CN105701680B - Personalized recommendation method and system based on opposite attribute knowledge base - Google Patents
Personalized recommendation method and system based on opposite attribute knowledge base Download PDFInfo
- Publication number
- CN105701680B CN105701680B CN201511034894.7A CN201511034894A CN105701680B CN 105701680 B CN105701680 B CN 105701680B CN 201511034894 A CN201511034894 A CN 201511034894A CN 105701680 B CN105701680 B CN 105701680B
- Authority
- CN
- China
- Prior art keywords
- user
- recommendation
- opposite
- attribute
- recommendation result
- 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.)
- Active
Links
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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
- G06Q30/0256—User search
Abstract
The invention discloses a personalized recommendation method and a system, wherein the method comprises the following steps: acquiring a recommendation result sequence recommended to a user by a current recommendation system; acquiring a plurality of preset recommendation results in a preset direction from the recommendation result sequence as a primary recommendation result sequence; inquiring whether to store the opposite attribute of the user in an opposite attribute knowledge base according to the identity information of the user; when the query result is yes, matching each recommendation result in the primary recommendation result sequence with the opposite attribute of the user; deleting recommendation results, of which the matching results of the opposite attributes of the users in the initial recommendation result sequence meet preset conditions; obtaining a final recommendation result sequence according to the rest recommendation results of the primary recommendation result sequence; and outputting a final recommendation result sequence. According to the invention, the accuracy of recommending the user is improved, the adoption rate of the user to the recommending result is further improved, and the value of the recommending system to the user is improved.
Description
Technical Field
The invention relates to the technical field of recommendation, in particular to an individualized recommendation method and system based on an opposite attribute knowledge base.
Background
With the continuous expansion of the electronic commerce scale, the number and the types of the commodities are rapidly increased, and a user needs to spend a great deal of time to find the commodity which the user wants to buy. The process of browsing through large amounts of extraneous information and products is undoubtedly a constant loss to consumers. To solve these problems, personalized recommendation techniques have been developed. The personalized recommendation technology is a high-level business intelligent platform established on the basis of mass data mining to help an e-commerce website to provide completely personalized decision support and information service for shopping of customers.
However, when the existing personalized recommendation system recommends based on the analysis of the historical data of the purchased goods, wrong recommendations may occur. For example, if the recommendation system finds that the past interest features and purchases of users a and B are similar, and recently the user a purchased a sanitary napkin, the result recommendation system recommends a sanitary napkin to the user B, is this recommendation accurate? The interest characteristics and purchasing behaviors of the user A and the user B in the past are similar because the user A and the user B are sisters, but the user A is a female and the menstruation starts to begin to take a sanitary towel for the first time, but the user B is a male, and the sanitary towel is recommended to the user B, which is obviously wrong recommendation. Therefore, the recommendation result obtained by the prior recommendation technology is often not consistent with the commodity which the user wants to buy, so that wrong recommendation is caused, the adoption rate of the recommendation result by the user is reduced, and the value of a recommendation system to the user is reduced.
Disclosure of Invention
Based on the situation, the invention provides a personalized recommendation method and system, which can improve the accuracy of recommendation for users, further improve the adoption rate of recommendation results for users, and improve the value of a recommendation system for users.
In order to achieve the above purpose, the embodiment of the technical scheme of the invention is as follows:
a personalized recommendation method, comprising the steps of:
acquiring a recommendation result sequence recommended to a user by a current recommendation system;
acquiring a plurality of preset recommendation results in a preset direction from the recommendation result sequence as a primary recommendation result sequence, wherein the number of the preset recommendations is less than or equal to the total number of the recommendation results in the recommendation result sequence;
inquiring whether to store the opposite attribute of the user in an opposite attribute table of the user pre-stored in an opposite attribute knowledge base according to the identity information of the user;
when the query result is yes, matching each recommendation result in the primary recommendation result sequence with the opposite attribute of the user respectively;
deleting the recommendation result of which the matching result of the opposite attributes of the user in the primary recommendation result sequence meets the preset condition;
obtaining a final recommendation result sequence according to the rest recommendation results of the primary recommendation result sequence;
and outputting the final recommendation result sequence.
A personalized recommendation system, comprising:
the recommendation result sequence acquisition module is used for acquiring a recommendation result sequence recommended to a user by the current recommendation system;
the primary recommendation result sequence acquisition module is used for acquiring a plurality of preset recommendation results in a preset direction from the recommendation result sequence as a primary recommendation result sequence, wherein the number of the preset recommendations is less than or equal to the total number of the recommendation results in the recommendation result sequence;
the attribute query module is used for querying whether the opposite attribute of the user is stored in a user opposite attribute table pre-stored in an opposite attribute knowledge base according to the identity information of the user;
the result matching module is used for respectively matching each recommendation result in the primary recommendation result sequence with the opposite attribute of the user when the query result is yes;
the result deleting module is used for deleting the recommendation result of which the matching result of the opposite attributes of the user in the initial recommendation result sequence meets the preset condition;
a final recommendation result sequence obtaining module, configured to obtain a final recommendation result sequence according to the remaining recommendation results of the primary recommendation result sequence;
and the sequence output module is used for outputting the final recommendation result sequence.
Compared with the prior art, the invention has the beneficial effects that: the personalized recommendation method and the system are based on the opposite attribute knowledge base, match a plurality of preset recommendation results recommended to the user by the current recommendation system with the user opposite attributes pre-stored in the opposite attribute knowledge base, and acquire a final recommendation result sequence according to the matching results, so that the accuracy of recommending the user is improved, the personalized recommendation requirements of the user are met, the adoption rate of the user on the recommendation results is improved, the value of the recommendation system on the user is improved, and the personalized recommendation method and the system are suitable for application.
Drawings
FIG. 1 is a flow chart illustrating a personalized recommendation method according to an embodiment of the present invention;
FIG. 2 is a flow chart of a personalized recommendation method based on one specific example of the method shown in FIG. 1;
fig. 3 is a schematic structural diagram of a personalized recommendation system according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the detailed description and specific examples, while indicating the scope of the invention, are intended for purposes of illustration only and are not intended to limit the scope of the invention.
In one embodiment, the personalized recommendation method, as shown in fig. 1, includes the following steps:
step S101: acquiring a recommendation result sequence recommended to a user by a current recommendation system;
the current recommendation system can be various existing recommendation systems or newly developed recommendation systems; the recommendation result recommended to the user by the recommendation system can be various types of recommendation results, such as commodity recommendation, clothes recommendation, book recommendation, video recommendation, picture recommendation, paper recommendation or friend recommendation and the like;
step S102: acquiring a plurality of preset recommendation results in a preset direction from the recommendation result sequence as a primary recommendation result sequence, wherein the number of the preset recommendations is less than or equal to the total number of the recommendation results in the recommendation result sequence;
for example, the number of recommendation results recommended to a user by the recommendation system is recorded as p, the first n recommendation results in the p recommendation results are used as n first recommendation results to obtain a primary recommendation result sequence, wherein p can be a natural number, the recommendation system can recommend at least one recommendation result to the user, and all or part of all recommendation results recommended to the user by the recommendation system are selected as the first recommendation results;
step S103: inquiring whether to store the opposite attribute of the user in an opposite attribute table of the user pre-stored in an opposite attribute knowledge base according to the identity information of the user;
for example, the opposite attribute of the user is retrieved from the user opposite attribute table in the opposite attribute knowledge base, the user opposite attribute table is retrieved through the identity information of the user, and when the identity information of the corresponding user is retrieved, the opposite attribute of the user corresponding to the identity information of the user is taken out; the opposite attributes of the user can be collected in advance and stored in an opposite attribute knowledge base;
step S104: when the query result is yes, matching each recommendation result in the primary recommendation result sequence with the opposite attribute of the user respectively;
the matching degree of a recommendation result and the opposite attribute of the user in the opposite attribute knowledge base is essentially the contradiction degree of the recommendation result and the user attribute; from the matching degree of the recommendation result and the opposite attribute of the user, the contradiction degree of the recommendation result and the attribute of the user can be seen, and the higher the matching degree of the recommendation result and the opposite attribute of the user is, the higher the contradiction degree of the recommendation result and the attribute of the user is;
step S105: deleting the recommendation result of which the matching result of the opposite attributes of the user in the primary recommendation result sequence meets the preset condition;
for example, when the matching degree of a recommendation result and the opposite attribute of the user is greater than a preset matching degree value (the preset matching degree value is minimum 0), the recommendation result is indicated to be contradictory to the attribute of the user, and the recommendation result is deleted; the calculation of the matching degree can be converted into the calculation of the matching degree or the similarity of the character strings, and the existing calculation method of the matching degree or the similarity of the character strings can be adopted, such as an Edit distance method (an Edit distance is used for calculating the minimum number of insertions, deletions and replacements required for converting an original string(s) into a target string (t); the matching degree can also be calculated by using a new matching degree algorithm, such as the common number of characters of two character strings as the matching degree;
step S106: obtaining a final recommendation result sequence according to the rest recommendation results of the primary recommendation result sequence;
if the matching degree of one recommendation result and the opposite attribute of the user is greater than the preset matching degree value, the fact that the recommendation result is inconsistent with the attribute of the user is indicated, and the recommendation result which is left after the recommendation result with the matching degree of the opposite attribute of the user in the initial recommendation result sequence which is greater than the preset matching degree value is deleted is used as a final recommendation result sequence;
step S107: and outputting the final recommendation result sequence.
The way of outputting the final recommendation result to the user may be the way adopted by the existing recommendation system, and may also adopt other information output ways, such as a web page way and a file way.
From the above description, the personalized recommendation method of the invention is based on the opposite attribute knowledge base, so that recommendation results inconsistent with the user attributes are greatly eliminated, the requirements of personalized recommendation of users are met, the recommendation accuracy is improved, the adoption rate of the recommendation results by the users is improved, and the value of the recommendation system to the users is improved.
In addition, in a specific example, the identity information of the user includes a user ID (identification number), the user opposite attribute table includes a user field and a user opposite attribute field, the user field stores the user ID, the user opposite attribute field stores the user opposite attribute, the user opposite attribute is obtained according to the user attribute, and the user attribute includes any one or any combination of the age, gender, occupation, academic calendar, profession, speciality, hobby and geographic location of the user.
The user opposite attribute table in the opposite attribute knowledge base comprises a user field and a user opposite attribute field, wherein the user field stores a user ID, and the user opposite attribute field stores the opposite attribute of the user. And retrieving the opposite attribute of the user from the opposite attribute knowledge base, namely retrieving the opposite attribute knowledge base through the user ID, and when the corresponding user ID is retrieved, retrieving the opposite attribute of the user corresponding to the user ID. The opposite attributes of the user are obtained according to the attributes of the user, the attributes of the user can comprise information related to the user, such as the age, the sex, the occupation, the academic calendar, the specialty, the speciality, the hobby, the geographic position and the like of the user, and the requirements of various applications are met.
And acquiring opposite attributes of the user: firstly, inquiring the antisense words of keywords in the attributes of the user; when the antisense words can be inquired, the antisense words are used as opposite attributes of the user; and when the anti-synonym cannot be inquired, inquiring the keywords of the same type which are farthest away from the keywords in the database according to the keywords in the attributes of the user as the opposite attributes of the user. The database stores various types of keywords and distances between the keywords in advance, where the distances refer to differences, for example, the keyword which is the same as the academic calendar type is farthest from the primary school and obviously is "postdoctor".
In addition, in a specific example, when the query result is negative, whether the user is a registered user of the current recommendation system is judged;
when the judgment result is yes, acquiring the attribute of the user from the registration information of the user of the current recommendation system, obtaining the opposite attribute of the user according to the attribute of the user, and storing the opposite attribute of the user in the opposite attribute knowledge base;
and when the judgment result is negative, generating an information acquisition window, acquiring the attributes of the user, obtaining the opposite attributes of the user according to the attributes of the user, and storing the opposite attributes of the user in the opposite attribute knowledge base.
For example, the opposite attribute of the user is retrieved from the opposite attribute repository, and when the user or the opposite attribute of the user is not retrieved from the opposite attribute repository, judging whether the user is a registered user of the recommendation system, inquiring the user attribute in the registration information of the user when the user is the registered user, and the opposite attribute of the user is obtained according to the attribute of the user and added into an opposite attribute knowledge base, when the user is not a registered user, the pop-up dialog box inquires the user, or acquires the attribute of the user in other interactive modes or inquiry modes, acquires the opposite attribute of the user according to the attribute of the user and adds the opposite attribute into an opposite attribute knowledge base, if the user attribute information does not exist in the registration information of the user, the user can be inquired through a pop-up dialog box or the attribute of the user can be acquired in other interactive modes, and the opposite attribute of the user is acquired according to the attribute of the user and is added into the opposite attribute knowledge base.
In addition, in a specific example, the step of respectively matching each recommendation result in the primary recommendation result sequence with the opposite attribute of the user includes:
converting each recommendation result in the primary recommendation result sequence and the opposite attributes of the user into character strings respectively;
and respectively calculating the matching degree of the character string converted by each recommendation result in the primary recommendation result sequence and the character string converted by the opposite attribute of the user.
The recommendation results and the opposite attributes of the users in the primary recommendation result sequence can be converted into character strings, the calculation of the matching degree of the recommendation results and the opposite attributes of the users can be converted into the calculation of the matching degree or the similarity of the character strings, the degree of contradiction between the recommendation results and the attributes of the users can be seen from the matching degree of the recommendation results and the opposite attributes of the users, and the higher the matching degree of the recommendation results and the opposite attributes of the users is, the higher the degree of contradiction between the recommendation results and the attributes of the users is.
The calculation of the matching degree can be converted into the calculation of the matching degree or the similarity of the character strings, and the existing calculation method of the matching degree or the similarity of the character strings can be adopted, such as an Edit distance method (an Edit distance is used for calculating the minimum number of insertions, deletions and replacements required for converting an original string(s) into a target string (t); the matching degree may be calculated by using a new matching degree algorithm, such as a common number of characters of two character strings as the matching degree.
In addition, in a specific example, the step of deleting the recommendation result in the initial recommendation result sequence whose matching result with the opposite attribute of the user meets a preset condition includes:
respectively acquiring the same number of characters of the character string converted by each recommendation result in the primary recommendation result sequence and the character string converted by the opposite attribute of the user;
and deleting the recommendation result of which the number of the same characters of the character string converted from the opposite attribute of the user in the initial recommendation result sequence is greater than a preset value.
The number of the same characters of the character string converted by one recommendation result and the character string converted by the opposite attribute of the user is larger than a preset value (for example, larger than 0), which indicates that the recommendation result is contradictory to the attribute of the user, and the recommendation result contradictory to the attribute of the user is deleted, so that the recommendation accuracy is improved.
In order to better understand the above method, an application example of the personalized recommendation method of the present invention is described in detail below.
As shown in fig. 2, the application instance may include the following steps:
step S201: acquiring a recommendation result sequence recommended to a user A by a recommendation system of a shopping website;
step S202: acquiring the first 11 recommendation results in the recommendation result sequence as a primary recommendation result sequence, wherein the total number of the recommendation results in the recommendation result sequence is greater than or equal to 11; the 11 recommendation results are: (1) 45g of total ultraviolet-proof isolation emulsion SPF30 of a beauty sunscreen female waterproof genuine whitening sun-proof essence for resisting ultraviolet rays; (2) a new type big net shoe for men, a sandal, a student leisure sports shoe for men, a summer net cloth shoe and a men's big size man shoe; (3) the cover tide is covered after the cover tide is protected by ELIFE7 shell GN9006 transparent silica gel of the cell phone cover ELIFE7 shell of Zhongge Jinli S7; (4) iphone4s Mobile phone Shell apple 5s Shell ultrathin plastic frosted protection hard Shell Black and white Red Tiger men and women brief; (5) a bag post man bag, a thickened canvas double-shoulder bag, a leisure travel bag, a tide bag and a Korean male backpack; (6) the summer vest sling 8520300114 of the printed sleeveless garment worn outside summer by the new summer vest with Yinman 2015; (7) dark valley bird Korean tide 2015 spring and autumn women canvas shoes high-upper height increasing women's shoes with thick base cloth; (8) big sim south korea custom-made summer clothing is a simple round-neck loose T-shirt with a hole and simple color; (9) middle-aged and old women summer T-shirt chiffon shirt jacket big-size mother's suit loose embroidered short-sleeve old person clothes; (10) a millet 2s mobile phone protective shell II is covered with a mobile phone cover and a millet 2 leather cover shell m2 ultrathin hard flip cover for stamping; (11) a korean east gate 2015 newly-dressed women in summer wear a fashionable flower-breaking loose short-sleeve chiffon cake shirt short-style jacket;
step S203: inquiring whether to store the opposite attribute of the user A in a user opposite attribute table pre-stored in an opposite attribute knowledge base according to the ID of the user A; the user opposite attribute table comprises a user field and a user opposite attribute field, wherein the user field stores a user ID, the user opposite attribute field stores the user opposite attribute, the user opposite attribute is obtained according to the user attribute, and the user attribute comprises the age and the gender of the user; the opposite attribute knowledge base can store the opposite attributes of the user in advance; in one embodiment, the user attribute table is shown in table 1, and the user opposite attribute table is shown in table 2;
TABLE 1 user Attribute Table
| User ID | User attributes |
| 14233 | Old man |
| 14234 | Younger women |
| 14235 | Younger male |
| 14236 | Elderly female |
TABLE 2 user inverse Attribute Table
Step S204: when the query result is yes, respectively converting the 11 recommendation results and the opposite attributes of the user A into character strings; when the query result is negative, judging whether the user A is a registered user of the shopping website; if so, acquiring the attribute of the user A from the registration information of the user A of the shopping website, obtaining the opposite attribute of the user A according to the attribute of the user A, and storing the opposite attribute of the user A in an opposite attribute knowledge base; when the judgment result is negative, generating an information acquisition window, acquiring the attribute of the user A, obtaining the opposite attribute of the user A according to the attribute of the user A, and storing the opposite attribute of the user A in an opposite attribute knowledge base;
knowing that the ID of the user a is 14235, it can be queried from the user opposite attribute table pre-stored in the opposite attribute knowledge base that the opposite attribute of the user a is "old female";
if the opposite attribute of the user A cannot be inquired in the user opposite attribute table stored in the opposite attribute knowledge base in advance, judging whether the user A is a registered user of the shopping website, inquiring the attribute of the user A in the registration information of the user when the user A is the registered user, obtaining the opposite attribute of the user A according to the attribute of the user A, adding the opposite attribute into an opposite attribute knowledge base, when the user A is not a registered user, the user A is inquired by a pop-up dialog box, the attribute of the user A can be acquired in other interactive modes or inquiry modes, the opposite attribute of the user A is acquired according to the attribute of the user A and is added into an opposite attribute knowledge base, if the user registration information does not have the attribute of the user A, inquiring the user or acquiring the attribute of the user A in other interactive modes through a pop-up dialog box, and acquiring the opposite attribute of the user A according to the attribute of the user A and adding the opposite attribute into an opposite attribute knowledge base;
the attribute of the user A is 'young male', the keywords 'young' and 'male', the antisense words 'old' and 'female' of the keywords are obtained through inquiry, and the 'old female' is used as the opposite attribute of the user A;
step S205: respectively calculating the matching degree of the character strings converted by the 11 recommendation results and the character strings converted by the opposite attributes of the user A;
the calculation of the matching degree can be converted into the calculation of the matching degree or the similarity of the character strings, and the common character number of the two character strings can be used as the size of the matching degree; the degree of contradiction between the recommendation result and the attribute of the user can be seen from the matching degree between the recommendation result and the opposite attribute of the user; the higher the matching degree of one recommendation result and the opposite attribute of the user is, the higher the contradiction degree of the recommendation result and the attribute of the user is;
step S206: respectively acquiring the same number of characters of the character string converted by the 11 recommendation results and the character string converted by the opposite attribute of the user A;
the matching degree adopts a calculation mode: the same number of characters of the two character strings is used as the size of the matching degree:
(1) the same number of characters of character strings of 45g ultraviolet-proof whole-body aged women and the character strings of the reverse attribute conversion of the user A are 1;
(2) the same number of characters of character strings converted from the opposite attributes of the old men's shoes and the old men's shoes with big codes and the user's shoes with stamp wrapping, new-style big net shoes, men's sandals, students, leisure sports shoes, men's summer net cloth shoes and men's shoes is 0;
(3) the same number of characters of character strings which are covered by the reverse attribute conversion of the aged women and the user A after the soft sleeve shell accessory is protected by ELIFE7 shell GN9006 transparent silica gel of the Jinli S7 cell phone cover of Zhongge is 0;
(4) the same number of characters of character strings of opposite attribute conversion of simple and old men and women in black and white red tide with an iphone4s mobile phone shell, an apple 5s shell and ultrathin plastic frosted protective hard shell and a user A is 1;
(5) the same number of characters of character strings of the reversed attribute conversion between the old women of the men's bag and the old women of the Korean-version men's backpack and the user A is 0;
(6) the same number of characters of character strings converted from the opposite attributes of an old woman and a user A is 1 by using a Twenman 2015 summer new vest woman to wear a printed sleeveless shirt summer vest sling 8520300114 outside summer;
(7) the same number of characters of character strings of opposite attribute conversion of old women and user A of the Sengu bird Korean tide 2015 spring and autumn women canvas shoes is 2;
(8) the same number of characters of character strings of opposite attribute conversion of old women of big sim Korean summer clothing customized hole breaking simple round collar loose short-sleeved girl T-shirt and user A is 1;
(9) the same number of characters of character strings of opposite attribute conversion of the old women of the middle-aged and the old women summer T-shirt blouse, big code, mother dress, loose embroidery short-sleeve old person clothes and the user A is 3;
(10) the millet 2s mobile phone protective shell II is covered with a mobile phone cover II, the millet 2 leather cover shell m2 ultrathin hard flip cover covers the same characters of character strings with opposite attribute conversion between the old female and the user A, and the number of the characters is 0;
(11) the same number of characters of character strings converted by the opposite attributes of old women who wear fashionable, flower-breaking, loose and short sleeve chiffon cake shirt jacket for new women in summer and the user A is 1;
step S207: deleting the recommendation results of which the number of the same characters of the character string converted by the attribute opposite to that of the user A in the 11 recommendation results is not zero;
namely deletion:
45g of total ultraviolet-proof isolation emulsion SPF30 of a beauty sunscreen female waterproof genuine whitening sun-proof essence for resisting ultraviolet rays;
iphone4s Mobile phone Shell apple 5s Shell ultrathin plastic frosted protection hard Shell Black and white Red Tiger men and women brief;
the summer vest sling 8520300114 of the printed sleeveless garment worn outside summer by the new summer vest with Yinman 2015;
dark valley bird Korean tide 2015 spring and autumn women canvas shoes high-upper height increasing women's shoes with thick base cloth;
big sim south korea custom-made summer clothing is a simple round-neck loose T-shirt with a hole and simple color;
middle-aged and old women summer T-shirt chiffon shirt jacket big-size mother's suit loose embroidered short-sleeve old person clothes;
a korean east gate 2015 newly-dressed women in summer wear a fashionable flower-breaking loose short-sleeve chiffon cake shirt short-style jacket;
step S208: obtaining a final recommendation result sequence according to the rest recommendation results of the 11 recommendation result sequences:
(1) a new type big net shoe for men, a sandal, a student leisure sports shoe for men, a summer net cloth shoe and a men's big size man shoe;
(2) the cover tide is covered after the cover tide is protected by ELIFE7 shell GN9006 transparent silica gel of the cell phone cover ELIFE7 shell of Zhongge Jinli S7;
(3) a bag post man bag, a thickened canvas double-shoulder bag, a leisure travel bag, a tide bag and a Korean male backpack;
(4) a millet 2s mobile phone protective shell II is covered with a mobile phone cover and a millet 2 leather cover shell m2 ultrathin hard flip cover for stamping;
step S209: and outputting the final recommendation result sequence.
The way of outputting the final recommendation result to the user may be the way adopted by the existing recommendation system, and may also adopt other information output ways, such as a web page way and a file way.
In the application example, the recommendation result left after the recommendation result with the matching degree of the opposite attribute of the user A being not 0 in the recommendation results 11 is deleted is used as the final recommendation result, so that the recommendation result inconsistent with the user attribute is greatly eliminated, the requirement of the user on personalized recommendation is met, the recommendation accuracy is improved, the adoption rate of the user on the recommendation result is improved, and the value of the recommendation system on the user is improved.
In one embodiment, the personalized recommendation system, as shown in fig. 3, includes:
a recommendation result sequence obtaining module 301, configured to obtain a recommendation result sequence recommended to a user by a current recommendation system;
a primary recommendation result sequence obtaining module 302, configured to obtain, in the recommendation result sequence, a preset number of recommendation results in a preset direction as a primary recommendation result sequence, where the preset number of recommendations is less than or equal to a total number of recommendation results in the recommendation result sequence;
the attribute query module 303 is configured to query whether to store the opposite attribute of the user in an opposite attribute table of the user, which is pre-stored in an opposite attribute repository, according to the identity information of the user;
a result matching module 304, configured to, when the query result is yes, match each recommendation result in the primary recommendation result sequence with an opposite attribute of the user respectively;
a result deleting module 305, configured to delete a recommendation result in the primary recommendation result sequence, where a matching result of the opposite attribute of the user meets a preset condition;
a final recommendation result sequence obtaining module 306, configured to obtain a final recommendation result sequence according to the remaining recommendation results of the primary recommendation result sequence;
and a sequence output module 307, configured to output the final recommendation result sequence.
In addition, in a specific example, the identity information of the user includes a user ID, the user opposite attribute table includes a user field and a user opposite attribute field, the user field stores the user ID, the user opposite attribute field stores the user opposite attribute, the user opposite attribute is obtained according to the user attribute, and the user attribute includes any one or any combination of the age, the gender, the occupation, the academic calendar, the specialty, the speciality, the hobby and the geographic location of the user.
The user opposite attribute table in the opposite attribute knowledge base comprises a user field and a user opposite attribute field, wherein the user field stores a user ID, and the user opposite attribute field stores the opposite attribute of the user. And retrieving the opposite attribute of the user from the opposite attribute knowledge base, namely retrieving the opposite attribute knowledge base through the user ID, and when the corresponding user ID is retrieved, retrieving the opposite attribute of the user corresponding to the user ID. The opposite attributes of the user are obtained according to the attributes of the user, the attributes of the user can comprise information related to the user, such as the age, the sex, the occupation, the academic calendar, the specialty, the speciality, the hobby, the geographic position and the like of the user, and the requirements of various applications are met.
And acquiring opposite attributes of the user: firstly, inquiring the antisense words of keywords in the attributes of the user; when the antisense words can be inquired, the antisense words are used as opposite attributes of the user; and when the anti-synonym cannot be inquired, inquiring the keywords of the same type which are farthest away from the keywords in the database according to the keywords in the attributes of the user as the opposite attributes of the user. The database stores various types of keywords and distances between the keywords in advance, where the distances refer to differences, for example, the keyword which is the same as the academic calendar type is farthest from the primary school and obviously is "postdoctor".
As shown in fig. 3, in a specific example, the system further includes an attribute obtaining module 308, configured to determine whether the user is a registered user of the current recommendation system when the query result is negative;
when the judgment result is yes, acquiring the attribute of the user from the registration information of the user of the current recommendation system, obtaining the opposite attribute of the user according to the attribute of the user, and storing the opposite attribute of the user in the opposite attribute knowledge base;
and when the judgment result is negative, generating an information acquisition window, acquiring the attributes of the user, obtaining the opposite attributes of the user according to the attributes of the user, and storing the opposite attributes of the user in the opposite attribute knowledge base.
For example, the opposite attribute of the user is retrieved from the opposite attribute repository, and when the user or the opposite attribute of the user is not retrieved from the opposite attribute repository, judging whether the user is a registered user of the recommendation system, inquiring the user attribute in the registration information of the user when the user is the registered user, and the opposite attribute of the user is obtained according to the attribute of the user and added into an opposite attribute knowledge base, when the user is not a registered user, the pop-up dialog box inquires the user, or acquires the attribute of the user in other interactive modes or inquiry modes, acquires the opposite attribute of the user according to the attribute of the user and adds the opposite attribute into an opposite attribute knowledge base, if the user attribute information does not exist in the registration information of the user, the user can be inquired through a pop-up dialog box or the attribute of the user can be acquired in other interactive modes, and the opposite attribute of the user is acquired according to the attribute of the user and is added into the opposite attribute knowledge base.
As shown in fig. 3, in a specific example, the result matching module 304 includes:
a converting unit 3041, configured to convert each recommendation result in the primary recommendation result sequence and the opposite attribute of the user into a character string respectively;
a matching unit 3042, configured to calculate matching degrees between the character strings converted by the recommendation results in the primary recommendation result sequence and the character strings converted by the opposite attributes of the user, respectively.
The recommendation results and the opposite attributes of the users in the primary recommendation result sequence can be converted into character strings, the calculation of the matching degree of the recommendation results and the opposite attributes of the users can be converted into the calculation of the matching degree or the similarity of the character strings, the degree of contradiction between the recommendation results and the attributes of the users can be seen from the matching degree of the recommendation results and the opposite attributes of the users, and the higher the matching degree of the recommendation results and the opposite attributes of the users is, the higher the degree of contradiction between the recommendation results and the attributes of the users is.
The calculation of the matching degree can be converted into the calculation of the matching degree or the similarity of the character strings, and the existing calculation method of the matching degree or the similarity of the character strings can be adopted, such as an Edit distance method (an Edit distance is used for calculating the minimum number of insertions, deletions and replacements required for converting an original string(s) into a target string (t); the matching degree may be calculated by using a new matching degree algorithm, such as a common number of characters of two character strings as the matching degree.
As shown in fig. 3, in a specific example, the result deleting module 305 includes:
an obtaining unit 3051, configured to obtain the same number of characters of a character string converted by each recommendation result in the primary recommendation result sequence and a character string converted by an opposite attribute of the user, respectively;
a deleting unit 3052, configured to delete the recommendation result of which the number of the same characters in the string converted by the opposite attribute of the user in the primary recommendation result sequence is greater than a preset value.
The number of the same characters of the character string converted by one recommendation result and the character string converted by the opposite attribute of the user is larger than a preset value (for example, larger than 0), which indicates that the recommendation result is contradictory to the attribute of the user, and the recommendation result contradictory to the attribute of the user is deleted, so that the recommendation accuracy is improved.
Based on the system of the embodiment shown in fig. 3, a specific working process may be as follows:
firstly, a recommendation result sequence acquisition module 301 acquires a recommendation result sequence recommended to a user by a current recommendation system; the primary recommendation result sequence obtaining module 302 obtains a preset number of recommendation results in a preset direction from the recommendation result sequence as a primary recommendation result sequence, where the preset first number of recommendations is less than or equal to the total number of recommendation results in the recommendation result sequence; the attribute query module 303 queries whether to store the opposite attribute of the user in a user opposite attribute table pre-stored in an opposite attribute knowledge base according to the identity information of the user; when the query result is yes, the conversion unit 3041 in the result matching module 304 converts each recommendation result in the primary recommendation result sequence and the opposite attribute of the user into a character string respectively; the matching unit 3042 calculates the matching degree between the character string converted by each recommendation result in the primary recommendation result sequence and the character string converted by the opposite attribute of the user; when the query result is negative, the attribute obtaining module 308 determines whether the user is a registered user of the current recommendation system; when the judgment result is yes, acquiring the attribute of the user from the registration information of the user of the current recommendation system, obtaining the opposite attribute of the user according to the attribute of the user, and storing the opposite attribute of the user in the opposite attribute knowledge base; when the judgment result is negative, generating an information acquisition window, acquiring the attributes of the user, obtaining the opposite attributes of the user according to the attributes of the user, and storing the opposite attributes of the user in the opposite attribute knowledge base; an obtaining unit 3051 in the result deleting module 305 obtains the same number of characters of the character string converted by each recommendation result in the primary recommendation result sequence and the character string converted by the opposite attribute of the user respectively; the deleting unit 3052 deletes the recommendation result in which the number of the same characters of the character string converted from the opposite attribute of the user in the primary recommendation result sequence is greater than a preset value; the final recommendation result sequence obtaining module 306 obtains a final recommendation result sequence according to the rest recommendation results of the primary recommendation result sequence; the sequence output module 307 outputs the final recommendation result sequence.
From the above description, the personalized recommendation system of the invention is based on the opposite attribute knowledge base, so that recommendation results inconsistent with user attributes are greatly eliminated, the requirements of personalized recommendation of users are met, the recommendation accuracy is improved, the adoption rate of the recommendation results by the users is improved, and the value of the recommendation system to the users is improved.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (8)
1. A personalized recommendation method is characterized by comprising the following steps:
acquiring a recommendation result sequence recommended to a user by a current recommendation system;
acquiring a plurality of preset recommendation results in a preset direction from the recommendation result sequence as a primary recommendation result sequence, wherein the number of the preset recommendations is less than or equal to the total number of the recommendation results in the recommendation result sequence;
inquiring whether to store the opposite attribute of the user in an opposite attribute table of the user pre-stored in an opposite attribute knowledge base according to the identity information of the user; the step of obtaining the opposite attribute of the user comprises the following steps: when the antisense words of the keywords in the attributes of the user can be inquired in a database, taking the antisense words of the keywords as the opposite attributes of the user; when the antisense words of the keywords in the attributes of the user are not inquired in the database, the keywords of the same type which are farthest from the keywords in the database are used as the opposite attributes of the user; the distance refers to the difference;
when the query result is yes, matching each recommendation result in the primary recommendation result sequence with the opposite attribute of the user respectively;
when the query result is negative, judging whether the user is a registered user of the current recommendation system; when the judgment result is yes, acquiring the attribute of the user from the registration information of the user of the current recommendation system, obtaining the opposite attribute of the user according to the attribute of the user, and storing the opposite attribute of the user in the opposite attribute knowledge base; when the judgment result is negative, generating an information acquisition window, acquiring the attributes of the user, obtaining the opposite attributes of the user according to the attributes of the user, and storing the opposite attributes of the user in the opposite attribute knowledge base;
deleting the recommendation result of which the matching result of the opposite attributes of the user in the primary recommendation result sequence meets the preset condition; the preset condition is that the matching degree of the recommendation result and the opposite attribute of the user is greater than a preset matching degree value;
obtaining a final recommendation result sequence according to the rest recommendation results of the primary recommendation result sequence;
and outputting the final recommendation result sequence.
2. The personalized recommendation method according to claim 1, wherein the identity information of the user comprises a user ID, the user opposite attribute table comprises a user field and a user opposite attribute field, the user field stores the user ID, the user opposite attribute field stores the user opposite attribute, the user opposite attribute is obtained according to the user attribute, and the user attribute comprises any one or any combination of age, gender, occupation, academic history, specialty, hobby and geographic location of the user.
3. The personalized recommendation method according to claim 1, wherein the step of respectively matching each recommendation result in the primary recommendation result sequence with an opposite attribute of the user comprises:
converting each recommendation result in the primary recommendation result sequence and the opposite attributes of the user into character strings respectively;
and respectively calculating the matching degree of the character string converted by each recommendation result in the primary recommendation result sequence and the character string converted by the opposite attribute of the user.
4. The personalized recommendation method according to claim 3, wherein the step of deleting the recommendation result of which the matching result with the opposite attribute of the user in the primary recommendation result sequence meets a preset condition comprises:
respectively acquiring the same number of characters of the character string converted by each recommendation result in the primary recommendation result sequence and the character string converted by the opposite attribute of the user;
and deleting the recommendation result of which the number of the same characters of the character string converted from the opposite attribute of the user in the initial recommendation result sequence is greater than a preset value.
5. A personalized recommendation system, comprising:
the recommendation result sequence acquisition module is used for acquiring a recommendation result sequence recommended to a user by the current recommendation system;
the primary recommendation result sequence acquisition module is used for acquiring a plurality of preset recommendation results in a preset direction from the recommendation result sequence as a primary recommendation result sequence, wherein the number of the preset recommendations is less than or equal to the total number of the recommendation results in the recommendation result sequence;
the attribute query module is used for querying whether the opposite attribute of the user is stored in a user opposite attribute table pre-stored in an opposite attribute knowledge base according to the identity information of the user; the step of obtaining the opposite attribute of the user comprises the following steps: when the antisense words of the keywords in the attributes of the user can be inquired in a database, taking the antisense words of the keywords as the opposite attributes of the user; when the antisense words of the keywords in the attributes of the user are not inquired in the database, the keywords of the same type which are farthest from the keywords in the database are used as the opposite attributes of the user;
the result matching module is used for respectively matching each recommendation result in the primary recommendation result sequence with the opposite attribute of the user when the query result is yes;
the attribute acquisition module is used for judging whether the user is a registered user of the current recommendation system or not when the query result is negative; when the judgment result is yes, acquiring the attribute of the user from the registration information of the user of the current recommendation system, obtaining the opposite attribute of the user according to the attribute of the user, and storing the opposite attribute of the user in the opposite attribute knowledge base; when the judgment result is negative, generating an information acquisition window, acquiring the attributes of the user, obtaining the opposite attributes of the user according to the attributes of the user, and storing the opposite attributes of the user in the opposite attribute knowledge base;
the result deleting module is used for deleting the recommendation result of which the matching result of the opposite attributes of the user in the initial recommendation result sequence meets the preset condition; the preset condition is that the matching degree of the recommendation result and the opposite attribute of the user is greater than a preset matching degree value;
a final recommendation result sequence obtaining module, configured to obtain a final recommendation result sequence according to the remaining recommendation results of the primary recommendation result sequence;
and the sequence output module is used for outputting the final recommendation result sequence.
6. The personalized recommendation system according to claim 5, wherein the identity information of the user comprises a user ID, the user opposite attribute table comprises a user field and a user opposite attribute field, the user field stores the user ID, the user opposite attribute field stores the user opposite attribute, the user opposite attribute is obtained according to the user attribute, and the user attribute comprises any one or any combination of age, gender, occupation, academic history, specialty, speciality, hobby and geographic location of the user.
7. The personalized recommendation system of claim 5, wherein the result matching module comprises:
the conversion unit is used for respectively converting each recommendation result in the primary recommendation result sequence and the opposite attributes of the user into character strings;
and the matching unit is used for respectively calculating the matching degree of the character string converted by each recommendation result in the primary recommendation result sequence and the character string converted by the opposite attribute of the user.
8. The personalized recommendation system of claim 7, wherein the result deletion module comprises:
the obtaining unit is used for respectively obtaining the same number of characters of the character string converted by each recommendation result in the primary recommendation result sequence and the character string converted by the opposite attribute of the user;
and the deleting unit is used for deleting the recommendation result of which the number of the same characters of the character string converted from the opposite attribute of the user in the primary recommendation result sequence is greater than a preset value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511034894.7A CN105701680B (en) | 2015-12-31 | 2015-12-31 | Personalized recommendation method and system based on opposite attribute knowledge base |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511034894.7A CN105701680B (en) | 2015-12-31 | 2015-12-31 | Personalized recommendation method and system based on opposite attribute knowledge base |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105701680A CN105701680A (en) | 2016-06-22 |
| CN105701680B true CN105701680B (en) | 2020-08-25 |
Family
ID=56226854
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511034894.7A Active CN105701680B (en) | 2015-12-31 | 2015-12-31 | Personalized recommendation method and system based on opposite attribute knowledge base |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105701680B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111151858B (en) * | 2020-01-13 | 2021-10-15 | 吉利汽车研究院（宁波）有限公司 | Spot welding parameter application system and setting method |
| CN112269928A (en) * | 2020-10-23 | 2021-01-26 | 百度在线网络技术（北京）有限公司 | User recommendation method and device, electronic equipment and computer readable medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101329674A (en) * | 2007-06-18 | 2008-12-24 | 北京搜狗科技发展有限公司 | System and method for providing personalized searching |
| US20130091164A1 (en) * | 2011-10-11 | 2013-04-11 | Microsoft Corporation | Recommending data based on user and data attributes |
-
2015
- 2015-12-31 CN CN201511034894.7A patent/CN105701680B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101329674A (en) * | 2007-06-18 | 2008-12-24 | 北京搜狗科技发展有限公司 | System and method for providing personalized searching |
| US20130091164A1 (en) * | 2011-10-11 | 2013-04-11 | Microsoft Corporation | Recommending data based on user and data attributes |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105701680A (en) | 2016-06-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220121701A1 (en) | Retrieval support system, retrieval support method and retrieval support program | |
| Vaccaro et al. | The elements of fashion style | |
| KR101649322B1 (en) | Systems and methods for image-to-text and text-to-image association | |
| CN105678575B (en) | Personalized recommendation method and system based on user attribute knowledge base | |
| CN105701680B (en) | Personalized recommendation method and system based on opposite attribute knowledge base | |
| US20090281922A1 (en) | Method and system for selecting clothing items according to predetermined criteria | |
| KR102295459B1 (en) | A method of providing a fashion item recommendation service to a user using a date | |
| JP2014229129A (en) | Combination presentation system and computer program | |
| CN105955451A (en) | Travel prompt method based on user image | |
| CN105335386B (en) | A kind of method and device that navigation tag is provided | |
| US20130097053A1 (en) | Method and system for recommending a combined service by taking into account situation information on a target user and the degree of complementarity of a service | |
| US10346901B1 (en) | Computer generation of composite images | |
| JP2019197460A (en) | Cluster generator, cluster generation method, and cluster generation program | |
| CN105677817B (en) | Personalized recommendation method and system based on positive and negative attribute knowledge base | |
| CN105512341B (en) | Personalized recommendation method and system based on big data search | |
| KR20210131198A (en) | Method, apparatus and computer program for advertising recommended product | |
| WO2020141802A2 (en) | Method for providing fashion item recommendation service to user by using date | |
| JP2020064496A (en) | Information processing system and program | |
| KR102271090B1 (en) | A method of providing a fashion item recommendation service to a user using calendar data | |
| CN112750004A (en) | Cross-domain commodity cold start recommendation method and device and electronic equipment | |
| US20210334877A1 (en) | System and method for fashion style recommendation | |
| CN112036521B (en) | Information consistency detection method, device, equipment and storage medium | |
| KR20230000687A (en) | Customized fashion information provision system using online clothing related search information | |
| KR20200085106A (en) | A method of providing a fashion item recommendation service to a user | |
| CN112149007A (en) | Method, device and equipment for clothing recommendation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |