CN108022150B

CN108022150B - Recommendation method and system based on O2O data

Info

Publication number: CN108022150B
Application number: CN201711236424.8A
Authority: CN
Inventors: 刘津防
Original assignee: Jinse Jiayuan Network Technology Co ltd
Current assignee: Jinse Jiayuan Network Technology Co ltd
Priority date: 2017-11-29
Filing date: 2017-11-29
Publication date: 2021-07-06
Anticipated expiration: 2037-11-29
Also published as: CN108022150A

Abstract

The invention provides a recommendation method and a recommendation system based on O2O data, which comprise the following steps: collecting service data information, and updating the service data information within a preset time; counting consumption data information, wherein the consumption data information comprises the number of users consumed by each merchant, the number of users consumed by the same user in different merchants and the consumption times of each user in different merchants; calculating the number of users consumed by each merchant and the number of users consumed by the same user in different merchants to obtain the similarity of the surplus numbers; obtaining the interest degree of each user according to the consumption times of each user in different merchants; according to the interest degree and the distance weight model of the user, the preference degree of the user is obtained, more accurate recommendation service can be provided, and user experience is improved.

Description

Recommendation method and system based on O2O data

Technical Field

The invention relates to the technical field of information processing, in particular to a recommendation method and a recommendation system based on O2O data.

Background

At present, a conventional recommendation system generally obtains a user preference degree according to online user transaction data based on online user transaction data, so as to perform recommendation according to the user preference degree, and this recommendation manner generally only considers online user transaction data, but does not consider offline factors, such as information of location and distance of offline service stores, for example, if a recommended service is too far away from a user, the user may give up consumption, thereby resulting in poor user experience.

Disclosure of Invention

In view of this, the present invention provides a recommendation method and system based on O2O data, which can provide more accurate recommendation service and improve user experience.

In a first aspect, an embodiment of the present invention provides a recommendation method based on O2O data, where the method includes:

acquiring service data information, and updating the service data information within a preset time;

counting consumption data information, wherein the consumption data information comprises the number of users consumed by each merchant, the number of users consumed by the same user in different merchants and the consumption times of each user in different merchants;

calculating the number of users consumed by each merchant and the number of users consumed by the same user in different merchants to obtain the similarity of the residual sound;

obtaining the interestingness of each user according to the consumption times of each user in different merchants;

and obtaining the preference of the user according to the interest degree of the user and the weight model of the distance.

With reference to the first aspect, an embodiment of the present invention provides a first possible implementation manner of the first aspect, where the calculating the number of users consumed by each merchant and the number of users consumed by the same user in different merchants to obtain the residual sea similarity includes:

calculating the number of the users consumed by each merchant and the number of the users consumed by the same user in different merchants to obtain a first matrix and a second matrix;

and obtaining the cosine similarity according to the first matrix and the second matrix.

With reference to the first possible implementation manner of the first aspect, an embodiment of the present invention provides a second possible implementation manner of the first aspect, where the obtaining the cosine similarity according to the first matrix and the second matrix includes:

calculating the cosine similarity according to the following formula:

wherein, W_uvFor the cosine similarity, n (u) n (v) is the first matrix, and | n (u) | × | n (v) | is the second matrix.

With reference to the first aspect, an embodiment of the present invention provides a third possible implementation manner of the first aspect, wherein the obtaining of the interestingness of each user according to the number of times of consumption of the user in the different merchants includes:

calculating the interestingness of the user according to the following formula:

wherein, P_ujFor the userIs a set of items liked by the user, S (j, k) is a set of k items similar to item j, W_ijIs the cosine similarity of items j and i, and r is the interest of the user u in item i.

With reference to the first aspect, an embodiment of the present invention provides a fourth possible implementation manner of the first aspect, where the service data information includes user data information, user consumption data information, and article data information, and the method further includes:

judging whether the user data information or the article data information is updated or not;

and if the user data information or the article data information is updated, updating or deleting the user data information or the article data information.

In a second aspect, an embodiment of the present invention further provides a recommendation system based on O2O data, where the system includes:

the first updating unit is used for collecting service data information and updating the service data information within preset time;

the statistical unit is used for counting consumption data information, wherein the consumption data information comprises the number of users consumed by each merchant, the number of users consumed by the same user in different merchants and the consumption times of each user in different merchants;

the calculating unit is used for calculating the number of the users consumed by each merchant and the number of the users consumed by the same user in different merchants to obtain the similarity of the surplus numbers;

the interestingness obtaining unit is used for obtaining the interestingness of each user according to the consumption times of each user in different merchants;

and the preference degree unit is used for obtaining the preference degree of the user according to the interest degree of the user and the weight model of the distance.

With reference to the second aspect, an embodiment of the present invention provides a first possible implementation manner of the second aspect, where the computing unit includes:

the matrix acquisition module is used for calculating the number of the users consumed by each merchant and the number of the users consumed by the same user in different merchants to obtain a first matrix and a second matrix;

and the cosine similarity obtaining module is used for obtaining the cosine similarity according to the first matrix and the second matrix.

With reference to the first possible implementation manner of the second aspect, an embodiment of the present invention provides a second possible implementation manner of the second aspect, where the cosine similarity obtaining module includes:

calculating the cosine similarity according to the following formula:

With reference to the second aspect, an embodiment of the present invention provides a third possible implementation manner of the second aspect, where the interestingness obtaining unit includes:

calculating the interestingness of the user according to the following formula:

wherein, P_ujFor the interest of the user, N (u) is a set of items liked by the user, S (j, k) is a set of k items similar to item j, W_ijIs the cosine similarity of items j and i, and r is the interest of the user u in item i.

With reference to the second aspect, an embodiment of the present invention provides a fourth possible implementation manner of the second aspect, where the system further includes:

a judging unit, configured to judge whether the user data information or the article data information is updated;

and the second updating unit is used for updating or deleting the user data information or the article data information if the user data information or the article data information is updated.

The embodiment of the invention provides a recommendation method and a recommendation system based on O2O data, which comprises the following steps: collecting service data information, and updating the service data information within a preset time; counting consumption data information, wherein the consumption data information comprises the number of users consumed by each merchant, the number of users consumed by the same user in different merchants and the consumption times of each user in different merchants; calculating the number of users consumed by each merchant and the number of users consumed by the same user in different merchants to obtain the similarity of the surplus numbers; obtaining the interest degree of each user according to the consumption times of each user in different merchants; according to the interest degree and the distance weight model of the user, the preference degree of the user is obtained, more accurate recommendation service can be provided, and user experience is improved.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.

FIG. 1 is a flowchart of a recommendation method based on O2O data according to an embodiment of the present invention;

fig. 2 is a flowchart of step S103 in the recommendation method based on O2O data according to an embodiment of the present invention;

FIG. 3 is a flowchart of an API provided by the second embodiment of the present invention;

fig. 4 is a schematic diagram of a recommendation system based on O2O data according to a third embodiment of the present invention.

Icon:

10-a first update unit; 20-a statistical unit; 30-a calculation unit; 40-an interestingness obtaining unit; 50-preference unit.

Detailed Description

To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

For the understanding of the present embodiment, the following detailed description will be given of the embodiment of the present invention.

The first embodiment is as follows:

fig. 1 is a flowchart of a recommendation method based on O2O data according to an embodiment of the present invention.

Referring to fig. 1, the method includes the steps of:

step S101, collecting service data information, and updating the service data information within a preset time;

here, the service data information includes user data information, user consumption data information, and article data information; the user data information, the user consumption data information, and the article data information specifically include the following information. In the process of acquiring the service data information, it is necessary to determine whether the service data information is accessed for the first time or not.

If the access is the first access, all information of class C users, all information of class C consumption data, coupons and all information of affiliated merchants are acquired in batch by ETL (or a timed task JOB), and the data are extracted and enter a data warehouse (the data can be extracted according to the updating time);

if the access is not the first access, all newly registered information of the yesterday C class user, all information of yesterday C class consumption data, yesterday coupon and all information of the affiliated merchant are taken at zero point every day, and extraction is carried out to enter a data warehouse (extraction can be carried out according to update time);

if the class C user information in the service system is updated, the updated data and the updated data warehouse can be obtained at zero point every day;

and if the service system C type coupon and the information of the affiliated merchant are updated and deleted, the data warehouse is updated or deleted at zero point every day, so that the data of the data warehouse are real and effective.

Step S102, counting consumption data information, wherein the consumption data information comprises the number of users consumed by each merchant, the number of users consumed by the same user in different merchants and the consumption times of each user in different merchants;

step S103, calculating the number of users consumed by each merchant and the number of users consumed by the same user in different merchants to obtain the similarity of the surplus numbers;

step S104, obtaining the interest degree of each user according to the consumption times of each user in different merchants;

and step S105, obtaining the preference of the user according to the interest degree of the user and the weight model of the distance.

Compared with the traditional recommendation algorithm RNN algorithm, the method can reflect the continuous change process of the interests and hobbies of the user at a certain stage, and is high in accuracy. The recommendation algorithm provided by the embodiment of the invention is combined with the traditional coordination filtering algorithm and the RNN neural algorithm, and provides more accurate recommendation service for O2O application scenes.

Wherein, the application scenarios of O2O are specifically: with the popularization of mobile internet, at present, more and more users begin to use mobile payment, and the mobile payment platform is mobile for merchants, so that not only is the payment collection convenient, but also more marketing services can be provided, for example, coupons are issued through merchant WeChat public numbers and merchant applets to attract surrounding customers to pick up, when the customers use the mobile payment, the picked coupons can be directly used, and meanwhile, after the payment is completed, the payment platform recommends the coupons to the customers according to the previous transaction records of the users.

Further, referring to fig. 2, step S103 includes the steps of:

step S201, calculating the number of users consumed by each merchant and the number of users consumed by the same user in different merchants to obtain a first matrix and a second matrix;

here, the number of users consumed by the same user at different merchants may be the number of users consumed by the same user for counting two different merchants. Specifically, reference is made to tables 1 and 2:

TABLE 1

	Merchant 1	Merchant 2	Merchant 3	Merchant 4	Merchant 5
						C user	5	4	4	2	0

TABLE 2

	Merchant 1	Merchant 2	Merchant 3	Merchant 4	Merchant 5
						Merchant 1	0	3	3	1	0
Merchant 2	3	0	3	2	0
						Merchant 3	2	3	0	4	0
Merchant 4	2	2	4	0	0
						Merchant 5	0	0	1	1	0

In addition, the consumption times of each user at different merchants can be used for expressing the preference of the C user to different merchants. With specific reference to table 3:

TABLE 3

C user	Business company	Number of consumption
			A	Merchant 1	2
A	Merchant 2	1
			A	Merchant 3	1
A	Merchant 4	0
			A	Merchant 5	0
B	Merchant 1	1
			B	Merchant 2	1
B	Merchant 3	0
			B	Merchant 4	0
B	Merchant 5	0

In step S202, a cosine similarity is obtained according to the first matrix and the second matrix.

Further, step S202 includes:

calculating the similarity of the cosine and the sine according to the formula (1):

wherein, W_uvFor the cosine similarity, n (u) n (v) is the first matrix, and | n (u) | × | n (v) | is the second matrix. Specifically, see table 4:

TABLE 4

	Merchant 1	Merchant 2	Merchant 3	Merchant 4	Merchant 5
						Merchant 1		0.47	0.63	0.71
Merchant 2	0.47		0.77	0.57
						Merchant 3	0.63	0.77		0.89	0.45
Merchant 4	0.71	0.57	0.89		0.5
						Merchant 5			0.45	0.5

Further, step S104 includes:

calculating the interestingness of the user according to the formula (2):

Here, the interest degree of the user U in the item j is the similarity degree of K items most similar to the item j and the preference degree of the user U in the K items.

For example, the merchants that the user has consumed are merchant 1 and merchant 2, and the outstanding merchants are merchant 3, merchant 4, and merchant 5.

Then, the interestingness of each merchant is:

merchant 3 has an interest of 1.47 + 1.0.63-1.1; merchant 4's interestingness 1 × 0.57+1 × 0.89 ═ 1.46; the merchant 5 has an interest level of 1 × 0+1 × 0.45 — 0.45. With specific reference to table 5:

TABLE 5

C user	Business company	Degree of interest
			B	Merchant 3	1.1
B	Merchant 4	1.46
			B	Merchant 5	0.45

In addition, the RNN is based on the location information of the user's merchant, specifically: obtaining a weight model S of the interest degree and the distance of the user based on the RNN model training sequence data, predicting the preference degree of the user to a new merchant based on the weight model S, and specifically referring to a table 6:

TABLE 6

Further, the service data information includes user data information, user consumption data information and article data information, and the method further includes:

Here, the service data information is updated every day within a preset time, which may be 1 point every morning, and yesterday's C-consumption data (which may include fields, i.e., user ID and merchant ID) for successful payment is taken, so that merchants who are closed and have all coupon statuses disabled are excluded. Specifically, tables 7 to 9 can be referred to:

TABLE 7

Application id	WeChat user id
		1	A
1	B
		1	C
1	D
		1	E

TABLE 8

Application id	Affiliated merchant	Longitude (G)	Dimension (d) of
				1	Merchant 1	xxx	xxx
1	Merchant 2	xxx	xxx
				1	Merchant 3	xxx	xxx
1	Merchant 4	xxx	xxx
				1	Merchant 5	xxx	xxx

TABLE 9

Application id	C user	Business company	Date
				1	A	Merchant 1
1	A	Merchant 1
				1	A	Merchant 2
1	A	Merchant 3
				1	B	Merchant 1
1	B	Merchant 2
				1	C	Merchant 2
1	C	Merchant 3
				1	C	Merchant 4
1	D	Merchant 4
				1	D	Merchant 3
1	D	Merchant 1
				1	E	Merchant 1
1	E	Merchant 2
				1	E	Merchant 3

Example two:

fig. 3 is a flowchart of an API provided in the second embodiment of the present invention.

Referring to fig. 3, a recommendation result table API (Application Programming Interface) needs to distinguish between new and old users, and if the new and old users are old users and have a past consumption record, there may be a recommendation result; if it is a new user and there is no record of consumption, recommendations can only be made according to rules (location based).

The method comprises the following steps:

step S201, an API requests an entrance, and input parameters are a user ID, the longitude and latitude of a merchant, the recommended quantity C and the province and city of the merchant;

step S202, inquiring a city where the user is located and a list of all merchants to be recommended;

step S203, judging whether the number of merchants in the merchant list is greater than 0, if so, taking nth page data from the paging, and checking available coupons in all merchants of nth page to be recommended;

step S204, recommending available coupon lists;

step S205, judging whether the number of the lists is not less than the recommended number C, if so, performing a final coupon list; if not, judging whether the current page number is equal to the total page number or not;

step S206, if yes, calculating a distance list based on the distance between the user and all merchants;

step S207, taking nth page data from the paging, and checking available coupons in all merchants of nth page to be recommended;

step S208, recommending available coupon lists and the number of the lists, and thus recommending a final coupon list;

step S209, if not, taking the (n + 1) th page data, and executing step S203;

step S210, if the number of merchants in the merchant list is not more than 0, judging whether the longitude and latitude is null;

step S211, if the number is empty, C coupons of the hottest merchants in the city are taken, and C coupons to be recommended are verified;

step S212, judging whether the number of the lists is larger than or equal to the number C of the coupons, and if so, taking C finally recommended coupon lists; if not, go to step S211;

step S213, if the latitude and longitude are not null, calculating a distance list based on the distance between the user and all merchants;

step S214, fetching the nth page data from the paging;

step S215, checking available coupons in all merchants of the nth page to be recommended;

step S216, recommending available coupon lists;

step S217, judging whether the number of the lists is larger than or equal to the recommended number C, if so, taking the front C as a final recommended coupon list; if not, the (n + 1) th page data is taken, and step S214 is executed.

Example three:

Referring to fig. 4, the system includes:

referring to fig. 4, the system includes a first updating unit 10, a counting unit 20, a calculating unit 30, an interestingness obtaining unit 40, and a preference unit 50.

The first updating unit 10 is configured to collect service data information and update the service data information within a preset time;

the statistical unit 20 is configured to count consumption data information, where the consumption data information includes the number of users consumed by each merchant, the number of users consumed by the same user in different merchants, and the number of times of consumption of each user in different merchants;

a calculating unit 30, configured to calculate the number of users consumed by each merchant and the number of users consumed by the same user in different merchants, so as to obtain a residual similarity;

the interestingness obtaining unit 40 is configured to obtain the interestingness of each user according to the consumption times of each user in different merchants;

and the preference degree unit 50 is configured to obtain the preference degree of the user according to the interest degree of the user and the weight model of the distance.

Further, the calculation unit 30 includes:

a matrix obtaining module (not shown) configured to calculate the number of users consumed by each merchant and the number of users consumed by the same user in different merchants to obtain a first matrix and a second matrix;

an index similarity obtaining module (not shown) configured to obtain the index similarity according to the first matrix and the second matrix.

Further, the cosine similarity obtaining module (not shown) includes:

and (4) calculating the similarity of the cosine and the sine according to the formula (1).

Further, the interestingness obtaining unit 40 includes:

calculating the interestingness of the user according to formula (2):

further, the system further comprises:

a judging unit (not shown) for judging whether the user data information or the article data information is updated;

a second updating unit (not shown) for updating or deleting the user data information or the article data information if the user data information or the article data information is updated.

The embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and when the processor executes the computer program, the steps of the recommendation method based on O2O data provided in the foregoing embodiment are implemented.

The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the recommendation method based on O2O data of the above embodiment are executed.

The computer program product provided in the embodiment of the present invention includes a computer-readable storage medium storing a program code, where instructions included in the program code may be used to execute the method described in the foregoing method embodiment, and specific implementation may refer to the method embodiment, which is not described herein again.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In addition, in the description of the embodiments of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A recommendation method based on O2O data, characterized in that the method comprises:

obtaining the preference of the user according to the interest degree of the user and a weight model of the distance;

the obtaining the interestingness of the user according to the consumption times of each user in different merchants comprises:

calculating the interestingness of the user according to the following formula:

2. The O2O data-based recommendation method according to claim 1, wherein the calculating the number of users consumed by each merchant and the number of users consumed by the same user at different merchants to obtain the inter-edge similarity comprises:

3. The O2O data-based recommendation method according to claim 2, wherein the deriving the cosine similarity according to the first matrix and the second matrix comprises:

calculating the cosine similarity according to the following formula:

4. The O2O data-based recommendation method according to claim 1, wherein the business data information includes user data information, user consumption data information and item data information, the method further comprising:

5. A recommendation system based on O2O data, the system comprising:

the preference degree unit is used for obtaining the preference degree of the user according to the interest degree of the user and the weight model of the distance;

the interestingness acquisition unit includes:

calculating the interestingness of the user according to the following formula:

6. The O2O data-based recommendation system according to claim 5, wherein the computing unit comprises:

7. The O2O data-based recommendation system according to claim 6, wherein the cosine similarity obtaining module comprises:

calculating the cosine similarity according to the following formula:

8. The O2O data-based recommendation system according to claim 5, further comprising: