matching similar sequence is marked as the matching flower data, that is, the most matching flower data that the application customer wants to search for;

[0108] The comprehensive flower data of the flower coefficient curve corresponding to the similar matching difference of the remaining rankings are marked as a pushed flower set, and the obtained pushed flower set is pushed to the application customers to improve the recommendation accuracy of related searches.

[0109] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A flower matching and push system based on big data, including a control center, characterized in that: The control center is connected to a data acquisition module, a data preprocessing module, a data processing module, a matching analysis module and a comprehensive push module; The data acquisition module is used to collect comprehensive data of flowers; The data preprocessing module is used to perform data stream conversion and node splitting on the flower comprehensive data, obtain flower data stream segments, split and mark the flower data stream segments, obtain monitoring code elements and identification code elements, perform front-end combination and feature transformation on the obtained flower data stream segments, and obtain a sample stream segment matrix; The data processing module is used to set a matching input end to input keywords for application customers, obtain customer keywords, perform data stream conversion and match pre-selection on customer keywords, obtain pre-selected sample codes, perform status analysis and data tracing on the pre-selected sample codes, and obtain flower matching data; The matching analysis module is used to perform signal conversion on the flower matching data to obtain the flower matching signal, set the change parameter to perform interval extraction on the matching decision coefficient to obtain the interval interval, divide the matching decision coefficient into sections according to the change parameter and the interval interval to obtain the matching decision section, match and combine the matching decision sections to obtain the combined decision section, match and collect the flower matching signal according to the combined decision section to obtain the flower matching coefficient; The comprehensive recommendation module is used to perform signal conversion and matching collection on customer keywords, obtain keyword coefficients, build a keyword library and a matching database, push and compare the keyword library according to the matching database, obtain similar matching differences, match and push application customers according to the similar matching differences, obtain matching flower data and push flower collections.

2. A flower matching and pushing system based on big data according to claim 1, characterized in that: The process of collecting comprehensive data of flowers by the data collection module includes: Set up the login terminal and collection terminal; Enter the user's identity information and login password through the login terminal, and the control center will review the user's identity information, and conduct personal login for the user who passes the review to obtain the application customer; Through the collection end, information research on flowers is carried out to obtain comprehensive data of flowers.

3. A flower matching and pushing system based on big data according to claim 2, characterized in that: The process of obtaining monitoring code elements and identification code elements includes: Performing data stream conversion on the obtained flower comprehensive data to obtain a flower comprehensive data stream; The obtained flower comprehensive data stream is node-splitting to obtain flower data stream segments, and the obtained flower data stream segments are split and marked to obtain monitoring code elements and identification code elements.

4. A flower matching and pushing system based on big data according to claim 3, characterized in that: The process of obtaining the sample flow segment matrix includes: Performing front-end combination on the obtained flower data stream segments according to the obtained monitoring code elements and identification code elements to obtain a flower stream segment matrix; Perform feature transformation on the obtained flower flow segment matrix to obtain a sample flow segment matrix; Construct a sample matrix set, and upload the obtained sample flow segment matrix to the sample matrix set.

5. A flower matching and pushing system based on big data according to claim 4, characterized in that: The process of obtaining flower matching data includes: Set a matching input terminal, and use the matching input terminal for the application customer to input keywords to obtain the customer keywords; Perform data stream conversion and node splitting on the obtained customer keywords to obtain keyword flow segments; Pre-select the keyword segments according to the obtained sample matrix set to obtain pre-selected sample codes; Performing status analysis on the obtained pre-selected sample code to obtain a pre-selected matching code and a pre-selected non-matching sample code; Obtain the sample flow segment matrix corresponding to the pre-selected matching code, trace the data of the sample flow segment matrix, and obtain the flower matching data.

6. A flower matching and pushing system based on big data according to claim 5, characterized in that: The process of obtaining a matching decision segment includes: Performing signal conversion on the obtained flower matching data to obtain a flower matching signal; Setting a matching decision coefficient according to the obtained flower matching signal; Set the variation parameters to extract the matching decision coefficients in intervals to obtain the interval intervals, and make a variation response to the flower matching signal according to the variation parameters and the interval intervals to obtain the matching cardinality; The matching decision coefficient is segmented according to the matching cardinality to obtain the matching decision segments.

7. The flower matching and pushing system based on big data according to claim 6 is characterized in that: The process of obtaining the flower matching coefficient includes: Matching and combining the matching decision segments to obtain a combined decision segment; Performing a combined response on the matching decision coefficients according to the obtained combined decision segments to obtain matching combination coefficients; The flower matching signal is matched and collected according to the obtained matching combination coefficient to obtain the flower matching coefficient.

8. The flower matching and pushing system based on big data according to claim 7 is characterized in that: The process of obtaining matching flower data and pushing flower collections includes: Perform signal conversion on the customer keywords of the application customers to obtain keyword signals; Match and collect keyword signals according to the matching combination coefficients to obtain keyword coefficients, build a keyword library according to the keyword coefficients, perform spectrum conversion on the keyword coefficients, and obtain a keyword coefficient spectrum diagram; Building a matching database based on flower matching coefficients, performing spectrum conversion on the flower matching coefficients, and obtaining a flower coefficient spectrum diagram; Push and compare the keyword library according to the obtained matching database to obtain similar matching difference; The similar matching differences are sorted to obtain a matching similarity sequence, and matching push is performed to application customers according to the obtained matching similarity sequence to obtain matching flower data and push flower collections.

Images