CN111506816A - Recommendation method, device, equipment and storage medium - Google Patents
Recommendation method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111506816A CN111506816A CN202010311448.0A CN202010311448A CN111506816A CN 111506816 A CN111506816 A CN 111506816A CN 202010311448 A CN202010311448 A CN 202010311448A CN 111506816 A CN111506816 A CN 111506816A
- Authority
- CN
- China
- Prior art keywords
- item
- recommended
- item set
- historical
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Recommending goods or services
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a recommendation method, a recommendation device, recommendation equipment and a storage medium, wherein the method comprises the following steps: determining a historical favorite item set of a target user according to the historical behaviors of the target user, determining a first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item, determining a second pre-recommended item set corresponding to the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item, and determining an intersection item of the first pre-recommended item set and the second pre-recommended item set as a target recommended item recommended to the target user. On one hand, compared with the method of only adopting first-order information recommendation at present, the recommendation precision of the embodiment is higher, and on the other hand, the recommendation method can improve the recommendation diversity on the premise of ensuring the recommendation precision.
Description
Technical Field
The embodiment of the invention relates to the technical field of big data processing, in particular to a recommendation method, a recommendation device, recommendation equipment and a storage medium.
Background
With the continuous development of big data processing technology, recommendation systems for recommending articles to users are more and more widely applied. The items can be multimedia items such as movies, music, albums, and the like.
Currently, recommendation systems often use collaborative filtering methods to make recommendations: and calculating related items of the items, and recommending the items most related to the favorite items of the user through the favorite items of the user.
However, the recommendation method only uses the information about the item that is the favorite item of the user, and the information is considered to be single, so that the recommendation precision is low.
Disclosure of Invention
The invention provides a recommendation method, a recommendation device, recommendation equipment and a storage medium, and aims to solve the technical problem of low recommendation precision in the conventional recommendation method.
In a first aspect, an embodiment of the present invention provides a recommendation method, including:
determining a historical favorite item set of a target user according to the historical behavior of the target user;
determining a first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item;
determining a second pre-recommended item set corresponding to the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item;
and determining the intersection article of the first pre-recommended article set and the second pre-recommended article set as the target recommended article recommended to the target user.
In a second aspect, an embodiment of the present invention provides a recommendation apparatus, including:
the first determination module is used for determining a historical favorite item set of a target user according to the historical behavior of the target user;
the second determining module is used for determining a first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item;
a third determining module, configured to determine, according to the first pre-recommended item set and the related item set of each item, a second pre-recommended item set corresponding to the first pre-recommended item set;
a fourth determining module, configured to determine an intersection item of the first pre-recommended item set and the second pre-recommended item set as a target recommended item recommended to the target user.
In a third aspect, an embodiment of the present invention further provides a computer device, where the computer device includes:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the recommendation method as provided in the first aspect.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the recommendation method as provided in the first aspect.
The embodiment provides a recommendation method, a recommendation device, recommendation equipment and a storage medium, wherein the method comprises the following steps: determining a historical favorite item set of a target user according to the historical behaviors of the target user, determining a first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item, determining a second pre-recommended item set corresponding to the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item, and determining an intersection item of the first pre-recommended item set and the second pre-recommended item set as a target recommended item recommended to the target user. It has the following technical effects: when a target user is recommended, a first pre-recommended item set corresponding to a historical favorite item set of the user and a second pre-recommended item set corresponding to the first pre-recommended item set are combined to determine a target recommended item, in other words, first-order information and second-order information of the user are combined to recommend the target recommended item.
Drawings
Fig. 1 is a schematic flow chart of a recommendation method according to an embodiment of the present invention;
FIG. 2A is a flowchart illustrating a process for determining a first pre-recommended set of items corresponding to a historical favorite set of items;
FIG. 2B is a flowchart illustrating a process of determining a second pre-recommended set of items corresponding to the first pre-recommended set of items;
FIG. 3 is a schematic structural diagram of a recommendation device according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Fig. 1 is a schematic flowchart of a recommendation method according to an embodiment of the present invention. The method and the device are suitable for determining the target recommended item recommended to the target user according to the historical behaviors of the target user. The present embodiment may be performed by a recommendation device, which may be implemented in software and/or hardware, which may be integrated in a computer device. As shown in fig. 1, the recommendation method provided in this embodiment includes the following steps:
step 101: and determining the historical favorite item set of the target user according to the historical behaviors of the target user.
Specifically, the computer device in this embodiment may be a server or a terminal device. The terminal device may be a handheld device, a vehicle-mounted device, a wearable device, various User Equipment (UE), a Mobile Station (MS), a terminal (terminal), and the like. Illustratively, the terminal device of the embodiment of the invention may be a smart phone, a tablet computer, a smart television, or the like. The items in this embodiment may be multimedia items such as music, albums, movies, pictures, articles, and the like, or may be physical items that can be purchased in a shopping website. This embodiment is not limited thereto. The target user in this embodiment refers to a user to whom an item needs to be recommended.
Optionally, when the item is a multimedia item, the historical behavior of the user is play behavior. When the item is a physical item which can be purchased in a shopping website, the historical behavior of the user is browsing behavior or purchasing behavior.
In one implementation, the item that the target user played, browsed or purchased within the preset time period may be determined as a history favorite item of the target user.
In another implementation manner, an item whose playing times, browsing times or purchasing times of the target user are greater than a preset time threshold within a preset time period may be determined as a historical favorite item of the target user.
In another implementation, the articles whose playing time or browsing time of the target user is greater than the preset time threshold within the preset time period may be determined as the historical favorite articles of the target user.
The set of historical liked items of the target user is a set of historical liked items of the target user.
Step 102: and determining a first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item.
Specifically, in this embodiment, the related item set of each item may be predetermined. The related item set of an item in this embodiment refers to a set of items related to the item. An item associated with the item refers to an item that has some commonality with the item. The commonalities here may be the same or similar for the type and style of the article, etc.
In one implementation, the recommending apparatus may determine the related item set of the first item according to the correlation coefficient between each other item except the first item and the first item in the database. The quantity of the related items in the related item set is a first preset quantity, and the first item is any item in the database.
For convenience of description, in this embodiment, any item in the database is referred to as a first item. When calculating the related item set of the first item, the correlation coefficient between each other item except the first item in the database and the first item may be calculated, and then the related item set of the first item may be determined according to the correlation coefficient. The specific process can be as follows: calculating the Jacard coefficient or cosine similarity of each other article and the first article, and taking the Jacard coefficient or cosine similarity as a correlation coefficient; and arranging the correlation coefficients according to the sequence from large to small, and determining the set of the items with the corresponding correlation coefficients arranged in the front by the first preset number as the related item set of the first item. Illustratively, the first preset number here may be 30.
The above-mentioned process is executed to obtain the related item set of each first item in the database, i.e. the related item set of each item. The database in this embodiment includes all items.
More specifically, when the article is a multimedia object, the jkcard coefficients of the second article and the first article may be determined according to the number of users playing the first article and the number of users playing the second article among other articles. Wherein the second article is any one of the other articles.
Illustratively, according to a formula
Determining the Jacard coefficients of the second article and the first article. Wherein, wijRepresenting the vicard coefficient of the second item and the first item, n (j) representing the number of users playing the second item, | n (i) ∩ n (j) | representing the number of users playing the first item and the second item simultaneously, | n (i) ∪ n (j) | representing the number of users playing the first item or the second item, α being a first preset parameter, β being a second preset parameter.
Optionally, the value of α can be between 0.1 and 0.3, the α can be used for inhibiting hot goods, the value of β can be between 5 and 100, and the β can be used for inhibiting cold goods.
Illustratively, it can also be according to a formula
Determining the Jacard coefficients of the second article and the first article.
Fig. 2A is a flowchart illustrating a process of determining a first pre-recommended set of items corresponding to a historical favorite set of items. As shown in fig. 2A, one possible implementation of step 102 is to include the following steps:
step 1021: and determining a first related item set of each historical favorite item and a correlation coefficient between each first related item in the first related item set and the historical favorite item in the historical favorite item set according to the historical favorite item set and the related item set of each item.
As indicated previously, the associated set of items for each item has been predetermined. After the historical favorite item sets of the target user are determined, a first related item set of each historical favorite item in the historical favorite item sets can be determined. The target user's history likes how many items there are, and thus how many first related item sets can be determined.
For each first set of related items, each first related item in the first set of related items has a correlation coefficient with the historical liked item.
Step 1022: and adding the correlation coefficients corresponding to the same first related items in the first related item sets of all the historical favorite items, keeping the correlation coefficients corresponding to different first related items unchanged, and determining the updated correlation coefficient of each first related item and the historical favorite items.
Since there may be the same first related item in the plurality of first related item sets, in step 1022, the correlation coefficients corresponding to the same first related item in all the first related item sets are added, and the correlation coefficients corresponding to different first related items remain unchanged, so that the updated correlation coefficient of each first related item and the historical favorite item can be determined.
Step 1023: and arranging the updated correlation coefficients of all the first related articles and the historical favorite articles according to the descending order, arranging the corresponding updated correlation coefficients in a second preset number of first related article sets, and determining the first related article sets as the first pre-recommended article sets corresponding to the historical favorite article sets.
The above process is described below as a specific example. Assume that the target user's historical favorite collection is { A, B, C }.
The predetermined related item set of A is { H, I, J, K }, and the correlation coefficient of H and A is 0.1, the correlation coefficient of I and A is 0.2, the correlation coefficient of J and A is 0.8, and the correlation coefficient of K and A is 0.78. The predetermined related item set of B is { M, N, J, K }, and the correlation coefficient of M and B is 0.3, the correlation coefficient of N and B is 0.4, the correlation coefficient of J and B is 0.8, and the correlation coefficient of K and B is 0.7. The predetermined related item set of C is { M, H, S, Y }, and the correlation coefficient of M and C is 0.3, the correlation coefficient of H and C is 0.1, the correlation coefficient of S and C is 0.5, and the correlation coefficient of Y and C is 0.6.
Based on the historical favorite item sets, the determined first related item set of each historical favorite item is respectively as follows: the first related item set of a is { H, I, J, K }, and the correlation coefficient of each first related item with a is: {0.1, 0.2, 0.8, 0.78 }; the first related item set of B is { M, N, J, K }, and the correlation coefficient of each first related item with B is: {0.3, 0.4, 0.8, 0.7 }; c is { M, H, S, Y }, and the correlation coefficient of each first related item with C is: {0.3,0.1,0.5,0.6}.
It can be seen that there are identical first related items H, J, K, M in these three first related item sets. Adding correlation coefficients corresponding to the same first related articles, namely adding correlation coefficient 0.1 of H and A and correlation coefficient 0.1 of H and C to obtain 0.2; adding the correlation coefficient 0.8 of J and A and the correlation coefficient 0.8 of J and B to obtain 1.6; adding the correlation coefficient 0.78 of K and A and the correlation coefficient 0.7 of K and B to obtain 1.48; the correlation coefficient of M to B, 0.3, was added to the correlation coefficient of M to C, 0.3, to yield 0.6. The correlation coefficient of the other first related item I, N, S, Y remains unchanged, and the updated correlation coefficient of each first related item with the historical favorite items can be determined as follows: the updated correlation coefficient of H with the historical favorite is 0.2, the updated correlation coefficient of J with the historical favorite is 1.6, the updated correlation coefficient of K with the historical favorite is 1.48, the updated correlation coefficient of M with the historical favorite is 0.6, the updated correlation coefficient of I with the historical favorite is 0.2, the updated correlation coefficient of N with the historical favorite is 0.4, the updated correlation coefficient of S with the historical favorite is 0.5, and the updated correlation coefficient of Y with the historical favorite is 0.6.
And then, arranging the updated correlation coefficients of all the first related articles and the historical favorite articles in descending order, wherein the arranged order is {1.6, 1.48, 0.6, 0.6, 0.5, 0.4, 0.2, 0.2}, and the corresponding first related articles are { J, K, M, Y, S, N, H, I }. And determining the corresponding updated related coefficients to be a first pre-recommended item set corresponding to the historical favorite item set, wherein the first pre-recommended item set comprises a first related item set { J, K, M, Y, S } with the corresponding updated related coefficients arranged in a first preset number, for example, the first 5 related items.
Step 103: and determining a second pre-recommended item set corresponding to the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item.
Specifically, fig. 2B is a flowchart illustrating a process of determining a second pre-recommended item set corresponding to the first pre-recommended item set. As shown in fig. 2B, in step 103, one possible implementation includes the following steps:
step 1031: and determining a second related item set of each first pre-recommended item and a correlation coefficient between each second related item in the second related item set and the first pre-recommended item in the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item.
After the first pre-recommended item set is determined, a second related item set of each first pre-recommended item in the first pre-recommended item set can be determined. How many first pre-recommended items there are, how many second sets of related items can be determined.
For each second related item set, each second related item in the second related item set has a correlation coefficient with the first pre-recommended item.
Step 1032: and adding the correlation coefficients corresponding to the same second related items in the second related item sets of all the first pre-recommended items, keeping the correlation coefficients corresponding to different second related items unchanged, and determining the updated correlation coefficient of each second item and the first pre-recommended item.
Since there may be second related items that are the same in the plurality of sets of second related items, in step 1032, the correlation coefficients corresponding to the second related items that are the same in all the sets of second related items are added, and the correlation coefficients corresponding to different second related items remain unchanged, so that the updated correlation coefficient of each second related item and the first pre-recommended item can be determined.
Step 1033: and arranging the updated correlation coefficients of all the second related articles and the first pre-recommended articles according to the descending order, arranging the corresponding updated correlation coefficients in a third preset number of second related article sets, and determining the second related article sets as second pre-recommended article sets corresponding to the first pre-recommended article sets.
The following example follows step 102, and the above process is specifically described. Based on step 102, the first pre-recommended set of items is { J, K, M, Y, S }.
For convenience of description, assume that the second related item set of J is { O, P, L, K }, and the correlation coefficient of each second related item with J is {0.1, 0.2, 0.3, 0.4}, respectively, assume that the second related item set of K is { X, P, N, L }, and the correlation coefficient of each second related item with K is {0.3, 0.5, 0.8, 0.78}, assume that the second related item set of M is { O, X, N, K }, and the correlation coefficient of each second related item with M is {0.32, 0.2, 0.8, 0.6}, respectively, assume that the second related item set of Y is { H, I, S, T }, and the correlation coefficient of each second related item with Y is {0.1, 0.3, 0.8, 0.5}, respectively, assume that the second related item set of Y is {0.4, 0.5} and the correlation coefficient of each second related item with Y is {0.1, 0.3, 0.8, 0.5}, respectively, and the correlation coefficient of each second related item set of S, 4, K }, and the correlation coefficient of each second related item is 0.4, 0.5, 0..
In the five second related item sets, the same second related items O, P, L, K, X and N are present, and the correlation coefficients corresponding to the same first related items are added, the correlation coefficients of the other second related items H, I, S, T, J are kept unchanged, and the updated correlation coefficients of each second related item and the first pre-recommended item are determined as follows, wherein the updated correlation coefficients of O, P, L and K, X, N, H, I, S, T, J and the first pre-recommended item are respectively {0.82, 0.9, 1.08, 1.5, 0.5, 1.6, 0.1, 0.3, 0.8, 0.5 and 0.3 }.
Then, the updated correlation coefficients of all the second related items and the first pre-recommended item are arranged in descending order, the arrangement order is {1.6, 1.5, 1.08, 0.9, 0.82, 0.8, 0.5, 0.5, 0.3, 0.3, 0.1}, the corresponding second related items are { N, K, L, P, O, S, T, X, I, J, H }, the corresponding updated correlation coefficients are arranged in the first third preset number, for example, the set of the first 5 second related items { N, K, L, P, O } is determined as the second pre-recommended item set corresponding to the first pre-recommended item set.
Step 104: and determining the intersection article of the first pre-recommended article set and the second pre-recommended article set as the target recommended article recommended to the target user.
Specifically, after the first pre-recommended item set and the second pre-recommended item set are determined, the intersection item in the two sets is determined as the target recommended item recommended to the target user.
With continued reference to steps 102 and 103, for example, if the first pre-recommended set of items is { J, K, M, Y, S }, and the second pre-recommended set of items is { N, K, L, P, O }, and the intersection item is K, then K is determined as the target recommended item recommended to the target user.
Optionally, after the target recommended item is determined, the target recommended item may be recommended to the target user.
In this embodiment, the historical favorite item set of the target user is zeroth order information of the target user. And according to the zeroth-order information of the target user and the related item set, the obtained first pre-recommended item set is the first-order information of the target user. And according to the first pre-recommended item set and the related item set, obtaining second-order information of the second pre-recommended item set as the target user. And finally, intersecting the first-order information and the second-order information of the target user, and preferentially recommending the objects in the intersection.
It should be noted that, in order to avoid recommending an item that the user already likes to the user, if an intersection item of the first pre-recommended item set and the second pre-recommended item set is a history favorite item of the user, the history favorite item is filtered out.
The recommendation method provided by the embodiment comprises the following steps: determining a historical favorite item set of a target user according to the historical behaviors of the target user, determining a first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item, determining a second pre-recommended item set corresponding to the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item, and determining an intersection item of the first pre-recommended item set and the second pre-recommended item set as a target recommended item recommended to the target user. It has the following technical effects: when a target user is recommended, a first pre-recommended item set corresponding to a historical favorite item set of the user and a second pre-recommended item set corresponding to the first pre-recommended item set are combined to determine a target recommended item, in other words, first-order information and second-order information of the user are combined to recommend the target recommended item.
Fig. 3 is a schematic structural diagram of a recommendation device according to an embodiment of the present invention. As shown in fig. 3, the recommendation apparatus provided in this embodiment includes: a first determination module 31, a second determination module 32, a third determination module 33 and a fourth determination module 34.
The first determining module 31 is configured to determine a historical favorite item set of the target user according to the historical behavior of the target user.
Optionally, the apparatus further comprises: and the fifth determining module is used for determining the related item set of the first item according to the correlation coefficient of each other item except the first item in the database and the first item. The quantity of the related items in the related item set is a first preset quantity, and the first item is any item in the database.
In an implementation manner, the fifth determining module is specifically configured to: calculating the Jacard coefficient or cosine similarity of each other article and the first article, and taking the Jacard coefficient or cosine similarity as a correlation coefficient; and arranging the correlation coefficients according to the sequence from large to small, and determining the set of the items with the corresponding correlation coefficients arranged in the front by the first preset number as the related item set of the first item.
More specifically, in terms of calculating the jaccard coefficient of each of the other items and the first item, the fifth determining module specifically includes a determining sub-module, configured to determine the jaccard coefficients of the second item and the first item according to the number of users playing the first item and the number of users playing the second item of the other items. Wherein the second article is any one of the other articles.
Optionally, the determining sub-module is specifically adapted to determine the sub-module according to a formula
Determining the Jacard coefficients of the second article and the first article. Wherein, wijRepresenting the Jacard coefficient of the second item and the first item, N (j) representing the number of users playing the second item, | N (i) ∩ N (j) representing the number of users playing the first item and the second item simultaneously, | N (i) ∪ N (j) survival rateIndicating the number of users playing the first or second item, α is a first preset parameter, and β is a second preset parameter.
And the second determining module 32 is configured to determine, according to the historical favorite item set and a predetermined related item set of each item, a first pre-recommended item set corresponding to the historical favorite item set.
In an implementation, the second determining module 32 is specifically configured to: determining a first related item set of each historical favorite item and a correlation coefficient between each first related item in the first related item set and the historical favorite item in the historical favorite item set according to the historical favorite item set and the related item set of each item; adding the correlation coefficients corresponding to the same first related articles in the first related article set of all historical favorite articles, keeping the correlation coefficients corresponding to different first related articles unchanged, and determining the updated correlation coefficient of each first related article and the historical favorite articles; and arranging the updated correlation coefficients of all the first related articles and the historical favorite articles according to the descending order, arranging the corresponding updated correlation coefficients in a second preset number of first related article sets, and determining the first related article sets as the first pre-recommended article sets corresponding to the historical favorite article sets.
The third determining module 33 is configured to determine, according to the first pre-recommended item set and the related item set of each item, a second pre-recommended item set corresponding to the first pre-recommended item set.
In an implementation manner, the third determining module 33 is specifically configured to: determining a second related item set of each first pre-recommended item and a correlation coefficient between each second related item and the first pre-recommended item in the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item; adding correlation coefficients corresponding to the same second related articles in a second related article set of all the first pre-recommended articles, keeping the correlation coefficients corresponding to different second related articles unchanged, and determining updated correlation coefficients of each second article and the first pre-recommended articles; and arranging the updated correlation coefficients of all the second related articles and the first pre-recommended articles according to the descending order, arranging the corresponding updated correlation coefficients in a third preset number of second related article sets, and determining the second related article sets as second pre-recommended article sets corresponding to the first pre-recommended article sets.
And a fourth determining module 34, configured to determine an intersection item of the first pre-recommended item set and the second pre-recommended item set as a target recommended item recommended to the target user.
The recommendation device provided by the embodiment of the invention can execute the recommendation method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Fig. 4 is a schematic structural diagram of a computer device according to an embodiment of the present invention. As shown in fig. 4, the computer device includes a processor 40 and a memory 41. The number of the processors 40 in the computer device may be one or more, and one processor 40 is taken as an example in fig. 4; the processor 40 and the memory 41 of the computer device may be connected by a bus or other means, as exemplified by the bus connection in fig. 4.
The memory 41 is a computer-readable storage medium, and can be used for storing software programs, computer-executable programs, and modules, such as program instructions and modules corresponding to the recommendation method in the embodiment of the present invention (for example, the first determination module 31, the second determination module 32, the third determination module 33, and the fourth determination module 34 in the recommendation apparatus). The processor 40 executes various functional applications and recommendations of the computer device by executing software programs, instructions and modules stored in the memory 41, i.e. implements the recommendation method described above.
The memory 41 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the computer device, and the like. Further, the memory 41 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, memory 41 may further include memory located remotely from processor 40, which may be connected to a computer device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The present invention also provides a storage medium containing computer-executable instructions which, when executed by a computer processor, are operable to perform a method of recommendation, the method comprising:
determining a historical favorite item set of a target user according to the historical behavior of the target user;
determining a first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item;
determining a second pre-recommended item set corresponding to the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item;
and determining the intersection article of the first pre-recommended article set and the second pre-recommended article set as the target recommended article recommended to the target user.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the method operations described above, and may also perform related operations in the recommendation method provided by any embodiment of the present invention.
Based on the understanding that the technical solutions of the present invention can be embodied in the form of software products, such as floppy disks, Read-Only memories (ROMs), Random Access Memories (RAMs), flash memories (F L ASHs), hard disks or optical disks of a computer, etc., and include instructions for enabling a computer device (which may be a personal computer, a computer device, or a network device, etc.) to execute the method recommendation described in the embodiments of the present invention.
It should be noted that, in the embodiment of the recommendation device, the included units and modules are merely divided according to the functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A recommendation method, comprising:
determining a historical favorite item set of a target user according to the historical behavior of the target user;
determining a first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item;
determining a second pre-recommended item set corresponding to the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item;
and determining the intersection article of the first pre-recommended article set and the second pre-recommended article set as the target recommended article recommended to the target user.
2. The method of claim 1, wherein before determining the first pre-recommended set of items corresponding to the historical favorite set of items according to the historical favorite set of items and a predetermined related set of items for each item, the method further comprises:
determining a related item set of the first item according to the correlation coefficient of each other item except the first item and the first item in the database; the number of the related items in the related item set is a first preset number, and the first item is any item in the database.
3. The method of claim 2, wherein determining the related item set of the first item according to the correlation coefficient of each other item except the first item with the first item in the database comprises:
calculating Jacard coefficients or cosine similarity of each other article and the first article, and taking the Jacard coefficients or the cosine similarity as the correlation coefficients;
and arranging the correlation coefficients according to the sequence from large to small, and determining a set of the articles with the corresponding correlation coefficients arranged in the first preset number as a related article set of the first article.
4. The method of claim 2, wherein determining the first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item comprises:
determining a first related item set of each historical favorite item and a correlation coefficient between each first related item in the first related item set and the historical favorite item in the historical favorite item set according to the historical favorite item set and the related item set of each item;
adding the correlation coefficients corresponding to the same first related articles in the first related article set of all the historical favorite articles, keeping the correlation coefficients corresponding to different first related articles unchanged, and determining the updated correlation coefficient of each first related article and the historical favorite articles;
arranging the updated correlation coefficients of all the first related articles and the historical favorite articles according to the descending order, arranging the corresponding updated correlation coefficients in a first preset number of first related article sets, and determining the first related article sets as the first pre-recommended article sets corresponding to the historical favorite article sets.
5. The method according to claim 2, wherein the determining a second pre-recommended item set corresponding to the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item comprises:
determining a second related item set of each first pre-recommended item and a correlation coefficient between each second related item in the second related item set and the first pre-recommended item in the first pre-recommended item set according to the first pre-recommended item set and the related item set of each item;
adding correlation coefficients corresponding to the same second related items in a second related item set of all the first pre-recommended items, keeping the correlation coefficients corresponding to different second related items unchanged, and determining updated correlation coefficients of each second item and the first pre-recommended items;
and arranging the updated correlation coefficients of all the second related articles and the first pre-recommended articles according to a descending order, arranging the corresponding updated correlation coefficients in a third preset number of second related article sets, and determining the second related article sets as second pre-recommended article sets corresponding to the first pre-recommended article sets.
6. The method of claim 3, wherein the item is a multimedia item;
said calculating Jacard coefficients for each of said other articles and said first article, comprising:
determining Jacard coefficients of a second item and a first item according to the number of users playing the first item and the number of users playing the second item in other items; wherein the second article is any one of the other articles.
7. The method of claim 6, wherein determining the Jacard coefficients for a first item and a second item of other items based on the number of users playing the first item and the number of users playing the second item comprises:
according to the formula
Determining the Jacard coefficients of the second article and the first article; wherein, wijRepresenting the vicard coefficient of the second item and the first item, n (j) representing the number of users playing the second item, | n (i) ∩ n (j) | representing the number of users playing the first item and the second item simultaneously, | n (i) ∪ n (j) | representing the number of users playing the first item or the second item, α being a first preset parameter, and β being a second preset parameter.
8. A recommendation device, comprising:
the first determination module is used for determining a historical favorite item set of a target user according to the historical behavior of the target user;
the second determining module is used for determining a first pre-recommended item set corresponding to the historical favorite item set according to the historical favorite item set and a predetermined related item set of each item;
a third determining module, configured to determine, according to the first pre-recommended item set and the related item set of each item, a second pre-recommended item set corresponding to the first pre-recommended item set;
a fourth determining module, configured to determine an intersection item of the first pre-recommended item set and the second pre-recommended item set as a target recommended item recommended to the target user.
9. A computer device, characterized in that the computer device comprises:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the recommendation method as claimed in any one of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the recommendation method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010311448.0A CN111506816B (en) | 2020-04-20 | 2020-04-20 | Recommendation method, recommendation device, recommendation equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010311448.0A CN111506816B (en) | 2020-04-20 | 2020-04-20 | Recommendation method, recommendation device, recommendation equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111506816A true CN111506816A (en) | 2020-08-07 |
| CN111506816B CN111506816B (en) | 2023-05-09 |
Family
ID=71876190
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010311448.0A Active CN111506816B (en) | 2020-04-20 | 2020-04-20 | Recommendation method, recommendation device, recommendation equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111506816B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112634009A (en) * | 2020-12-14 | 2021-04-09 | 中国建设银行股份有限公司 | Product pushing method and device, storage medium and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140214592A1 (en) * | 2013-01-31 | 2014-07-31 | International Business Machines Corporation | Method and system for online recommendation |
| CN107609201A (en) * | 2017-10-25 | 2018-01-19 | 广东工业大学 | A kind of recommended models generation method and relevant apparatus based on commending system |
| CN109495770A (en) * | 2018-11-23 | 2019-03-19 | 武汉斗鱼网络科技有限公司 | A kind of direct broadcasting room recommended method, device, equipment and medium |
-
2020
- 2020-04-20 CN CN202010311448.0A patent/CN111506816B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140214592A1 (en) * | 2013-01-31 | 2014-07-31 | International Business Machines Corporation | Method and system for online recommendation |
| CN107609201A (en) * | 2017-10-25 | 2018-01-19 | 广东工业大学 | A kind of recommended models generation method and relevant apparatus based on commending system |
| CN109495770A (en) * | 2018-11-23 | 2019-03-19 | 武汉斗鱼网络科技有限公司 | A kind of direct broadcasting room recommended method, device, equipment and medium |
Non-Patent Citations (1)
| Title |
|---|
| 曹夏琳;周健勇;: "基于物品的协同过滤算法对"宁波地铁go"用户个性化推荐系统研究" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112634009A (en) * | 2020-12-14 | 2021-04-09 | 中国建设银行股份有限公司 | Product pushing method and device, storage medium and electronic equipment |
| CN112634009B (en) * | 2020-12-14 | 2024-05-14 | 中国建设银行股份有限公司 | Product pushing method and device, storage medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111506816B (en) | 2023-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12124404B2 (en) | Method of and system for enhanced local-device content discovery | |
| US20230185857A1 (en) | Method and system for providing context based query suggestions | |
| CN104809154B (en) | The method and device recommended for information | |
| US7685200B2 (en) | Ranking and suggesting candidate objects | |
| US9699490B1 (en) | Adaptive filtering to adjust automated selection of content using weightings based on contextual parameters of a browsing session | |
| US9727906B1 (en) | Generating item clusters based on aggregated search history data | |
| CN110634047B (en) | Method and device for recommending house resources, electronic equipment and storage medium | |
| CN102184199A (en) | Network information recommending method and system | |
| CN107704560B (en) | Information recommendation method, device and equipment | |
| CN104969224A (en) | Tailoring user experience for unrecognized and new users | |
| CN110110206B (en) | Method, device, computing equipment and storage medium for mining and recommending relationships among articles | |
| CN102054003A (en) | Methods and systems for recommending network information and creating network resource index | |
| US9767204B1 (en) | Category predictions identifying a search frequency | |
| CN110825977A (en) | A data recommendation method and related equipment | |
| CN105719164A (en) | Paid multimedia resource recommending method and paid multimedia resource recommending device | |
| CN105894310A (en) | Personalized recommendation method | |
| CN109587527B (en) | Personalized video recommendation method and device | |
| CN111506816B (en) | Recommendation method, recommendation device, recommendation equipment and storage medium | |
| CN108319622A (en) | A kind of media content recommendations method and device | |
| US20140325378A1 (en) | Method and apparatus for generating a personalized page | |
| CN112507220B (en) | Information push method, device and medium | |
| CN114265972A (en) | Content recommendation method and device, storage medium and electronic device | |
| CN111491198B (en) | Small video searching method and device | |
| JPWO2013011728A1 (en) | Communication support device, communication support method, and program | |
| CN111191056A (en) | A kind of multimedia recommendation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |