CN115983952A - Custom garment design recommendation system - Google Patents
Custom garment design recommendation system Download PDFInfo
- Publication number
- CN115983952A CN115983952A CN202310279066.8A CN202310279066A CN115983952A CN 115983952 A CN115983952 A CN 115983952A CN 202310279066 A CN202310279066 A CN 202310279066A CN 115983952 A CN115983952 A CN 115983952A
- Authority
- CN
- China
- Prior art keywords
- user
- browsing
- data
- clothing
- historical
- 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
- 238000013461 design Methods 0.000 title claims abstract description 22
- 238000004458 analytical method Methods 0.000 claims abstract description 64
- 238000000034 method Methods 0.000 claims abstract description 10
- 238000005516 engineering process Methods 0.000 claims abstract description 6
- 238000012216 screening Methods 0.000 claims description 24
- 238000011156 evaluation Methods 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 9
- 238000007619 statistical method Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the technical field of customized garment design recommendation, in particular to a customized garment design recommendation system, which comprises: the data acquisition module is used for acquiring historical browsing data and historical order data of a user based on a big data technology; the type analysis module is used for carrying out user type analysis on the historical browsing data and the historical order data of the user; the style analysis module is used for carrying out user style analysis on the historical browsing data and the historical order data of the user; and the recommending module is used for recommending the clothing to the user. According to the method and the system, the type analysis and the style analysis are carried out on the user by acquiring the historical browsing data and the historical order data of the user to obtain the actual consumption analysis result of the user, and the recommendation of the related content of the customized clothing design is carried out on the user in a targeted manner according to the actual consumption analysis result, so that the user who actually needs the clothing customization service can better understand the clothing customization.
Description
Technical Field
The invention relates to the technical field of customized garment design recommendation, in particular to a customized garment design recommendation system.
Background
The clothes can be customized to enable a user to have clothes suitable for the wearing style of the user, and people who have difficulty in selecting clothes can find clothes suitable for the user faster through the clothes customization. Some businesses recommend clothing customization to users by means of broadcast network, and want more users to know the relevant knowledge of clothing customization, so as to guide some users who actually need the service to better enjoy the benefits brought by clothing customization.
However, in the current information age, the information received by people is too complicated, some information is easily missed, and the recommendation mode of the broad casting network does not consider the actual consumption condition of the user and does not push the content to the user in a targeted manner, so that part of users actually needing the clothing customization service still have little contact with the clothing customization service.
Disclosure of Invention
The application provides a customized garment design recommendation system which carries out customized garment design recommendation according to the actual consumption condition of a user in a targeted manner so as to solve the technical problem.
A custom garment design recommendation system is disclosed, the system comprising:
the data acquisition module is used for acquiring historical browsing data and historical order data of a user based on a big data technology; in the step, an information acquisition application is sent to a user, and after an agreement reply of the user is received, historical browsing data and historical order data of the user are acquired, wherein the historical browsing data is browsing data of the user on the clothes, and the historical order data is order data of the user for purchasing the clothes.
The type analysis module is used for carrying out user type analysis on the historical browsing data and the historical order data of the user to obtain a type analysis result, and adding a category label to the user according to the type analysis result; performing user type analysis includes: acquiring historical browsing data and historical order data of a user, determining the consumption level and the clothing size level of the user according to the historical order data, and determining the clothing selection tendency level of the user according to the historical browsing data and the historical order data.
The style analysis module is used for carrying out user style analysis on the historical browsing data and the historical order data of the user to obtain a style analysis result, and adding style labels to the user according to the style analysis result; the performing the user style analysis comprises:
the method comprises the steps of obtaining historical browsing data and historical order data of a user, dividing the historical browsing data into a plurality of subdata sets according to a time sequence, and creating a plurality of order sets according to the historical order data, wherein each order set comprises a historical order and a subdata set corresponding to the historical order.
For any order set, identifying a first label of a subdata set in the order set, sorting the first label according to the occurrence frequency, identifying a second label of a historical order in the order set, judging whether the sorting of the second label in the first label is greater than a preset ranking, and if so, taking the second label as a third label of the order set.
Performing statistical analysis on third labels of all order sets, determining a frequency weight parameter of each label according to occurrence frequency, determining a time weight parameter of each label according to seasonal attributes, determining a correction parameter of each user according to the type of the user, calculating a reference value of each label according to the frequency weight parameter, the time weight parameter and the correction parameter, and sequencing the third labels according to the reference values to obtain style analysis results;
the reference value for any tag is calculated as follows:
R e =((n 1 P)(n 2 +n 3 ))/(n 1 +n 2 +n 3 );
in the formula, R e Denotes a reference value, P denotes the frequency of occurrence of the tag, n 1 As a frequency weight parameter, n 2 Is a time weight parameter, n 3 To correct the parameters.
The style labels are a preset number of labels selected from the style analysis results according to the size sequence.
The recommendation module is used for carrying out label identification on the user to obtain a category label and a style label of the user, selecting a first target garment from a garment database according to the category label, selecting a second target garment from the first target garment according to the style label, and recommending the garment to the user based on the second target garment.
The performing the user type analysis includes:
acquiring the historical browsing data and the historical order data of the user, determining the consumption grade and the clothing size grade of the user according to the historical order data, and determining the clothing selection tendency grade of the user according to the historical browsing data and the historical order data; determining a clothing selection preference rating of the user according to the historical browsing data and the historical order data comprises: dividing the historical browsing data into a plurality of subdata sets according to a time sequence, and creating a plurality of order sets according to the historical order data, wherein each order set comprises a historical order and a subdata set corresponding to the historical order; for any order set, calculating the number of times of browsing the clothes corresponding to the order set according to the subdata set in the order set; and performing statistical analysis on the clothing browsing times of all the order sets, calculating to obtain an average value of the clothing browsing times, and determining the clothing selection tendency level of the user according to the average value of the clothing browsing times.
If the clothing size grade of the user does not belong to the preset clothing size range, marking the user as a first type; if the clothing size grade of the user belongs to the preset clothing size range, marking the user as a second type under the condition that the consumption grade of the user is smaller than a first preset grade and the clothing selection tendency grade is smaller than a second preset grade, otherwise, marking the user as a third type.
The category labels include a customer's consumption rating, a garment size rating, and a garment selection propensity rating, as well as the type of customer.
Selecting a first target garment from a garment database according to the category label comprises:
and determining the type of the user according to the category label, and selecting the clothing corresponding to the type of the user from the clothing database to obtain the preselected clothing.
For the users marked as the first type, screening the preselected clothes according to the clothes size grade to obtain the first target clothes; for the users marked as the second type, screening the preselected clothes according to the consumption grades to obtain the first target clothes; for users tagged as a third type, treat the pre-selected garment as the first target garment.
Selecting a second target garment from the first target garments according to the style label comprises:
and primarily screening the first target clothes according to the style labels, determining the weight of each label in the style labels, calculating the recommended value of the clothes subjected to primary screening based on the weight of each label in the style labels, secondarily screening the clothes subjected to primary screening according to a preset recommended threshold, and sequencing the rest clothes according to the recommended value to obtain the second target clothes.
The recommending clothing to the user based on the second target clothing comprises:
determining a recommendation cycle according to the clothing selection tendency grade, acquiring a recommendation sequence of the second target clothing, and recommending the clothing in the second target clothing to the user according to the recommendation cycle and the recommendation sequence.
Calculating the number of times of browsing the clothing corresponding to the order set according to the subdata set in the order set, and further comprising:
and screening the historical browsing data in the sub-data set according to the preset browsing time length to obtain target browsing data with the browsing time length larger than the preset browsing time length, and counting the clothing browsing times of the target browsing data.
The system further comprises:
the feedback module is used for acquiring browsing feedback data of the user and adjusting a recommendation sequence for recommending the clothes in the second target clothes to the user according to the browsing feedback data; the method comprises the following specific steps: the browsing feedback data is the actual browsing condition of the recommended clothing by the user, including the picture browsing duration, the clothing detail browsing duration and the evaluation browsing duration, and the recommended sequence adjustment parameter is calculated according to the following formula:
T=α 1 m 1 (c 1 D 1 +c 2 D 2 +c 3 D 3 );
wherein T is a recommended sequence adjustment parameter, m 1 For the total number of picture views, garment details views and evaluation data views, D 1 For picture browsing duration, D 2 For the duration of the browsing of the details of the garment, D 3 To evaluate the duration of the browsing, α 1 Is m 1 Weight parameter of c 1 、c 2 And c 3 Respectively, the picture browsing duration D 1 Garment detail browsing duration D 2 And evaluating the browsing duration D 3 The weight parameter of (2);
wherein, for m 1 M is the time length of any one of the picture browsing time length, the clothing detail browsing time length and the evaluation browsing time length of the user is greater than zero 1 Plus one;
after the recommendation sequence adjustment parameter T is obtained through calculation, the style labels of the recommended garments corresponding to the recommendation sequence adjustment parameter T are obtained, the weight of the style labels is adjusted through the recommendation sequence adjustment parameter T, and the recommendation values of the rest unrendered garments are corrected and updated to obtain a new recommendation sequence.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the method and the system, the type analysis and the style analysis are carried out on the user by acquiring the historical browsing data and the historical order data of the user to obtain the actual consumption analysis result of the user, and the recommendation of the related content of the customized clothing design is carried out on the user in a targeted manner according to the actual consumption analysis result, so that the user who actually needs the clothing customization service can better understand the clothing customization.
2. According to the method and the device, the browsing feedback data of the user are obtained, the browsing feedback data are analyzed and processed, the recommended sequence adjusting parameters are calculated, the recommended sequence of the clothes which are not recommended is adjusted through the recommended sequence adjusting parameters, and the recommending accuracy is improved.
Drawings
Fig. 1 is a schematic structural diagram of a recommendation system for custom garment design provided in an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, some embodiments of the present application will be described in detail below with reference to the accompanying drawings and embodiments.
Embodiment 1, referring to fig. 1, a custom garment design recommendation system, comprising:
the data acquisition module is used for acquiring historical browsing data and historical order data of a user based on a big data technology;
it is to be added that, in the process of obtaining the historical browsing data and the historical order data of the user based on the big data technology, an information obtaining application may be sent to the user first, and after receiving an agreement reply from the user, the historical browsing data and the historical order data of the user are obtained based on the big data technology, where the historical browsing data may be related browsing data of the user on the clothing, and the historical order data may be related order data of the user on the clothing.
The type analysis module is used for carrying out user type analysis on historical browsing data and historical order data of the user to obtain a type analysis result, and adding a category label to the user according to the type analysis result;
the style analysis module is used for carrying out user style analysis on historical browsing data and historical order data of the user to obtain a style analysis result, and adding style labels to the user according to the style analysis result;
and the recommendation module is used for carrying out label identification on the user to obtain a category label and a style label of the user, selecting a first target garment from the garment database according to the category label, selecting a second target garment from the first target garment according to the style label, and recommending the garments to the user based on the second target garment.
According to the customized clothing design recommendation system provided by the embodiment of the invention, the historical browsing data and the historical order data of the user are acquired through the data acquisition module, the type analysis module is used for carrying out the type analysis on the user, the style analysis module is used for carrying out the style analysis on the user, the personalized analysis on the user is completed, the category label and the style label are added for the user according to the analysis result, the related content of the customized clothing design is sent to the user according to the category label and the style label of the user, and the targeted content recommendation is realized, so that the user actually needing clothing customization service can better know the clothing customization.
A customized garment design recommendation system provided in the embodiments of the present application is described in detail below.
In an alternative embodiment, for the type analysis module, performing the user type analysis includes:
acquiring historical browsing data and historical order data of a user, determining the consumption level and the clothing size level of the user according to the historical order data, and determining the clothing selection tendency level of the user according to the historical browsing data and the historical order data.
It needs to be supplemented that, for the consumption grade, the total consumption amount of the user can be determined according to historical order data, the consumption grade is divided into 1 to 10 grades, and each grade corresponds to different total consumption amount ranges; for garment size grades, the garment size of the user can be determined according to historical order data, the garment size grades are divided into 1-10 grades, and each grade corresponds to different garment size ranges.
For the clothing selection tendency grade, the selection parameters of the user can be determined according to historical browsing data and historical order data, specifically, according to a time relation, the historical browsing data is divided into a plurality of parts based on the historical order data, each historical order corresponds to one part of the historical browsing data, clothing browsing times of the user under each historical order are calculated, the clothing browsing times of all the historical orders are subjected to statistical analysis, an average value of the clothing browsing times is calculated, the clothing selection tendency grade is divided into 1-10 grades, and each grade corresponds to different clothing browsing time ranges.
Dividing the user into different types according to the consumption grade, the clothing size grade and the clothing selection tendency grade of the user, and marking the user as a first type if the clothing size grade of the user does not belong to a preset clothing size range;
it should be added that the meaning of the preset size range of the clothes is to divide the people with the special height and weight, and the users with the special height and weight are marked as the first type.
If the clothing size grade of the user belongs to the preset clothing size range, in this case, for the condition that the consumption grade of the user is smaller than a first preset grade and the clothing selection tendency grade is smaller than a second preset grade, marking the user as a second type, otherwise, marking the user as a third type;
it should be added that the first preset level is to screen out users who consume less clothing, the second preset level is to screen out users who choose easier in clothing selection, for example, some users may choose the clothing to be purchased from the users after browsing several pieces of clothing, such users are users who choose easier in clothing selection, while some users choose the clothing to be purchased after browsing dozens of or even more pieces of clothing, such users are users who choose harder in clothing selection; and marking the users meeting the two conditions at the same time as the second type, wherein the rest users belong to the third type.
After the type analysis module is used for carrying out type analysis on the user, a type analysis result is obtained, the type analysis result specifically comprises the consumption grade, the clothing size grade, the clothing selection tendency grade and the type of the user, and the data are used as the category label of the user.
In an alternative embodiment, in the process of calculating the number of times of browsing the clothing of the user placed in each historical order, the method further comprises:
and screening the historical browsing data under each historical order according to the preset browsing duration, screening out the data of which the browsing duration is not more than the preset browsing duration to obtain target browsing data, and counting the clothing browsing times of the target browsing data.
In an alternative embodiment, for the style analysis module, performing the user style analysis comprises:
acquiring historical browsing data and historical order data of a user, dividing the historical browsing data into a plurality of subdata sets according to a time sequence, and creating a plurality of order sets according to the historical order data, wherein each order set comprises a historical order and a subdata set corresponding to the historical order;
it is to be added that the number of the subdata sets obtained by dividing the historical browsing data is the same as the number of the historical orders, each historical order corresponds to one subdata set, and the subdata sets represent the garments browsed by the user when purchasing the garments of a certain historical order.
For any order set, identifying a first label of a subdata set in the order set, sorting the first label according to the occurrence frequency, identifying a second label of a historical order in the order set, judging whether the sorting of the second label in the first label is greater than a preset ranking, and if so, taking the second label as a third label of the order set.
It should be added that, for a subdata set in any one order set, the total number of the tags in the first tag may be one or multiple, the number of times of occurrence of any one tag in the first tag may be one or multiple, and statistics is performed, wherein the frequency of occurrence of each tag in the first tag is sorted according to the order from large to small; for a historical order in any order set, a second label of the order may be determined, and it is conceivable that the number of the second label is one, and the second label is included in the first label, find out an ordering of the second label in the first label, and determine a third label of the order set according to the ordering of the second label in the first label, specifically, the ordering of the second label in the first label is greater than a preset ordering, which indicates that the second label has a reference value, and use it as the third label of the order set, otherwise, discard the order set, which is preset as a fourth label in this embodiment, and when the ordering of the second label in the first label belongs to the top three, indicate that the second label has the reference value.
Performing statistical analysis on third labels of all order sets, determining a frequency weight parameter of each label according to occurrence frequency, determining a time weight parameter of each label according to seasonal attributes, determining a correction parameter of each user according to the type of the user, calculating a reference value of each label according to the frequency weight parameter, the time weight parameter and the correction parameter, and sequencing the third labels according to the reference values to obtain style analysis results.
It needs to be supplemented that the third labels of each order set are counted, the occurrence frequency of each third label is calculated, and the frequency weight parameter of each third label is determined according to the occurrence frequency, and it is conceivable that the occurrence frequency is high and the frequency weight parameter is larger; the season attribute can be determined according to the current season and the transaction time of the historical order, and according to the time change sequence, the longer the season corresponding to the transaction time of the historical order is separated from the current season, the smaller the time weight parameter is; the correction parameters are related to the types of the users, the correction parameters are used for correcting the time weight parameters, and for the three types of users of the first type, the second type and the third type, the correction parameter size relationship of the users is that the first type is larger than the second type, and the third type is larger than the second type.
Specifically, the calculation formula of the reference value of any label is as follows:
R e =((n 1 P)(n 2 +n 3 ))/(n 1 +n 2 +n 3 );
in the formula, R e Denotes a reference value, P denotes the frequency of occurrence of the tag, n 1 As a frequency weight parameter, n 2 Is a time weight parameter, n 3 To correct the parameters.
The style labels are a preset number of labels selected from the style analysis results according to the size sequence.
In an alternative embodiment, for the recommendation module, selecting the first target garment from the garment database based on the category label comprises:
determining the type of the user according to the category label, and selecting clothing corresponding to the type of the user from a clothing database to obtain preselected clothing;
it is necessary to supplement that the clothing database records the clothing corresponding to the types of different users, and after the types of the users are determined, the clothing recorded in the clothing database can be screened according to the types of the users to obtain a clothing set corresponding to the types of the users, and the clothing set is marked as a preselected clothing.
For the users marked as the first type, screening preselected clothes according to the clothes size grade to obtain first target clothes; for the user marked as the second type, screening the preselected clothes according to the consumption level to obtain a first target clothes; for users tagged as a third type, the pre-selected garment is taken as the first target garment.
It is to be supplemented that the first type of user is specifically a user with a relatively special height and weight, the preselected garments are screened according to the garment size grades, and garments relatively fitting with the user can be obtained, the second type of user has a relatively low consumption grade and is relatively easy to select, when the type of user selects garments, the user can easily find own targets in a limited number of garments, and does not need to spend much time and effort to select garments.
In an alternative embodiment, for the recommendation module, selecting the second target garment from the first target garments based on the style label includes:
the first target clothes are primarily screened according to the style labels, the weight of each label in the style labels is determined, the recommended value of the clothes after primary screening is calculated based on the weight of each label in the style labels, the clothes after primary screening are secondarily screened according to a preset recommended threshold value, and the rest clothes are sorted according to the recommended value to obtain second target clothes.
The method includes the steps of firstly screening first target clothes according to style labels, specifically screening clothes containing any one of the style labels, extracting a sorting mode of the labels in the style labels, determining the weight of each label in the style labels according to the sorting mode, calculating a recommended value of the clothes according to the weight of each label, screening the clothes through a preset recommended threshold value in order to improve the recommendation accuracy, screening the clothes with the recommended value smaller than the preset recommended threshold value, and sorting the rest of the clothes according to the descending order of the recommended value to obtain second target clothes.
In an alternative embodiment, making a garment recommendation to the user based on the second target garment comprises:
and determining a recommendation period according to the clothing selection tendency level, acquiring a recommendation sequence of the second target clothing, and recommending the clothing in the second target clothing to the user according to the recommendation period and the recommendation sequence.
It is to be supplemented that, for different users, the clothing selection tendency levels may be the same or different, for users with lower clothing selection tendency levels, clothing selection is simpler, it is conceivable that the recommendation cycle may be longer, for users with higher clothing selection tendency levels, clothing selection is more difficult, so that the recommendation cycle for such users may be appropriately promoted, the recommendation sequence of the second target clothing is the sorting manner of the clothing in the second target clothing, and clothing recommendation is periodically performed to the users according to the recommendation sequence.
In an alternative embodiment, the present invention provides a system for recommending customized garment design, further comprising:
and the feedback module is used for acquiring browsing feedback data of the user and adjusting the recommendation sequence for recommending the clothes in the second target clothes to the user according to the browsing feedback data.
It should be added that the browsing feedback data is specifically the actual browsing condition of the recommended clothing by the user, including the picture browsing duration, the clothing detail browsing duration and the evaluation browsing duration, and the recommendation order adjustment parameter is calculated according to the following formula:
T=α 1 m 1 (c 1 D 1 +c 2 D 2 +c 3 D 3 );
wherein T is a recommended sequence adjustment parameter, m 1 For the total number of picture views, garment details views and evaluation data views, D 1 For picture browsing duration, D 2 For the duration of the browsing of the details of the garment, D 3 To evaluate the duration of the browsing, α 1 Is m 1 Weight parameter of c 1 、c 2 And c 3 Respectively, the picture browsing duration D 1 Garment detail browsing duration D 2 And evaluating the browsing duration D 3 The weight parameter of (2).
Wherein, for m 1 M is the time length of any one of the picture browsing time length, the clothing detail browsing time length and the evaluation browsing time length of the user is greater than zero 1 The numerical value of (a) is increased by one, and it is conceivable that m is the sum of the picture browsing time length, the clothing detail browsing time length and the evaluation browsing time length of the user which are all greater than zero 1 The value of (2) is 3.
After the recommendation sequence adjustment parameter T is obtained through calculation, the style labels of the recommended garments corresponding to the recommendation sequence adjustment parameter T are obtained, the weight of the style labels is adjusted through the recommendation sequence adjustment parameter T, and the recommendation values of the rest unrendered garments are corrected and updated to obtain a new recommendation sequence.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims. Parts of the description that are not described in detail are known to the person skilled in the art.
Claims (7)
1. A custom garment design recommendation system, the system comprising:
a data acquisition module;
the type analysis module is used for carrying out user type analysis on historical browsing data and historical order data of a user and adding a category label to the user;
the style analysis module is used for carrying out user style analysis on the historical browsing data and the historical order data of the user to obtain a style analysis result, and adding style labels to the user according to the style analysis result; the performing of the user style analysis comprises:
acquiring historical browsing data and historical order data of a user, dividing the historical browsing data into a plurality of subdata sets according to a time sequence, and creating a plurality of order sets according to the historical order data, wherein each order set comprises a historical order and a subdata set corresponding to the historical order;
for any order set, identifying a first label of a subdata set in the order set, sorting the first label according to the occurrence frequency, identifying a second label of a historical order in the order set, judging whether the sorting of the second label in the first label is greater than a preset ranking, and if so, taking the second label as a third label of the order set;
performing statistical analysis on third labels of all order sets, determining a frequency weight parameter of each label according to occurrence frequency, determining a time weight parameter of each label according to seasonal attributes, determining a correction parameter of each user according to the type of the user, calculating a reference value of each label according to the frequency weight parameter, the time weight parameter and the correction parameter, and sequencing the third labels according to the reference values to obtain style analysis results;
the reference value for any tag is calculated as follows:
R e =((n 1 P)(n 2 +n 3 ))/(n 1 +n 2 +n 3 );
in the formula, R e Denotes a reference value, P denotes the frequency of occurrence of the label, n 1 As a frequency weight parameter, n 2 Is a time weight parameter, n 3 To correct the parameters;
the style labels are a preset number of labels selected from the style analysis results according to the size sequence;
the system also comprises a recommendation module;
the performing the user type analysis includes: acquiring the historical browsing data and the historical order data of the user, determining the consumption grade and the clothing size grade of the user according to the historical order data, and determining the clothing selection tendency grade of the user according to the historical browsing data and the historical order data; the apparel selection propensity ratings include: dividing the historical browsing data into a plurality of subdata sets according to a time sequence, and creating a plurality of order sets according to the historical order data, wherein each order set comprises a historical order and a subdata set corresponding to the historical order; for any order set, calculating the number of times of browsing the clothes corresponding to the order set according to the subdata set in the order set; performing statistical analysis on the clothing browsing times of all the order sets, calculating to obtain an average value of the clothing browsing times, and determining the clothing selection tendency level of the user according to the average value of the clothing browsing times;
the feedback module is used for acquiring browsing feedback data of the user and adjusting the recommendation sequence for recommending the clothing of the second target clothing to the user according to the browsing feedback data; the method specifically comprises the following steps: the browsing feedback data is the actual browsing condition of the recommended clothing by the user, including the picture browsing duration, the clothing detail browsing duration and the evaluation browsing duration, and the recommended sequence adjustment parameter is calculated according to the following formula:
T=α 1 m 1 (c 1 D 1 +c 2 D 2 +c 3 D 3 );
wherein T is a recommended sequence adjustment parameter, m 1 For the total number of picture views, garment details views and evaluation data views, D 1 For picture browsing duration, D 2 For the duration of the browsing of the details of the garment, D 3 To evaluate the browsing duration, α 1 Is m 1 Weight parameter of c 1 、c 2 And c 3 Respectively the picture browsing duration D 1 Garment detail browsing duration D 2 And evaluating the browsing duration D 3 The weight parameter of (2); wherein, for m 1 M is the time length of any one of the picture browsing time length, the clothing detail browsing time length and the evaluation browsing time length of the user is greater than zero 1 Plus one;
after the recommendation sequence adjustment parameter T is obtained through calculation, the style labels of the recommended garments corresponding to the recommendation sequence adjustment parameter T are obtained, the weight of the style labels is adjusted through the recommendation sequence adjustment parameter T, and the recommendation values of the rest unrendered garments are corrected and updated to obtain a new recommendation sequence.
2. The custom garment design recommendation system of claim 1, wherein the data acquisition module is configured to acquire historical browsing data and historical order data of the user based on big data technology; in the step, an information acquisition application is sent to the user, and after an agreement reply of the user is received, historical browsing data and historical order data of the user are acquired, wherein the historical browsing data is browsing data of the user on the clothes, and the historical order data is order data of the user for purchasing the clothes.
3. The system of claim 1, wherein the type analysis module is configured to perform user type analysis on the historical browsing data and the historical order data of the user to obtain a type analysis result, and add a category tag to the user according to the type analysis result; performing user type analysis includes: acquiring historical browsing data and historical order data of a user, determining a consumption grade and a clothing size grade of the user according to the historical order data, and determining a clothing selection tendency grade of the user according to the historical browsing data and the historical order data;
the recommendation module is used for carrying out label identification on a user to obtain a category label and a style label of the user, selecting a first target garment from a garment database according to the category label, selecting a second target garment from the first target garment according to the style label, and recommending the garment to the user based on the second target garment;
the apparel selection propensity rating further includes: if the clothing size grade of the user does not belong to the preset clothing size range, marking the user as a first type; if the clothing size grade of the user belongs to the preset clothing size range, marking the user as a second type under the condition that the consumption grade of the user is smaller than a first preset grade and the clothing selection tendency grade is smaller than a second preset grade, otherwise, marking the user as a third type;
the category labels include a customer's consumption rating, a garment size rating, and a garment selection propensity rating, as well as the type of customer.
4. The customized garment design recommendation system of claim 3, wherein selecting a first target garment from a garment database based on the category label comprises:
determining the type of the user according to the category label, and selecting the clothing corresponding to the type of the user from the clothing database to obtain preselected clothing;
for users marked as a first type, screening the preselected garments according to the garment size grades to obtain the first target garment; for the users marked as the second type, screening the preselected clothes according to the consumption grades to obtain the first target clothes; for a user tagged as a third type, the pre-selected garment is taken as the first target garment.
5. The customized garment design recommendation system of claim 4, wherein selecting a second target garment from the first target garments based on the style label comprises:
and primarily screening the first target clothes according to the style labels, determining the weight of each label in the style labels, calculating the recommended value of the clothes subjected to primary screening based on the weight of each label in the style labels, secondarily screening the clothes subjected to primary screening according to a preset recommended threshold, and sequencing the rest clothes according to the recommended value to obtain the second target clothes.
6. The customized garment design recommendation system of claim 3, wherein the garment recommendation to the user based on the second target garment comprises:
determining a recommendation cycle according to the clothing selection tendency grade, acquiring a recommendation sequence of the second target clothing, and recommending the clothing in the second target clothing to the user according to the recommendation cycle and the recommendation sequence.
7. The system of claim 4, wherein the system for calculating the number of browsing times of the garment corresponding to the order set according to the sub data sets in the order set further comprises:
and screening historical browsing data in the sub-data set according to the preset browsing duration to obtain target browsing data with the browsing duration larger than the preset browsing duration, and counting clothing browsing times of the target browsing data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310279066.8A CN115983952B (en) | 2023-03-21 | 2023-03-21 | Custom-made clothing design recommendation system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310279066.8A CN115983952B (en) | 2023-03-21 | 2023-03-21 | Custom-made clothing design recommendation system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115983952A true CN115983952A (en) | 2023-04-18 |
CN115983952B CN115983952B (en) | 2023-12-08 |
Family
ID=85970906
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310279066.8A Active CN115983952B (en) | 2023-03-21 | 2023-03-21 | Custom-made clothing design recommendation system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115983952B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117473003A (en) * | 2023-12-28 | 2024-01-30 | 山东街景智能制造科技股份有限公司 | Database management method for vehicle customization |
CN117495513A (en) * | 2024-01-03 | 2024-02-02 | 深圳市智百威科技发展有限公司 | Shopping recommendation method and system based on big data and artificial intelligence |
CN117788124A (en) * | 2024-02-26 | 2024-03-29 | 山东街景智能制造科技股份有限公司 | Data management system and method for customized products |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20120036414A (en) * | 2010-10-08 | 2012-04-18 | 주식회사 엘지유플러스 | Method and system for managing clothes by user personalized type and server having function therefore |
CN108171569A (en) * | 2017-12-11 | 2018-06-15 | 武汉纺织大学 | A kind of reaction type intelligent clothing recommends method and system |
CN109165428A (en) * | 2018-08-08 | 2019-01-08 | 浙江敦奴联合实业股份有限公司 | A kind of intelligent clothing customization platform |
CN110059247A (en) * | 2019-03-19 | 2019-07-26 | 江苏皓之睿数字科技有限公司 | A kind of personalization based on artificial intelligence technology, which is worn, takes recommender system |
CN112150239A (en) * | 2020-09-10 | 2020-12-29 | 浙江网安文化发展有限公司 | Wearing image information recommendation method and device |
CN113450178A (en) * | 2021-05-13 | 2021-09-28 | 孙昊盛 | Internet-based e-commerce accurate marketing system |
US20210390430A1 (en) * | 2020-06-12 | 2021-12-16 | Rokkcb10, Inc. | Machine Learning System and Method for Garment Recommendation |
CN114238777A (en) * | 2022-02-18 | 2022-03-25 | 优游宝科技(深圳)有限公司 | Negative feedback flow distribution method, device, equipment and medium based on behavior analysis |
CN114529359A (en) * | 2021-12-27 | 2022-05-24 | 北京服装学院 | Shoe money type recommendation system and method based on user characteristics |
CN115047974A (en) * | 2022-06-20 | 2022-09-13 | 江西服装学院 | Virtual fitting interaction method and device for clothes |
-
2023
- 2023-03-21 CN CN202310279066.8A patent/CN115983952B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20120036414A (en) * | 2010-10-08 | 2012-04-18 | 주식회사 엘지유플러스 | Method and system for managing clothes by user personalized type and server having function therefore |
CN108171569A (en) * | 2017-12-11 | 2018-06-15 | 武汉纺织大学 | A kind of reaction type intelligent clothing recommends method and system |
CN109165428A (en) * | 2018-08-08 | 2019-01-08 | 浙江敦奴联合实业股份有限公司 | A kind of intelligent clothing customization platform |
CN110059247A (en) * | 2019-03-19 | 2019-07-26 | 江苏皓之睿数字科技有限公司 | A kind of personalization based on artificial intelligence technology, which is worn, takes recommender system |
US20210390430A1 (en) * | 2020-06-12 | 2021-12-16 | Rokkcb10, Inc. | Machine Learning System and Method for Garment Recommendation |
CN112150239A (en) * | 2020-09-10 | 2020-12-29 | 浙江网安文化发展有限公司 | Wearing image information recommendation method and device |
CN113450178A (en) * | 2021-05-13 | 2021-09-28 | 孙昊盛 | Internet-based e-commerce accurate marketing system |
CN114529359A (en) * | 2021-12-27 | 2022-05-24 | 北京服装学院 | Shoe money type recommendation system and method based on user characteristics |
CN114238777A (en) * | 2022-02-18 | 2022-03-25 | 优游宝科技(深圳)有限公司 | Negative feedback flow distribution method, device, equipment and medium based on behavior analysis |
CN115047974A (en) * | 2022-06-20 | 2022-09-13 | 江西服装学院 | Virtual fitting interaction method and device for clothes |
Non-Patent Citations (2)
Title |
---|
KUAN HSIEN LIU ETAL: "clothing recommendation based on deep learning", 《IEEE》, pages 281 - 282 * |
崔丽芸: "服装设计专业毕业成衣作品现象解析", 《包装世界》, no. 6, pages 84 - 85 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117473003A (en) * | 2023-12-28 | 2024-01-30 | 山东街景智能制造科技股份有限公司 | Database management method for vehicle customization |
CN117473003B (en) * | 2023-12-28 | 2024-03-22 | 山东街景智能制造科技股份有限公司 | Database management method for vehicle customization |
CN117495513A (en) * | 2024-01-03 | 2024-02-02 | 深圳市智百威科技发展有限公司 | Shopping recommendation method and system based on big data and artificial intelligence |
CN117495513B (en) * | 2024-01-03 | 2024-03-29 | 深圳市智百威科技发展有限公司 | Shopping recommendation method and system based on big data and artificial intelligence |
CN117788124A (en) * | 2024-02-26 | 2024-03-29 | 山东街景智能制造科技股份有限公司 | Data management system and method for customized products |
CN117788124B (en) * | 2024-02-26 | 2024-05-14 | 山东街景智能制造科技股份有限公司 | Data management system and method for customized products |
Also Published As
Publication number | Publication date |
---|---|
CN115983952B (en) | 2023-12-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115983952B (en) | Custom-made clothing design recommendation system | |
CN109672939B (en) | Method and device for marking video content popularity | |
JP4426563B2 (en) | Information provision system | |
US20120179716A1 (en) | System for recommending an article not present in an image | |
KR101579376B1 (en) | Personalized place recommendation system and method by using subjectivity analysis for user classification | |
CN106355449A (en) | User selecting method and device | |
CN112102037A (en) | Live E-commerce platform commodity content intelligent pushing management system based on big data | |
KR101827345B1 (en) | Personalized recommendation system and its method using multiple algorithms and self-learning function | |
US11836813B2 (en) | Producing garments from patterns with size grid providing custom-like fit | |
CN105608650A (en) | Dish collocation recommending method and system | |
CN111915409B (en) | Item recommending method, device, equipment and storage medium based on item | |
KR101755653B1 (en) | user-tailored webtoon auto-recommendation system and method | |
CN108132964A (en) | A kind of collaborative filtering method to be scored based on user item class | |
CN106547808A (en) | Picture update method, classification sort method and device | |
CN110889043A (en) | Diet recommendation system and method based on big data | |
US20220075804A1 (en) | Method and device for providing guide information for enhancement of artist's reputation | |
JP4844669B2 (en) | Information provision system | |
CN113129078A (en) | Social software advertisement information delivery method based on feature recognition and data visualization analysis | |
CN110110206A (en) | Relation excavation and recommended method, device, calculating equipment, storage medium between article | |
CN115880037A (en) | Commodity recommendation method based on multi-project planning integration analysis | |
CN113625916B (en) | Entertainment device for interaction and data acquisition and transmission | |
CN113507623B (en) | Intelligent commodity selection system and method based on data analysis | |
CN112989227A (en) | Method and system for selecting target address of interested object | |
JP5370351B2 (en) | Information processing method, information processing apparatus, and information processing program | |
CN109886756A (en) | Communication user upshift prediction probability recognition methods and system based on integrated model |
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 | ||
GR01 | Patent grant |