CN114529359A - Shoe money type recommendation system and method based on user characteristics - Google Patents
Shoe money type recommendation system and method based on user characteristics Download PDFInfo
- Publication number
- CN114529359A CN114529359A CN202111614577.8A CN202111614577A CN114529359A CN 114529359 A CN114529359 A CN 114529359A CN 202111614577 A CN202111614577 A CN 202111614577A CN 114529359 A CN114529359 A CN 114529359A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- database
- module
- recommendation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012937 correction Methods 0.000 claims abstract description 77
- 238000007405 data analysis Methods 0.000 claims abstract description 45
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 230000004927 fusion Effects 0.000 claims description 46
- 239000000463 material Substances 0.000 claims description 28
- 230000002159 abnormal effect Effects 0.000 claims description 19
- 239000013598 vector Substances 0.000 claims description 17
- 241000233805 Phoenix Species 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000013075 data extraction Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 6
- 238000012544 monitoring process Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000011157 data evaluation Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Landscapes
- Business, Economics & Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a system and a method for recommending shoe money types based on user characteristics, wherein the system comprises a central database module, a data analysis module, a data correction module and a data recommendation module; the central database module is used for recording characteristic data of the user, and the characteristic data comprises historical order data of the user and browsing data of the user; the data analysis module is used for analyzing a second database in the central database module and transmitting a shoe money recommendation model obtained after analysis to the data recommendation module; the data correction module is used for analyzing and correcting a first database in the central database module and transmitting a corrected shoe money correction model to the data recommendation module; the data recommendation module is used for receiving the shoe money recommendation model sent by the data analysis module and the shoe money correction model received by the data correction module and recommending a user; the invention improves the accuracy and the diversity of the shoe money type recommendation system.
Description
Technical Field
The invention relates to the technical field of recommendation systems, in particular to a shoe money type recommendation system and method based on user characteristics.
Background
With the popularization of the internet and the rapid development of the e-commerce industry, compared with off-line shopping, people have higher and higher consumption frequency on the internet. The shoe as an indispensable article in daily life occupies a part of market of an electronic commerce platform. More and more people begin to buy the shoe money of own mood apparatus on line, but the full-looking commodity of online lin lan has surpassed commodity quantity, kind and style that show in the packing cupboard of daily off-line far away, can not accomplish to choose the shoe money type that accords with every user purchase demand best among numerous categories, and the not recommendation data that corresponds of different users for system's calculation is loaded down with trivial details, and when system's data appear not agreeing with user's historical data, can't accurately judge user's subjective wish, cause the inaccurate problem of recommendation shoe money type.
Disclosure of Invention
The invention aims to provide a shoe money type recommendation system and method based on user characteristics so as to solve the problems in the background technology.
In order to solve the technical problems, the invention provides the following technical scheme: a shoe money type recommendation system based on user characteristics comprises a central database module, a data analysis module, a data correction module and a data recommendation module;
the central database module is used for recording characteristic data of the user, the characteristic data comprises historical order data of the user and browsing data of the user, and the central database module stores the historical order data of the user as a first database and stores the browsing data of the user as a second database; the first database comprises the order placing time of the user, the order placing amount of the user and the order placing style of the user, and the second database comprises search engine keywords of the user, inquiry customer service keywords and related data of browsing contents of the user;
the data analysis module is used for analyzing a second database in the central database module and transmitting a shoe money recommendation model obtained after analysis to the data recommendation module, and the shoe money recommendation model comprises a recommendation model I and a recommendation model II;
the data correction module is used for analyzing and correcting a first database in the central database module and transmitting a corrected shoe money correction model to the data recommendation module, and the shoe money correction model comprises a fusion correction model and a special value correction model;
and the data recommendation module is used for receiving the shoe money recommendation model sent by the data analysis module and the shoe money correction model received by the data correction module and recommending the user.
Further, the data analysis module comprises a data extraction module, a leading data analysis module and a phoenix tail data analysis module; the data extraction module extracts data information of the central database module and transmits the information to the leading data analysis module and the phoenix tail data analysis module;
the bibcock data analysis module is used for analyzing search engine keywords input by a user when the user has a browsing intention and inquiry customer service keywords input by the user when the user has a purchasing intention, comprehensively analyzing the search engine keywords and the inquiry customer service keywords, and transmitting a recommendation model obtained by analysis to the data recommendation module;
the phoenix tail data analysis module is used for analyzing the browsing content related data of the user after the user searches the keywords, and transmitting a recommendation model II obtained by processing and analyzing the browsing content related data of the user to the data recommendation module.
Further, the data correction module comprises a fusion correction module and a characteristic value correction module;
the fusion correction module is used for judging whether the data information of different users in the first database of the central data module meets the fusion condition, if so, the fusion correction module fuses the recommendation models in the data analysis modules of the different users, records the fused models as fusion correction models and transmits the fusion correction models to the data recommendation module.
The characteristic value correction module is used for judging whether the data information of the central database module of the user has special conditions or not, processing and analyzing the characteristic value under the special conditions to obtain a characteristic value correction model, and judging whether the data analysis module needs to be corrected or not.
A shoe money type recommendation method based on user characteristics comprises the following specific steps:
step S1: acquiring historical order data of a user and browsing data of the user; storing historical order data of a user into a first database, wherein the first database comprises the order placing time of the user, the order placing amount of the user and the order placing style of the user; storing browsing data of a user into a second database, wherein the second database comprises search engine keywords, inquiry customer service keywords and browsing content related data of the user;
the data of the user are respectively stored into two different databases, which is beneficial to a subsequent system to extract and classify different data of the user; the first database is order information actually determined by the user and used as a reference for data evaluation, and the second database is real-time dynamic information of the user and used for judging real-time preference trend of the user;
step S2: extracting data information in the second database in the step S1, and performing keyword similarity analysis on the data information to obtain an analyzed recommendation model I;
analyzing a recommendation model, namely, in order to determine the primary shopping intention of a user, whether the keyword is searched in a search engine or a keyword for inquiring customer service indicates the purchase intention of the user and belongs to a leading trend;
step S3: extracting data information in the second database in the step S1, and analyzing data related to browsing content of the data to obtain an analyzed recommendation model II;
analyzing a recommendation model II to judge the preference of the displayed content after searching in order to determine the subsequent browsing trend of the user;
step S4: extracting the information of the first database and the second database in the step S1, judging the fusion correction of the information of the first database, and performing the fusion operation on the second database which meets the conditions to obtain a fusion correction model;
step S5: extracting information of the first database and the second database in the step S1, analyzing and processing the abnormal characteristic value appearing in the second database by combining the information in the first database, and judging whether the data analysis module needs to be corrected to obtain a characteristic value correction model;
step S6: recommending the model to the user according to rules based on the data analysis results of the steps S2-S5.
Further, the specific process of step S2 is:
step S20: setting a set of keywords in a search engine to { R }i}，i＝{1,2......n}，RiA set of representations RiThe ith keyword in the page; set key words for inquiring customer service as set Gi}，i＝{1,2......n}，GiRepresentation set GiThe ith key word in the page; keywords include, but are not limited to: the material, the size, the color and the heel height of the shoe style;
step S21: recording the keyword R in the search engine in the single complete search process based on the analysis process of step S20iHas a frequency of tiT is any natural number, tiIndicates the ith kind of keyword RiThe frequency of occurrence; recording keywords G for querying customer service in single complete search processiHas a frequency of piP is an arbitrary natural number, piIndicates the ith kind of keyword GiThe frequency of occurrence; and will be { RiAnd { G }iAll words appearing in the symbol are represented by corresponding frequencies in corresponding sets;
converting the frequency of occurrence of the keywords into a number of times indicating that the search directions representing the users are similar if the keywords are similar;
step S22: the frequency t in step S21iForming a frequency vector
The frequency p in step S21iForming a frequency vector
Using the formula:
determining the similarity cos (alpha) between the keywords of the search engine and the keywords when inquiring about the customer service, wherein
Representing a vector
The die of (a) is used,
representing a vector
The die of (2); and when the obtained similarity value is larger than a keyword similarity threshold value preset by the system, recording that the keyword at the moment forms a first recommendation model, wherein the first recommendation model is the corresponding shoe money of the keyword set under the condition.
The method comprises the steps of (1) imagining a keyword of a search engine and a keyword when inquiring customer service into two line segments in a space, starting from an origin point ([0, 0. ]) and pointing to different directions; an included angle is formed between the two line segments, if the included angle is 0 degree, the direction is the same, the line segments are overlapped, and the fact that the texts represented by the two vectors are completely equal is shown; if the included angle is 90 degrees, the right angle is formed, and the directions are completely dissimilar; if the angle is 180 degrees, it means the direction is exactly opposite. Therefore, the similarity degree of the vectors can be judged according to the size of the included angle; the smaller the included angle is, the more similar the included angle is; the closer the similarity cos (α) is to 1, the higher the similarity between the keyword indicating the search engine and the keyword at the time of inquiring about the customer service, whereas the closer the similarity cos (α) is to 0, the lower the similarity between the keyword indicating the search engine and the keyword at the time of inquiring about the customer service.
Further, the specific process of step S3 is:
step S30: dividing the data related to the browsed content of the user, including color related data, material related data, style related data and others, wherein the color related data includes the total number of pictures of the amplified browsed picture color and the browsed content, and the number of times of amplifying the browsed picture color is recorded as o1The total number of pictures of the browsed content is o2(ii) a The material related data comprises the time spent on the material introduction and the total time from entering the content introduction to exiting the content introduction browsing, and the time spent on the material introduction is p1And the total time from entering the content introduction to exiting the content introduction browsing is p2(ii) a The style related data comprises the number of strokes of the finger on the matching style detail page and the total number of strokes from entering the detail page to pushing out the detail page, and the number of strokes of the finger on the matching style detail page is recorded as q1And the total stroke number from entering the detail page to pushing out the detail page is q2；
Step S31: based on the data in step S30, using the formula
Calculating the color preference O of the user by using a formula
Calculating the user material preference P by using a formula
Calculating the style preference of the user; calculating the integrated average preference of the user as
Calculating the comprehensive average preference degree of the user, and comparing the comprehensive average preference degree with different data of the user after weighted assignment to analyze the real preference of the user under the influence of weighting;
step S32: setting a weight value a for the user ' S color preference, a weight value b for the user ' S material preference, and a weight for the user ' S style preference based on the data in step S31The value c, and { a, b, c } is less than or equal to 1; sorting the three kinds of preference degrees from large to small, and in the process of setting the corresponding weight values, making the order of the weight values opposite to the order of the preference degrees, and utilizing a formula
Obtaining the comprehensive weighted average preference of the user;
the purpose of calculating the weighted average is to increase interference factors, and a minimum weight value is added to the user data with the highest user preference degree to interfere the user preference degree, so that the real preference of the user under the interference condition is judged;
step S33: analyzing the final preference value of the user using the formula γ — N-W based on the data of step S31 and step S32; and if gamma is greater than 0, forming a second recommendation model by using the user browsing content data corresponding to the minimum weight value, and if gamma is less than or equal to 0, combining color related data, material related data and style related data contained in the user browsing content data to form the second recommendation model.
And comparing the difference between the comprehensive weighted average preference of the user and the comprehensive average preference, and when the difference is greater than 0, indicating that the comprehensive level is still improved under the condition that the influence of the weight value is light, further indicating that the influence of the preference of the user data corresponding to the minimum weight value is the highest, so that the data form a recommendation model and are pushed to the user.
Further, the specific process of step S4 is:
step S40: the sum of the orders made by the users in the first database of the users x and y is recorded as ux、uy(ii) a Giving a user ordering amount threshold value D of the system, and when the user ordering amount ux、uyWhen D is greater than or equal to D, the attribute characteristic value is recorded as 1, and when the user places an order, the amount ux、uy<D, recording the attribute characteristic value of 0 at the moment, and respectively recording the times of 1 of the attribute characteristic values of the user x and the user y as j11The number of times that the attribute eigenvalue of the user x is 1 and the attribute eigenvalue of the user y is 0 is j10J belongs to any natural number; recording attribute eigenvalues of both user x and user y as 0Number of times r00The number of times that the attribute eigenvalue of user x is 0 and the attribute eigenvalue of user y is 1 is r01；
Step S41: using the formula:
finding the similarity between user x and user y in the order amount, wherein J (u)x,uy) Denotes a similarity coefficient, d (u)x,uy) Representing a similarity distance; and d (u)x,uy) E (0, 1); the intra-system similarity threshold is set when d (u)x,uy) If the similarity is smaller than the built-in similarity threshold value of the system, the similarity is recorded as that the user x and the user y have the similarity, and fusion can be performed; if d (u) is not satisfiedx,uy) E (0,1) or d (u)x,uy) If the similarity is larger than or equal to the similarity threshold value in the system, the fusion is not carried out;
the difference between the order-placing amounts of two users is represented by the formula, when the system analyzes the order-placing amounts corresponding to different users in the same order-placing sequence, the two users are associated to form an asymmetric binary attribute, the distance between the two users is calculated to represent the similarity, the similarity coefficient is the ratio of the number of the sample intersection sets to the number of the sample union sets, the similarity coefficient with opposite similarity coefficients is the similarity distance, the similarity between the two samples is measured by the proportion of different elements in the two sets to the elements, and the similarity distance d (u) is measuredx,uy) A larger size indicates a lower similarity between user x and user y in the amount of the order.
Step S42: and on the basis of the fusion in the step S41, performing mutual fusion recommendation on the recommendation model i and the recommendation model ii in the second database of the user x and the user y to form a fusion correction model, and performing recommendation in combination with the first database at the ordering time corresponding to each user.
Further, the specific process of step S5 is:
step S50: recording the order form of the user in the first database, wherein the order form of the user comprises the shoe money number and the shoe money type, respectively drawing the shoe money number and the shoe money type in a two-dimensional coordinate system according to the purchase sequence, comparing the shoe money number and the shoe money type in the second database, and depicting the shoe money number and the shoe money type in the corresponding two-dimensional coordinate system; the abscissa of the two-dimensional coordinate system is a shoe money purchasing sequence, the ordinate is a shoe money number, the shoe money number is distributed in sequence from small to large at equal intervals by taking the origin of coordinates as a starting point, the other ordinate is a shoe money type, the shoe money types comprise spring and summer child shoes, autumn and winter child shoes, spring and summer women's shoes, autumn and winter women's shoes, spring and summer men's shoes and autumn and winter men's shoes, and the shoe money types are sequenced at equal intervals in sequence from the origin of coordinates;
step S51: when the data corresponding to the second database in the two-dimensional coordinate system has an extreme value, the extreme value comprises a maximum value and a minimum value, and the data corresponding to the first database after the extreme value also has the same extreme value, the extreme value is judged to be an abnormal characteristic value, at the moment, the recommended model is corrected, and the corrected characteristic value correction model is pushed to a user; and continuing to push the corresponding recommendation model I and recommendation model II after the abnormal characteristic value disappears.
The monitoring of the abnormal characteristic values is set to improve the accuracy and diversification of shoe purchasing of the user, because the phenomenon that the user purchases shoes for children or parents sometimes occurs when the user purchases the shoes, under the monitoring of the second database, whether the abnormal browsing contents are mistaken touches or real purchasing intentions needs to be judged, so that the actual orders of the first database can be accurately judged by checking, the abnormal phenomena are corrected, and related characteristic value correction models are pushed to the user.
Compared with the prior art, the invention has the following beneficial effects: according to the invention, analysis is carried out according to the historical orders and browsing data of the users, a recommendation model which meets the user satisfaction degree and can form consumption is quickly and effectively pushed, and similar shoe money selection preferences among different users can be fused, so that the selectivity among the users is increased, the pressure of system data storage is reduced, besides, the invention can also accurately correct abnormal data of the users, carry out diversified judgment on the selection of the users, and carry out diversified pushing on shoe money types.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic block diagram of a shoe type recommendation system based on user characteristics according to the present invention;
FIG. 2 is an overall step diagram of a method for recommending shoe money types based on user characteristics according to the present invention;
FIG. 3 is a diagram of a first step of a recommendation model of a shoe money type recommendation method based on user characteristics according to the present invention;
FIG. 4 is a diagram of the steps of forming a second recommendation model of the shoe money type recommendation method based on the user characteristics according to the invention;
FIG. 5 is a diagram of the steps of forming a fusion correction model of a method for recommending shoe money types based on user characteristics according to the present invention;
FIG. 6 is a diagram of steps for forming a characteristic value correction model of a shoe money type recommendation method based on user characteristics according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1 to 6, the present invention provides a technical solution: a shoe money type recommendation system based on user characteristics comprises a central database module, a data analysis module, a data correction module and a data recommendation module;
the central database module is used for recording characteristic data of the user, the characteristic data comprises historical order data of the user and browsing data of the user, and the central database module stores the historical order data of the user as a first database and stores the browsing data of the user as a second database; the first database comprises the order placing time of the user, the order placing amount of the user and the order placing style of the user, and the second database comprises search engine keywords of the user, inquiry customer service keywords and related data of browsing contents of the user;
the data analysis module is used for analyzing a second database in the central database module and transmitting a shoe money recommendation model obtained after analysis to the data recommendation module, and the shoe money recommendation model comprises a recommendation model I and a recommendation model II;
the data correction module is used for analyzing and correcting a first database in the central database module and transmitting a corrected shoe money correction model to the data recommendation module, and the shoe money correction model comprises a fusion correction model and a special value correction model;
and the data recommendation module is used for receiving the shoe money recommendation model sent by the data analysis module and the shoe money correction model received by the data correction module and recommending the user.
The data analysis module comprises a data extraction module, a leading data analysis module and a phoenix tail data analysis module; the data extraction module extracts data information of the central database module and transmits the information to the leading data analysis module and the phoenix tail data analysis module;
the bibcock data analysis module is used for analyzing search engine keywords input by a user when the user has a browsing intention and inquiry customer service keywords input by the user when the user has a purchasing intention, comprehensively analyzing the search engine keywords and the inquiry customer service keywords, and transmitting a recommendation model obtained by analysis to the data recommendation module;
the phoenix tail data analysis module is used for analyzing the browsing content related data of the user after the user searches the keywords, and transmitting a recommendation model II obtained by processing and analyzing the browsing content related data of the user to the data recommendation module.
The data correction module comprises a fusion correction module and a characteristic value correction module;
the fusion correction module is used for judging whether the data information of different users in the first database of the central data module meets the fusion condition, if so, the fusion correction module fuses the recommendation models in the data analysis modules of different users, and records the fused models as fusion correction models and transmits the fusion correction models to the data recommendation module.
The characteristic value correction module is used for judging whether the data information of the central database module of the user has special conditions or not, processing and analyzing the characteristic value under the special conditions to obtain a characteristic value correction model, and judging whether the data analysis module needs to be corrected or not.
A shoe money type recommendation method based on user characteristics comprises the following specific steps:
step S1: acquiring historical order data of a user and browsing data of the user; storing historical order data of a user into a first database, wherein the first database comprises the order placing time of the user, the order placing amount of the user and the order placing style of the user; storing browsing data of a user into a second database, wherein the second database comprises search engine keywords, inquiry customer service keywords and browsing content related data of the user;
the data of the user are respectively stored into two different databases, which is beneficial to a subsequent system to extract and classify different data of the user; the first database is order information actually determined by the user and used as a reference for data evaluation, and the second database is real-time dynamic information of the user and used for judging the real-time preference trend of the user;
step S2: extracting data information in the second database in the step S1, and performing keyword similarity analysis on the data information to obtain an analyzed recommendation model I;
analyzing a recommendation model, namely, in order to determine the primary shopping intention of a user, whether the keyword is searched in a search engine or a keyword for inquiring customer service indicates the purchase intention of the user and belongs to a leading trend;
the specific process of step S2 is:
step S20: setting a set of keywords in a search engine to { R }i}，i＝{1,2......n}，RiRepresentation set RiThe ith keyword in the page; set key words for inquiring customer service as set Gi}，i＝{1,2......n}，GiRepresentation set GiThe ith key word in the page; keywords include, but are not limited to: the material, the size, the color and the heel height of the shoe style;
for example: keyword set in search engine R3The query is a set of keywords { cortex, black, low heel, code 37 }, and the query is for customer service { G }3{ cortex, brown, heel, code 37 };
step S21: recording the keyword R in the search engine in the single complete search process based on the analysis process of step S20iHas a frequency of tiT is any natural number, tiIndicates the ith kind of keyword RiThe frequency of occurrence; recording keywords G for inquiring customer service in single complete search processiHas a frequency of piP is an arbitrary natural number, piIndicates the ith kind of keyword GiThe frequency of occurrence; and will be { RiAnd { G }iAll words appearing in the symbol are represented by corresponding frequencies in corresponding sets;
converting the frequency of occurrence of the keywords into a number of times indicating that the search directions representing the users are similar if the keywords are similar;
step S22: the frequency t in step S21iForming a frequency vector
The frequency p in step S21iForming a frequency vector
Using the formula:
determining the similarity cos (alpha) between the keywords of the search engine and the keywords when inquiring about the customer service, wherein
Representing a vector
The die of (a) is used,
representing a vector
The mold of (4); and when the obtained similarity value is larger than a keyword similarity threshold value preset by the system, recording that the keyword at the moment forms a first recommendation model, wherein the first recommendation model is the corresponding shoe money of the keyword set under the condition.
For example: keyword set { R3The frequencies in { cortex 2, black 2, brown 0, low heel 1, middle heel 0, code 37 2} respectively, and the keyword set { G3The frequencies in the { cortex 1, black 0, brown 1, low heel 0, middle heel 2, code 37 1} are respectively;
if the similarity threshold of the keywords preset by the system is 0.6, 0.64 is greater than 0.6, the keywords form a first recommendation model, and the first recommendation model is the shoe money corresponding to the cortex of the keywords and the number 37; the cortex 2 represents the cortex, the keyword appears twice in the search engine, the analogy is repeated, all the keywords are recorded in two sets, and the corresponding frequency coefficient is extracted to obtain the corresponding frequency vector.
The method comprises the steps of (1) imagining a keyword of a search engine and a keyword when inquiring customer service into two line segments in a space, starting from an origin point ([0, 0. ]) and pointing to different directions; an included angle is formed between the two line segments, if the included angle is 0 degree, the direction is the same, the line segments are overlapped, and the fact that the texts represented by the two vectors are completely equal is shown; if the included angle is 90 degrees, the right angle is formed, and the directions are completely dissimilar; if the angle is 180 degrees, it means the direction is exactly opposite. Therefore, the similarity degree of the vectors can be judged according to the size of the included angle; the smaller the included angle is, the more similar the included angle is; the closer the similarity cos (α) is to 1, the higher the similarity between the keyword indicating the search engine and the keyword at the time of inquiring about the customer service, whereas the closer the similarity cos (α) is to 0, the lower the similarity between the keyword indicating the search engine and the keyword at the time of inquiring about the customer service.
Step S3: extracting data information in the second database in the step S1, and analyzing data related to browsing content of the data to obtain an analyzed recommendation model II;
analyzing a recommendation model II to judge the preference of the displayed content after searching in order to determine the subsequent browsing trend of the user;
the specific process of step S3 is:
step S30: dividing the data related to the browsed content of the user, including color related data, material related data, style related data and others, wherein the color related data includes the total number of pictures of the amplified browsed picture color and the browsed content, and the number of times of amplifying the browsed picture color is recorded as o1The total number of pictures of the browsed content is o2(ii) a The material related data comprises the time spent on the material introduction and the total time from entering the content introduction to exiting the content introduction browsing, and the time spent on the material introduction is p1And the total time from entering the content introduction to exiting the content introduction browsing is p2(ii) a The style related data comprises the number of strokes of the finger on the matching style detail page and the total number of strokes from entering the detail page to pushing out the detail page, and the number of strokes of the finger on the matching style detail page is recorded as q1And the total stroke number from entering the detail page to pushing out the detail page is q2；
Step S31: based on the data in step S30, using the formula
Calculating the color preference O of the user by using a formula
Calculating the user material preference P by using a formula
Calculating the style preference of the user; calculating the average preference of the users
For example: the total number of pictures browsed by the user is 6, the times of amplifying and browsing the colors of the pictures is 4, and at the moment
The total time from the user entering the content introduction to the user exiting the content introduction browsing is 20min, the time for staying in the material introduction is 3min, and at the moment
The total sliding times from the user entering the detail page to pushing out the detail page is 8, the sliding times of the finger on the matching style detail page is 2, and at the moment, the user enters the detail page and pushes out the detail page
The average preference of the user is integrated
Calculating the comprehensive average preference degree of the user, and comparing the comprehensive average preference degree with different data of the user after weighted assignment to analyze the real preference of the user under the influence of weighting;
step S32: based on the data in step S31, a weight value a and a user material preference are set for the user' S color preferenceSetting a weight value b for the goodness and a weight value c for the user style liking, wherein the weight values are more than or equal to 1 { a, b and c }; sorting the three kinds of preference degrees from large to small, and in the process of setting the corresponding weight values, making the order of the weight values opposite to the order of the preference degrees, and utilizing a formula
Obtaining the comprehensive weighted average preference of the user;
for example: analyzing the value of the user color preference, the user material preference and the user style preference to obtain
I.e. user colour preference>User style preference>The user material preference, the weight value sequence set at this time is: user material preference weighted value>User style preference weight value>The user color preference, namely setting a to 2, b to 8 and c to 5; then
The purpose of calculating the weighted average is to increase interference factors, and a minimum weight value is added to the user data with the highest user preference degree to interfere the user preference degree, so that the real preference of the user under the interference condition is judged;
step S33: analyzing the final preference value of the user using the formula γ ═ N-W based on the data of step S31 and step S32; and if gamma is greater than 0, forming a second recommendation model by using the user browsing content data corresponding to the minimum weight value, and if gamma is less than or equal to 0, combining color related data, material related data and style related data contained in the user browsing content data to form the second recommendation model.
For example: at this time
At the moment, the user data corresponding to the minimum weight value is the color preference of the user, and the shoe money type corresponding to the color preference of the user is recommended to the user by the recommendation model II;
and comparing the difference between the comprehensive weighted average preference of the user and the comprehensive average preference, and when the difference is greater than 0, indicating that the comprehensive level is still improved under the condition that the influence of the weight value is light, further indicating that the influence of the preference of the user data corresponding to the minimum weight value is the highest, so that the data form a recommendation model and are pushed to the user.
Step S4: extracting the information of the first database and the second database in the step S1, judging the fusion correction of the information of the first database, and performing the fusion operation on the second database which meets the conditions to obtain a fusion correction model;
the specific process of step S4 is:
step S40: the sum of the orders made by the users in the first database of the users x and y is recorded as ux、uy(ii) a Giving a user ordering amount threshold value D of the system, and when the user ordering amount ux、uyWhen D is greater than or equal to D, the attribute characteristic value is recorded as 1, and when the user places an order, the amount ux、uy<D, recording the attribute characteristic value of 0 at the moment, and respectively recording the times of 1 of the attribute characteristic values of the user x and the user y as j11The number of times that the attribute eigenvalue of the user x is 1 and the attribute eigenvalue of the user y is 0 is j10J belongs to any natural number; recording the times of the attribute characteristic values of the user x and the user y being 0 as r00The number of times that the attribute eigenvalue of user x is 0 and the attribute eigenvalue of user y is 1 is r01；
For example: ordering amounts of a first database of a user x are 320 yuan, 180 yuan, 230 yuan, 120 yuan and 540 yuan in sequence; ordering amounts of a first database corresponding to the user y are 230 yuan, 100 yuan, 120 yuan, 420 yuan and 380 yuan in sequence, and an ordering amount threshold value of the user of the given system is 200 yuan; the number of times that the attribute feature values of the user x and the user y are both 1 is 2, the number of times that the attribute feature value of the user x is 1 and the attribute feature value of the user y is 0 is 1, the number of times that the attribute feature values of the user x and the user y are both 0 is 1, and the number of times that the attribute feature value of the user x is 0 and the attribute feature value of the user y is 1;
step S41: using the formula:
finding the similarity between user x and user y in the order amount, wherein J (u)x,uy) Denotes a similarity coefficient, d (u)x,uy) Representing a similarity distance; and d (u)x,uy) E (0, 1); the intra-system similarity threshold is set when d (u)x,uy) If the similarity is smaller than the built-in similarity threshold value of the system, the similarity is recorded as that the user x and the user y have the similarity, and fusion can be performed; if d (u) is not satisfiedx,uy) E (0,1) or d (u)x,uy) If the similarity is larger than or equal to the similarity threshold value in the system, the fusion is not carried out;
for example: based on the result of step S41, then
At this time d (u)x,uy) E (0,1), and the threshold value of the similarity in the system is set as
Therefore d (u)x,uy) If the similarity is larger than or equal to the in-system similarity threshold, no fusion is performed.
The difference between the order-placing amounts of two users is represented by the formula, when the system analyzes the order-placing amounts corresponding to different users in the same order-placing sequence, the two users are associated to form an asymmetric binary attribute, the distance between the two users is calculated to represent the similarity, the similarity coefficient is the ratio of the number of sample intersections to the number of sample union sets, the similarity coefficient with opposite direction is the similarity distance, and the proportion of different elements in the two sets to the elements is used for measuring the similarity distanceSimilarity between two samples, and when the similarity distance d (u)x,uy) A larger size indicates a lower similarity between user x and user y in the amount of the order.
Step S42: and on the basis of the fusion in the step S41, performing mutual fusion recommendation on the recommendation model i and the recommendation model ii in the second database of the user x and the user y to form a fusion correction model, and performing recommendation in combination with the first database at the ordering time corresponding to each user.
Step S5: extracting information of the first database and the second database in the step S1, analyzing and processing the abnormal characteristic value appearing in the second database by combining the information in the first database, and judging whether the data analysis module needs to be corrected to obtain a characteristic value correction model;
the specific process of step S5 is:
step S50: recording the order form of the user in the first database, wherein the order form of the user comprises the shoe money number and the shoe money type, respectively drawing the shoe money number and the shoe money type in a two-dimensional coordinate system according to the purchase sequence, comparing the shoe money number and the shoe money type in the second database, and depicting the shoe money number and the shoe money type in the corresponding two-dimensional coordinate system; the abscissa of the two-dimensional coordinate system is a shoe money purchasing sequence, the ordinate is a shoe money number, the shoe money number is distributed in sequence from small to large at equal intervals by taking the origin of coordinates as a starting point, the other ordinate is a shoe money type, the shoe money types comprise spring and summer child shoes, autumn and winter child shoes, spring and summer women's shoes, autumn and winter women's shoes, spring and summer men's shoes and autumn and winter men's shoes, and the shoe money types are sequenced at equal intervals in sequence from the origin of coordinates;
step S51: when the data corresponding to the second database in the two-dimensional coordinate system has an extreme value, the extreme value comprises a maximum value and a minimum value, and the data corresponding to the first database after the extreme value also has the same extreme value, the extreme value is judged to be an abnormal characteristic value, at the moment, the recommended model is corrected, and the corrected characteristic value correction model is pushed to a user; and continuing to push the corresponding recommendation model I and recommendation model II after the abnormal characteristic value disappears.
For example: in a shoe money number two-dimensional coordinate system, data of a first database are stably displayed in the coordinate system to show that the numbers purchased by a user are all similar numbers, when the second database has a maximum value, the number is increased to show an abnormal characteristic value, and whether the first database has the same maximum value after the value is judged; if the abnormal characteristic value occurs, the user purchases a shoe money different from the usual purchasing habit, the recommended model can be corrected to obtain a characteristic value correction model, and the content in the characteristic value correction model is the browsing content of the corresponding user when the abnormal characteristic value occurs.
The monitoring of the abnormal characteristic values is set to improve the accuracy and diversification of shoe purchasing of the user, because the phenomenon that the user purchases shoes for children or parents sometimes occurs when the user purchases the shoes, under the monitoring of the second database, whether the abnormal browsing contents are mistaken touches or real purchasing intentions needs to be judged, so that the actual orders of the first database can be accurately judged by checking, the abnormal phenomena are corrected, and related characteristic value correction models are pushed to the user.
Step S6: recommending the model to the user according to rules based on the data analysis results of the steps S2-S5.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. A shoe money type recommendation system based on user characteristics is characterized by comprising a central database module, a data analysis module, a data correction module and a data recommendation module;
the central database module is used for recording characteristic data of the user, the characteristic data comprises historical order data of the user and browsing data of the user, and the central database module stores the historical order data of the user as a first database and stores the browsing data of the user as a second database; the first database comprises the order placing time of the user, the order placing amount of the user and the order placing style of the user, and the second database comprises search engine keywords of the user, inquiry customer service keywords and related data of browsing contents of the user;
the data analysis module is used for analyzing a second database in the central database module and transmitting a shoe money recommendation model obtained after analysis to the data recommendation module, and the shoe money recommendation model comprises a recommendation model I and a recommendation model II;
the data correction module is used for analyzing and correcting a first database in the central database module and transmitting a corrected shoe money correction model to the data recommendation module, wherein the shoe money correction model comprises a fusion correction model and a special value correction model;
the data recommendation module is used for receiving the shoe money recommendation model sent by the data analysis module and the shoe money correction model received by the data correction module and recommending the user.
2. The system and method for recommending shoe money types based on user characteristics as claimed in claim 1, wherein: the data analysis module comprises a data extraction module, a leading data analysis module and a phoenix tail data analysis module; the data extraction module extracts data information of the central database module and transmits the information to the leading data analysis module and the phoenix tail data analysis module;
the bibcock data analysis module is used for analyzing search engine keywords input by a user when the user wishes to browse and inquiry customer service keywords input when the user wishes to purchase, comprehensively analyzing the search engine keywords and the inquiry customer service keywords, and transmitting a recommendation model obtained by analysis to the data recommendation module;
the phoenix tail data analysis module is used for analyzing the browsing content related data of the user after the user searches the keywords, and transmitting a recommendation model II obtained by processing and analyzing the browsing content related data of the user to the data recommendation module.
3. The system and method for recommending shoe money types based on user characteristics as claimed in claim 1, wherein: the data correction module comprises a fusion correction module and a characteristic value correction module;
the fusion correction module is used for judging whether the data information of different users in the first database of the central data module meets the fusion condition, if so, the fusion correction module fuses the recommendation models in the data analysis modules of the different users, records the fused models as fusion correction models and transmits the fusion correction models to the data recommendation module.
The characteristic value correction module is used for judging whether the data information of the central database module of the user has special conditions or not, processing and analyzing the characteristic value under the special conditions to obtain a characteristic value correction model, and judging whether the data analysis module needs to be corrected or not.
4. A shoe money type recommendation method based on user characteristics is characterized by comprising the following steps: the method comprises the following specific steps:
step S1: acquiring historical order data of a user and browsing data of the user; storing historical order data of a user into a first database, wherein the first database comprises the order placing time of the user, the order placing amount of the user and the order placing style of the user; storing browsing data of a user into a second database, wherein the second database comprises search engine keywords, inquiry customer service keywords and browsing content related data of the user;
step S2: extracting data information in the second database in the step S1, and performing keyword similarity analysis on the data information to obtain an analyzed recommendation model I;
step S3: extracting data information in the second database in the step S1, and analyzing data related to browsing content of the data to obtain an analyzed recommendation model II;
step S4: extracting the information of the first database and the second database in the step S1, judging the fusion correction of the information of the first database, and performing the fusion operation on the second database which meets the conditions to obtain a fusion correction model;
step S5: extracting information of the first database and the second database in the step S1, analyzing and processing the abnormal characteristic value appearing in the second database by combining the information in the first database, and judging whether the data analysis module needs to be corrected to obtain a characteristic value correction model;
step S6: recommending the model to the user according to rules based on the data analysis results of the steps S2-S5.
5. The method for recommending shoe money types based on user characteristics as claimed in claim 4, wherein: the specific process of step S2 is as follows:
step S20: setting a set of keywords in a search engine to { R }i}，i＝{1,2......n}，RiA set of representations RiThe ith keyword in the page; set key words for inquiring customer service as set Gi}，i＝{1,2......n}，GiRepresentation set GiThe ith keyword in the page; the keywords include, but are not limited to: the material, the size, the color and the heel height of the shoe style;
step S21: recording the search index in the single complete search process based on the analysis process of step S20Optimus prime key word RiHas a frequency of tiT is any natural number, tiIndicates the ith kind of keyword RiThe frequency of occurrence; recording keywords G for inquiring customer service in single complete search processiHas a frequency of piP is an arbitrary natural number, piIndicates the ith kind of keyword GiThe frequency of occurrence;
step S22: the frequency t in step S21iForming a frequency vector
The frequency p in step S21iForming a frequency vector
Using the formula:
determining the similarity cos (alpha) between the keywords of the search engine and the keywords when inquiring about the customer service, wherein
Representing a vector
The die of (a) is used,
representing a vector
The mold of (4); and when the obtained similarity value is larger than a keyword similarity threshold value preset by the system, recording the keywords at the moment to form a first recommendation model, wherein the first recommendation model is the corresponding shoe money of the keyword set under the condition.
6. The method for recommending shoe money types based on user characteristics as claimed in claim 4, wherein: the specific process of step S3 is as follows:
step S30: dividing the data related to the browsed content of the user, wherein the data includes color related data, material related data, style related data and the like, the color related data includes the color of the amplified browsed picture and the total number of pictures of the browsed content, and the number of times of amplifying the color of the browsed picture is recorded as o1The total number of pictures of the browsing content is o2(ii) a The material related data comprises the time of staying at the material introduction and the total time from entering the content introduction to exiting the content introduction browsing, and the time of staying at the material introduction is recorded as p1And the total time from entering the content introduction to exiting the content introduction browsing is p2(ii) a The style related data comprises the number of strokes of the finger on the matching style detail page and the total number of strokes from entering the detail page to pushing out the detail page, and the number of strokes of the finger on the matching style detail page is recorded as q1And the total stroke number from entering the detail page to pushing out the detail page is q2；
Step S31: based on the data in step S30, using the formula
Calculating the color preference O of the user by using a formula
Calculating the user material preference P by using a formula
Calculating the style preference of the user; calculating the average preference of the users
Step S32: setting a weight value a for the user ' S color preference, a weight value b for the user ' S material preference, and a weight value c for the user ' S style preference based on the data in step S31, wherein 1 is equal to or greater than{ a, b, c }; sorting the three kinds of preference degrees from large to small, and in the process of setting the corresponding weight values, making the order of the weight values opposite to the order of the preference degrees, and utilizing a formula
Obtaining the comprehensive weighted average preference of the user;
step S33: analyzing the final preference value of the user using the formula γ ═ N-W based on the data of step S31 and step S32; and if gamma is greater than 0, forming a second recommendation model by using the user browsing content data corresponding to the minimum weight value, and if gamma is less than or equal to 0, combining color related data, material related data and style related data contained in the user browsing content data to form the second recommendation model.
7. The method for recommending shoe money types based on user characteristics as claimed in claim 4, wherein: the specific process of step S4 is as follows:
step S40: the sum of the orders made by the users in the first database of the users x and y is recorded as ux、uy(ii) a Giving a user ordering amount threshold value of D for the system, and when the user ordering amount ux、uyWhen D is greater than or equal to D, the attribute characteristic value is recorded as 1, and when the user places an order, the amount ux、uy<D, recording the attribute characteristic value of 0 at the moment, and respectively recording the times of 1 of the attribute characteristic values of the user x and the user y as j11The number of times that the attribute eigenvalue of the user x is 1 and the attribute eigenvalue of the user y is 0 is j10J belongs to any natural number; recording the times of the attribute characteristic values of the user x and the user y being 0 as r00The number of times that the attribute eigenvalue of user x is 0 and the attribute eigenvalue of user y is 1 is r01；
Step S41: using the formula:
finding the similarity between user x and user y in the order amount, wherein J (u)x,uy) Denotes a similarity coefficient, d (u)x,uy) Representing a similarity distance; and d (u)x,uy) E (0,1), in-system similarity threshold when d (u)x,uy) If the similarity is smaller than the built-in similarity threshold value of the system, the similarity is recorded as that the user x and the user y have the similarity, and fusion can be performed; if d (u) is not satisfiedx,uy) E (0,1) or d (u)x,uy) If the similarity is more than or equal to the in-system similarity threshold, the fusion is not carried out
Step S42: and performing mutual fusion recommendation on the first recommendation model and the second recommendation model in the second database of the user x and the user y to form a fusion correction model, and performing recommendation in combination with the first database at the ordering time corresponding to each user.
8. The method for recommending shoe money types based on user characteristics as claimed in claim 4, wherein: the specific process of step S5 is as follows:
step S50: recording the order form of the user in the first database, wherein the order form of the user comprises the shoe money number and the shoe money type, respectively drawing the shoe money number and the shoe money type in a two-dimensional coordinate system according to the purchase sequence, comparing the shoe money number and the shoe money type in the second database, and depicting the shoe money number and the shoe money type in the corresponding two-dimensional coordinate system; the abscissa of the two-dimensional coordinate system is a shoe money purchasing sequence, the ordinate is a shoe money number, the shoe money number is distributed in sequence from small to large at equal intervals by taking a coordinate origin as a starting point, the other ordinate is a shoe money type, the shoe money types comprise spring and summer child shoes, autumn and winter child shoes, spring and summer woman shoes, autumn and winter woman shoes, spring and summer man shoes and autumn and winter man shoes, and the shoe money types are sequenced at equal intervals in sequence from the coordinate origin;
step S51: when the data corresponding to the second database in the two-dimensional coordinate system has an extreme value, wherein the extreme value comprises a maximum value and a minimum value, and the data corresponding to the first database after the extreme value also has the same extreme value, the extreme value is judged to be an abnormal characteristic value, at the moment, the recommended model is corrected, and the corrected characteristic value correction model is pushed to a user; and continuing to push the corresponding recommendation model I and recommendation model II after the abnormal characteristic value disappears.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111614577.8A CN114529359B (en) | 2021-12-27 | 2021-12-27 | Shoe style recommendation system and method based on user characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111614577.8A CN114529359B (en) | 2021-12-27 | 2021-12-27 | Shoe style recommendation system and method based on user characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114529359A true CN114529359A (en) | 2022-05-24 |
| CN114529359B CN114529359B (en) | 2024-07-19 |
Family
ID=81618801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111614577.8A Active CN114529359B (en) | 2021-12-27 | 2021-12-27 | Shoe style recommendation system and method based on user characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114529359B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115983952A (en) * | 2023-03-21 | 2023-04-18 | 江西服装学院 | Custom garment design recommendation system |
| CN117688250A (en) * | 2024-02-04 | 2024-03-12 | 国网湖北省电力有限公司信息通信公司 | Unified data dynamic service management system and method suitable for electric power full scene |
| CN118134608A (en) * | 2024-05-06 | 2024-06-04 | 南京麦佳电子商务有限公司 | E-commerce user behavior analysis method and system based on data mining |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017193666A1 (en) * | 2016-05-10 | 2017-11-16 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for recommending personalized content |
| CN108960945A (en) * | 2017-05-18 | 2018-12-07 | 北京京东尚科信息技术有限公司 | Method of Commodity Recommendation and device |
| CN112541798A (en) * | 2020-09-27 | 2021-03-23 | 深圳莱尔托特科技有限公司 | Shared wardrobe service system |
| CN113592401A (en) * | 2021-07-30 | 2021-11-02 | 上海寻梦信息技术有限公司 | Address recommendation method, system, device and storage medium |
-
2021
- 2021-12-27 CN CN202111614577.8A patent/CN114529359B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017193666A1 (en) * | 2016-05-10 | 2017-11-16 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for recommending personalized content |
| CN108960945A (en) * | 2017-05-18 | 2018-12-07 | 北京京东尚科信息技术有限公司 | Method of Commodity Recommendation and device |
| CN112541798A (en) * | 2020-09-27 | 2021-03-23 | 深圳莱尔托特科技有限公司 | Shared wardrobe service system |
| CN113592401A (en) * | 2021-07-30 | 2021-11-02 | 上海寻梦信息技术有限公司 | Address recommendation method, system, device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 戴玉芳;李依璇;杜劲松;陈文;: "服装C2M定制模式中的关键技术", 服装学报, no. 05, 15 October 2018 (2018-10-15), pages 390 - 394 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115983952A (en) * | 2023-03-21 | 2023-04-18 | 江西服装学院 | Custom garment design recommendation system |
| CN115983952B (en) * | 2023-03-21 | 2023-12-08 | 江西服装学院 | Custom-made clothing design recommendation system |
| CN117688250A (en) * | 2024-02-04 | 2024-03-12 | 国网湖北省电力有限公司信息通信公司 | Unified data dynamic service management system and method suitable for electric power full scene |
| CN117688250B (en) * | 2024-02-04 | 2024-04-16 | 国网湖北省电力有限公司信息通信公司 | Unified data dynamic service management system and method suitable for electric power full scene |
| CN118134608A (en) * | 2024-05-06 | 2024-06-04 | 南京麦佳电子商务有限公司 | E-commerce user behavior analysis method and system based on data mining |
| CN118134608B (en) * | 2024-05-06 | 2024-07-09 | 南京麦佳电子商务有限公司 | E-commerce user behavior analysis method and system based on data mining |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114529359B (en) | 2024-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114529359B (en) | Shoe style recommendation system and method based on user characteristics | |
| US20220138831A1 (en) | Method of Providing Fashion Item Recommendation Service Using User's Body Type and Purchase History | |
| CN107122980B (en) | Method and device for identifying categories to which commodities belong | |
| CN112612973B (en) | Personalized intelligent clothing matching recommendation method combining knowledge graph | |
| CN103718212A (en) | Color determination device, color determination system, color determination method, information recording medium, and program | |
| KR102142126B1 (en) | Hierarchical Category Cluster Based Shopping Basket Associated Recommendation Method | |
| CN110619559B (en) | Method for accurately recommending commodities in electronic commerce based on big data information | |
| CN113034238B (en) | Commodity brand feature extraction and intelligent recommendation management method based on electronic commerce platform transaction | |
| KR20210031223A (en) | Method and System of recommending fashion based on vector based deep learning | |
| CN114581175A (en) | Commodity pushing method and device, storage medium and electronic equipment | |
| CN115392947A (en) | Demand prediction method and device | |
| CN107133811A (en) | The recognition methods of targeted customer a kind of and device | |
| CN115496555A (en) | Intelligent cross-border e-commerce website security quality assessment method and system | |
| KR20220039697A (en) | Method, apparatus and computer program for style recommendation | |
| KR20220019737A (en) | Method, apparatus and computer program for fashion item recommendation | |
| CN112508114A (en) | Intelligent clothing recommendation system and method | |
| CN115797014A (en) | Home decoration industry online and offline commodity recommendation method and device and storage medium | |
| KR102378072B1 (en) | Method, apparatus and computer program for style recommendation | |
| CN114021012A (en) | Information recommendation method and device, computer storage medium and electronic equipment | |
| CN110020136B (en) | Object recommendation method and related equipment | |
| KR20210063665A (en) | Recommendation item based on user event information and apparatus performing the same | |
| CN114648391B (en) | Online shopping information recommendation method | |
| CA2899529A1 (en) | Pricing intelligence for non-exact matching products | |
| KR102490621B1 (en) | Method and system for color-based flower product registration and retrieval service | |
| CN111316306A (en) | Intelligent terminal-based commodity recommendation method and commodity recommendation system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |