CN114564653A - Information recommendation method and device, server and storage medium - Google Patents
Information recommendation method and device, server and storage medium Download PDFInfo
- Publication number
- CN114564653A CN114564653A CN202011364852.0A CN202011364852A CN114564653A CN 114564653 A CN114564653 A CN 114564653A CN 202011364852 A CN202011364852 A CN 202011364852A CN 114564653 A CN114564653 A CN 114564653A
- Authority
- CN
- China
- Prior art keywords
- account
- information
- target
- recommended
- accounts
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06F16/951—Indexing; Web crawling techniques
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure relates to an information recommendation method, an information recommendation device, a server and a storage medium, and belongs to the technical field of internet. The method comprises the following steps: determining a target account set; a target second account with at least one dimension in the attribute information being the same as the dimension of the attribute information of the account to be recommended; determining the information preference type of the target second account as the target information preference type of the account to be recommended; acquiring a target first account with the type of the issued information matched with the target information preference type from a target account set; and recommending the information issued by the target first account to the account to be recommended. Because the registration time of the account to be recommended and the registration time of the second account are similar, the information preference type of the second account is more similar to the information preference type of the account to be recommended, and the information issued by the first account can trigger the account to be recommended to issue the information, the information issued by the target first account and the information preference type of the account to be recommended are recommended to the account to be recommended, and the recommended information is more accurate.
Description
Technical Field
The present disclosure relates to the field of internet technologies, and in particular, to an information recommendation method, an information recommendation apparatus, a server, and a storage medium.
Background
Generally, when a user uses an application program, the application platform recommends some information to the user based on the user's preference. Since the amount of information about the new user in the application platform is very small, how to recommend information to the new user becomes crucial at this time.
In the related art, a plurality of pieces of hot information are randomly acquired from a hot resource pool, and then the acquired plurality of pieces of hot information are recommended to a new user.
However, the information recommended to the new user by the related art is randomly selected, the new user may not be interested in the recommended information, and thus, the information recommended by the related art is not accurate.
Disclosure of Invention
The present disclosure provides an information recommendation method, device and system to at least solve the problem of inaccurate recommended information in the related art. The technical scheme of the disclosure is as follows:
according to a first aspect of the embodiments of the present disclosure, there is provided an information recommendation method, the method including:
determining a target account set according to attribute information of an account to be recommended, wherein at least one piece of target index information with the same dimension as the attribute information of the account to be recommended exists in multi-dimensional index information of the target account set, the target account set comprises a plurality of first accounts, the first accounts are used for triggering second account release information with at least one dimension being the same as the target index information in the attribute information, and the second accounts refer to accounts with registration time within a preset time range;
acquiring a target second account with at least one dimension in the attribute information being the same as the dimension of the attribute information of the account to be recommended according to the attribute information of the plurality of second accounts;
determining the information preference type of the target second account as the target information preference type of the account to be recommended;
according to the target information preference type, acquiring a target first account of which the type of the issued information is matched with the target information preference type from the target account set;
and recommending the information issued by the target first account to the account to be recommended.
In another embodiment of the present disclosure, the determining a target account set according to attribute information of an account to be recommended includes:
generating an attribute vector according to at least one dimension in the attribute information of the account to be recommended;
generating a set vector of each account set according to the multi-dimensional index information of each account set;
calculating the similarity between the attribute vector and a set vector of each account set;
and determining the account set with the similarity larger than a preset threshold as the target account set.
In another embodiment of the present disclosure, before determining the target account set according to the attribute information of the account to be recommended, the method further includes:
acquiring a plurality of first account numbers;
and dividing first account numbers, which are used for triggering second account numbers with attribute information of the same dimensionality, in the plurality of first account numbers to release information into the same account number set to obtain a plurality of account number sets.
In another embodiment of the present disclosure, before determining the information preference type of the target second account as the target information preference type of the account to be recommended, the method further includes:
acquiring a plurality of first account numbers;
determining the types of the information of which the plurality of second accounts are operated as the information preference types of the plurality of second accounts according to the operation behaviors of the plurality of second accounts on the information issued by the plurality of first accounts.
In another embodiment of the present disclosure, the obtaining a plurality of first account numbers includes:
determining a plurality of account numbers for issuing the information according to the information of the operation implemented by the plurality of second account numbers;
and acquiring a plurality of first account numbers from the plurality of account numbers.
In another embodiment of the present disclosure, the obtaining a plurality of first account numbers from the plurality of account numbers includes:
determining a second account number of which the number of issued information meets the target condition;
and determining the account concerned by the determined second account as the plurality of first accounts.
In another embodiment of the present disclosure, before recommending the information issued by the target first account to the account to be recommended, the method further includes:
determining a recommendation sequence of information issued by the target first account;
the recommending the information issued by the target first account to the account to be recommended includes:
and recommending the information issued by the target first account to the account to be recommended according to the recommendation sequence.
In another embodiment of the present disclosure, after recommending the information issued by the target first account to the account to be recommended, the method further includes:
and updating the target information preference type according to the operation behavior of the account to be recommended on the recommended information.
According to a second aspect of the embodiments of the present disclosure, there is provided an information recommendation apparatus, the apparatus including:
the system comprises a determining module, a recommending module and a recommending module, wherein the determining module is used for determining a target account set according to attribute information of an account to be recommended, at least one piece of target index information with the same dimension as the attribute information of the account to be recommended exists in multi-dimensional index information of the target account set, the target account set comprises a plurality of first accounts, the first accounts are used for triggering second account release information with at least one dimension in the attribute information being the same as the target index information, and the second accounts refer to accounts with the registration time within a preset time range;
the acquisition module is used for acquiring a target second account with at least one dimension in the attribute information being the same as the dimension of the attribute information of the account to be recommended according to the attribute information of the plurality of second accounts;
the determining module is further configured to determine an information preference type of the target second account as a target information preference type of the account to be recommended;
the acquisition module is further used for acquiring a target first account of which the type of the issued information is matched with the target information preference type from the target account set according to the target information preference type;
and the recommending module is used for recommending the information issued by the target first account to the account to be recommended.
In another embodiment of the present disclosure, the determining module is configured to generate an attribute vector according to at least one dimension in the attribute information of the account to be recommended; generating a set vector of each account set according to the multi-dimensional index information of each account set; calculating the similarity between the attribute vector and a set vector of each account set; and determining the account set with the similarity larger than a preset threshold as the target account set.
In another embodiment of the present disclosure, the apparatus further comprises:
the obtaining module is further configured to obtain a plurality of first account numbers;
and the account number dividing module is used for dividing the first account numbers of the plurality of first account numbers, which are used for triggering the second account numbers with the attribute information of the same dimensionality, to the same account number set to obtain a plurality of account number sets.
In another embodiment of the present disclosure, the obtaining module is further configured to obtain a plurality of first accounts;
the determining module is further configured to determine, according to operation behaviors of the plurality of second accounts on the information issued by the plurality of first accounts, types of information on which the plurality of second accounts have performed operations as information preference types of the plurality of second accounts.
In another embodiment of the present disclosure, the obtaining module is further configured to determine, according to information that the plurality of second accounts have performed operations, a plurality of accounts that issue the information; and acquiring a plurality of first account numbers from the plurality of account numbers.
In another embodiment of the present disclosure, the obtaining module is further configured to determine a second account with a quantity of published information meeting a target condition; and determining the account concerned by the determined second account as the plurality of first accounts.
In another embodiment of the present disclosure, the determining module is further configured to determine a recommendation order of information issued by the target first account;
and the recommending module is used for recommending the information issued by the target first account to the account to be recommended according to the recommending sequence.
In another embodiment of the present disclosure, the apparatus further comprises:
and the updating module is used for updating the target information preference type according to the operation behavior of the account to be recommended on the recommended information.
According to a third aspect of the embodiments of the present disclosure, there is provided a server, including:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the information recommendation method of an aspect.
According to a fourth aspect of embodiments of the present disclosure, there is provided a storage medium having instructions that, when executed by a processor of a server, enable the server to perform the information recommendation method of an aspect.
According to a fifth aspect of embodiments of the present disclosure, there is provided a computer program product, wherein instructions of the computer program product, when executed by a processor of a server, enable the server to perform the information recommendation method of the first aspect.
The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects:
because the registration time of the account to be recommended and the registration time of the second account are similar, the information preference type of the second account is more similar to the information preference type of the account to be recommended, and the information issued by the first account can trigger the account to be recommended to issue the information, the information issued by the target first account and the information preference type of the account to be recommended are recommended to the account to be recommended, and the recommended information is more accurate.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.
Fig. 1 is an illustration of an implementation environment in which a method for recommending information is involved, according to an example embodiment.
Fig. 2 is a flow chart illustrating an information recommendation method according to an example embodiment.
Fig. 3 is a flow chart illustrating an information recommendation method according to an example embodiment.
Fig. 4 is a block diagram illustrating an information recommendation apparatus according to an example embodiment.
FIG. 5 illustrates a server for information recommendation, according to an example embodiment.
Detailed Description
In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
The user information to which the present disclosure relates may be information authorized by the user or sufficiently authorized by each party.
In the field of internet technology, a main goal of a recommendation system is to recommend a large number of subjects to a potentially favorite large number of users. The object and the user are related, for any recommendation system, the object and the user are continuously changed, and new objects and new registered users are frequently encountered during the operation of the recommendation system, so that the cold start problem needs to be solved.
The cold start includes a user cold start, a subject cold start, a system cold start, and the like. The user cold start means that for a new registered user, because the operation behaviors are few or none, the original recommendation method based on the user operation behaviors is not applicable to the new registered user, and at the moment, how to recommend a subject matter to the new registered user enables the user to be satisfied, namely, the user cold start problem is solved. The cold start of the target object means that for the newly warehoused target object, before warehousing, it is not known which users are interested in the new target object, and therefore, how to recommend the new target object to the interested users is the problem of the cold start of the target object. The system cold start refers to a problem of system cold start, which is that for a newly developed application program, because the number of registered users at an initial stage is small, the operation behaviors of the users are also small, and common algorithms depending on a large number of user operation behaviors such as collaborative filtering, deep learning and the like cannot train an accurate recommendation model, at this time, how to operate the recommendation system makes the recommended information more accurate.
The embodiment of the disclosure provides an information recommendation method, which mainly solves the problem of cold start of a user. The method includes the steps of obtaining a plurality of first accounts capable of triggering a second account to release information, dividing the plurality of first accounts into a plurality of account sets, determining a target account set based on attribute information of an account to be recommended and the plurality of account sets, determining an information preference type of the account to be recommended based on operation behaviors of the plurality of first accounts and the plurality of second accounts, determining a target first account of which the type of the released information is the same as the information preference type of the account to be recommended based on the attribute information of the account to be recommended, the attribute information of the second account and the information preference type of a reference account, and then recommending the information released by the target first account to the account to be recommended.
Referring to fig. 1, a system architecture of an information recommendation system provided by an embodiment of the present disclosure is shown, and referring to fig. 1, the information recommendation system includes a terminal 101 and a server 102.
The terminal 101 is installed with an application having an information recommendation function, which may be a browser application, a news application, a shopping application, a social contact application, or the like. The terminal 101 has a display screen and is capable of displaying information recommended by the server 102, the terminal 101 may be a smart phone, a tablet computer, a notebook computer, or the like, and the embodiment of the present disclosure does not specifically limit the product type of the terminal 101.
The server 102 is a background server of an application program with an information recommendation function. The server 102 may be a single server or a server cluster composed of a plurality of servers. The server 102 has strong information storage capacity, can store information issued by different account numbers, and can also store operation information of each account number; the server 102 also has strong computing power, and can recommend information for the account to be recommended according to the information preference type corresponding to the second account with the registration time within the preset time range.
The terminal 101 and the server 102 may communicate with each other through a wired network or a wireless network.
Based on the implementation environment shown in fig. 1, fig. 2 is a flowchart illustrating an information recommendation method according to an exemplary embodiment, and as shown in fig. 2, the information recommendation method is used in a server and includes the following steps.
201. And determining a target account set according to the attribute information of the account to be recommended.
The target account set comprises a plurality of first accounts, the first accounts are used for triggering second account release information, at least one dimension of the first accounts is the same as the attribute information of the account to be recommended, and the second accounts refer to accounts with the registration time within a preset time range.
202. And acquiring a target second account with at least one dimension in the attribute information being the same as the dimension of the attribute information of the account to be recommended according to the attribute information of the plurality of second accounts.
203. And determining the information preference type of the target second account as the target information preference type of the account to be recommended.
204. And acquiring a target first account of which the type of the issued information is matched with the target information preference type from the target account set according to the target information preference type.
205. And recommending the information issued by the target first account to the account to be recommended.
According to the method provided by the embodiment of the disclosure, because the registration time of the account to be recommended and the registration time of the second account are similar, the information preference type of the second account is more similar to the information preference type of the account to be recommended, and because the information issued by the first account can trigger the account to be recommended to issue information, the information issued by the target first account and the information preference type of the account to be recommended are recommended to the account to be recommended, and the recommended information is more accurate.
In another embodiment of the present disclosure, determining a target account set according to attribute information of an account to be recommended includes:
generating an attribute vector according to at least one dimension in the attribute information of the account to be recommended;
generating a set vector of each account set according to the multi-dimensional index information of each account set;
calculating the similarity between the attribute vector and the set vector of each account set;
and determining the account set with the similarity larger than a preset threshold as a target account set.
In another embodiment of the present disclosure, before determining the target account set according to the attribute information of the account to be recommended, the method further includes:
acquiring a plurality of first account numbers;
dividing first accounts, which are used for triggering second accounts with attribute information of the same dimensionality and used for issuing information, in the multiple first accounts into the same account set to obtain multiple account sets.
In another embodiment of the present disclosure, before determining the information preference type of the target second account as the target information preference type of the account to be recommended, the method further includes:
acquiring a plurality of first account numbers;
and determining the types of the information subjected to the operation by the plurality of second accounts as the information preference types of the plurality of second accounts according to the operation behaviors of the plurality of second accounts on the information issued by the plurality of first accounts.
In another embodiment of the present disclosure, acquiring a plurality of first accounts includes:
determining a plurality of account numbers for issuing information according to the information of the operation implemented by the plurality of second account numbers;
a plurality of first account numbers are obtained from a plurality of account numbers.
In another embodiment of the present disclosure, acquiring a plurality of first account numbers from a plurality of account numbers includes:
determining a second account number of which the number of issued information meets the target condition;
and determining the account concerned by the determined second account as a plurality of first accounts.
In another embodiment of the present disclosure, before recommending the information issued by the target first account to the account to be recommended, the method further includes:
determining a recommendation sequence of information issued by a target first account;
recommending information issued by a target first account to an account to be recommended, wherein the recommending comprises the following steps:
and recommending the information issued by the target first account to the account to be recommended according to the recommendation sequence.
In another embodiment of the present disclosure, after recommending the information issued by the target first account to the account to be recommended, the method further includes:
and updating the target information preference type according to the operation behavior of the account to be recommended on the recommended information.
All the above optional technical solutions may be combined arbitrarily to form the optional embodiments of the present disclosure, and are not described herein again.
Based on the implementation environment shown in fig. 1, fig. 3 is a flowchart illustrating an information recommendation method according to an exemplary embodiment, where as shown in fig. 3, the information recommendation method is used in a server and includes the following steps.
In step 301, a server obtains a plurality of first accounts.
The first account is an account capable of triggering the second account to release information. For example, after any user publishes information through a registered account, if other users view the information, under the enlightenment of the information, the information with the same or similar content as the information is published, and then the account is the first account. The first account number may also be referred to as the account number of the high-quality author, depending on its role.
The second account is an account with registration time within a preset time range. The preset time range can ensure that the second account is an account which is not registered in the application soon but can acquire the operation behavior information, and the preset time range can be obtained by counting a plurality of second accounts and can be 7 days, 10 days and the like. In a possible implementation manner, the server may use all accounts with registration time within a preset time range as the second account. Considering that the number of accounts with the registration time within the preset time range is large, if all the accounts with the registration time within the preset time range are used as the second accounts, the processing pressure of the server is greatly increased, and in order to ensure the accuracy of information recommendation and reduce the consumption of computing resources in information recommendation, in another possible implementation manner, the server may obtain a plurality of accounts from all the accounts with the registration time within the preset time range as the second accounts. When a plurality of accounts are acquired as second accounts from all accounts whose registration time is within a preset time range, it is required to ensure that the second accounts of each attribute information can be acquired.
Because the registration time of the second account is short, and the operation behavior and preference of the second account are more similar to those of the newly registered account, the embodiment of the disclosure recommends information to the newly registered account based on the information preference type of the second account, and compared with the prior art that hot information is randomly acquired from a hot resource pool or information is recommended to the newly registered account based on the attribute information of the newly registered account, the recommended information is more accurate and more conforms to the preference of a new user.
When the server acquires a plurality of first account numbers, the following method can be adopted:
3011. and the server determines a plurality of accounts for issuing information according to the information of the operation implemented by the plurality of second accounts.
The operation carried out by the second account comprises at least one operation of viewing, praise, comment, collection, attention and the like. Because the server stores the operation information of each account in the embodiment of the present disclosure, based on the operation information of each account stored by the server, the server can acquire information that a plurality of second accounts have performed operations, and further determine the account issuing the information as a plurality of accounts.
3012. The server acquires a plurality of first account numbers from a plurality of account numbers.
When the server acquires a plurality of first accounts from a plurality of accounts, a second account with the number of release information meeting the target condition can be determined, and the account concerned by the determined second account is determined as the plurality of first accounts. The target condition may be that the information amount is greater than a first preset amount, and the first preset amount may be 10, 20, and so on. For example, if the first preset number is 10, the server acquires second account numbers with the number of the release information larger than 10, and determines the account numbers concerned by the 10 second account numbers as a plurality of first account numbers. The target condition may also rank the number of messages by a second predetermined number, which may be 5, 10, etc. For example, if the second preset number is 10, the server sorts the plurality of second account numbers in the order of decreasing the number of the issued information, obtains the second account number ranked at the top 10, and determines the account number concerned by the second account number ranked at the top 10 as the plurality of first account numbers.
In step 302, the server determines the type of the information operated by the plurality of second accounts as the information preference type of the plurality of second accounts according to the operation behavior of the plurality of second accounts on the information issued by the plurality of first accounts.
The type of the information includes gourmet, laugh, comedy, travel, emotion, and the like, and the embodiment of the present disclosure does not specifically limit the type of the information.
In an embodiment of the disclosure, if each piece of information has an explicit type, based on an operation behavior of each second account on information issued by a plurality of first accounts, the server obtains a type of the information on which each second account has performed an operation, and determines the type of the information on which each second account has performed the operation as an information preference type of each second account. For example, for any second account, if the type of the information of the second account subjected to the operation of praise, view, and the like is comedy and food, it is determined that the information preference type of the second account is comedy and food.
In another embodiment of the disclosure, considering that some information has no explicit type, for information without an explicit type, the server may obtain a type to which a first account issuing the information belongs, determine the type to which the first account issuing the information belongs as the type of the information, and determine the type of the information as the information preference type of the first account.
Before the server obtains the type of the first account of the information, the server needs to determine the type of each first account. Specifically, the server may divide the first accounts with the same information type into the same set according to the type of the information issued by the first account, so as to obtain a plurality of information type sets. When the server divides the first account numbers with the same information type into the same set according to the type of the information issued by the first account numbers, a type label can be manually labeled for each first account number, and the first account numbers with the same label are gathered into one type based on labeling results to obtain a plurality of information type sets; the server can also preset a plurality of reference types, and the first account numbers with the same type of the issued information and the same reference type are gathered into one type to obtain a plurality of information type sets; the server can also classify the plurality of first account numbers based on a pre-trained classification model, and further divide the first account numbers with the same information type into the same set based on a classification result to obtain a plurality of information type sets. Of course, the server may also adopt other modes, which are not described one by one here.
By adopting the method, the plurality of first accounts are classified, so that the first accounts in each information type set correspond to at least one type, and therefore when information is recommended to a new user, the information can be recommended according to the type of the information issued by the first accounts. For example, by classifying the plurality of first accounts, an information type set with a comedy release information type and an information type set with a beauty release information type can be obtained, and when recommending information to a new user, if the new user likes to see the comedy type information, the information released by the first account in the information type set corresponding to the comedy is recommended to the new user.
Further, in order to facilitate management of the information preference type of each second account, the server may establish an information preference list corresponding to each second account based on the information preference type of each second account, so that when recommending information to each second account, information can be recommended to each second account in a targeted manner based on the information preference list corresponding to each second account.
In step 303, the server divides first accounts, which are used for triggering second account publishing information having attribute information with the same dimensionality, of the multiple first accounts into the same account set to obtain multiple account sets.
The attribute information includes, among others, age, native place, sex, and the like. Considering that the age, native place, sex and the like influence the preference of users, the effect of information issued by the first account on users of different ages, native places, sexes and the like is different, and the effect on users of the same age, native places, sexes and the like may be the same, so that in order to facilitate the server to recommend information to the second account with different attribute information and the newly registered account for different attribute information, the server can divide the first accounts of the plurality of first accounts, which are used for triggering the second accounts with attribute information of the same dimension to issue information, into the same account set to obtain the plurality of account sets. For example, the server forms a first account set by the first accounts capable of triggering the female user to release information and forms an account set by the first accounts capable of triggering the male user to release information according to the gender attribute information; for another example, the server forms a set of accounts with first accounts capable of triggering the same native user to publish information according to the native place.
Further, for convenience of subsequent query, each account set has index information, the index information of each account set has at least one dimension, and the index information of each account set is used for indicating attribute information of a second account, in the account set, of which a first account can trigger information release, on the at least one dimension. By establishing the corresponding relation between the index information and the account set, the information retrieval efficiency is improved, and the fact that the user is full and complete is guaranteed.
In step 304, the server determines a target account set according to the attribute information of the account to be recommended.
The account to be recommended includes an account which is just registered, an account which is registered but does not perform any operation or has few operation actions on an application program, and the like. At least one piece of target index information with the same dimension as the first attribute information exists in the multi-dimensional index information of the target account set.
When the server determines the target account set according to the attribute information of the account to be recommended, the following method can be adopted:
3041. and the server generates an attribute vector according to at least one dimension in the attribute information of the account to be recommended.
The server extracts at least one dimensional attribute feature from the attribute information of the account to be recommended, and the extracted at least one dimensional attribute feature forms an attribute vector. For example, the server extracts two-dimensional attribute features of age and gender from the attribute information of the account to be recommended, and forms the extracted two-dimensional attribute features into an attribute vector.
3042. And the server generates a set vector of each account set according to the multi-dimensional index information of each account set.
And the server forms a set vector of each account set by using the index information of each dimension of each account set. For example, if the index information of the account set includes sex males and ages 20-30, the set vector of the account set is composed of two features of sex males and ages 20-30.
3043. The server calculates the similarity between the attribute vector and the collection vector of each account set.
When the server calculates the similarity between the attribute vector and the set vector of each account set, the Euclidean distance between the attribute vector and the set vector of each account set can be calculated, and the similarity between the attribute vector and the set vector of each account set is determined based on the calculated Euclidean distance; the server can also calculate the cosine distance between the attribute vector and the set vector of each account set, and further determine the calculated cosine distance as the similarity between the attribute vector and the set vector of each account set. Of course, the server may also use other methods to calculate the similarity between the attribute vector and the set vector of each account set, which is not described herein one by one.
3044. And the server determines the account set with the similarity larger than a preset threshold as a target account set.
Wherein, the preset threshold value can be 90%, 95%, etc. And when the similarity between the attribute vector and the set vector of any account number set is greater than a preset threshold value, the server determines the account number set as a target account number set.
In step 305, the server obtains a target second account with at least one dimension in the attribute information being the same as the dimension of the attribute information of the account to be recommended according to the attribute information of the plurality of second accounts.
And the server matches the attribute information of the account to be recommended with the attribute information of a plurality of second accounts, and takes the second account with at least one dimension in the attribute information being the same as the dimension of the attribute information of the account to be recommended as a target second account. For example, the attribute information of the account to be recommended includes sex male, native Beijing, and the like, the attribute information of the second account is matched with the attribute information of the account to be recommended, and the second account with male dimensionality in the attribute information of the second account is determined as the target second account.
In step 306, the server determines the information preference type of the target second account as the target information preference type of the account to be recommended.
Considering that at least one dimension in the attribute information of the target second account is the same as the dimension of the attribute information of the account to be recommended, the probability that the information preference types of the second account and the account to be recommended are similar is high, and under the condition that the information preference type of the account to be recommended cannot be obtained or is difficult to obtain based on the operation behavior of the account to be recommended, the information preference type corresponding to the target reference account is determined as the target information preference type of the account to be recommended, the recommendation is performed based on the determined target information preference type, and compared with the random recommendation or the recommendation performed based on the attribute information of the account to be recommended, the recommendation result is more accurate.
In step 307, the server obtains a target first account with the type of the issued information matching the target information preference type from the target account set according to the target information preference type.
The server acquires a first account with the type of the issued information matched with the target information preference type from the target account set based on the target information preference type of the account to be recommended, and then uses the first account as a target first account.
In step 308, the server recommends the information issued by the target first account to the account to be recommended.
In an embodiment of the disclosure, based on the determined target first account, the server acquires information issued by the target first account, and recommends the information issued by the target first account to the account to be recommended.
In another embodiment of the disclosure, in consideration of the fact that the number of information issued by the target first account is large, the size of a display screen of a terminal logging in the account to be recommended is limited, and the preference degrees of the account to be recommended to the information issued by the target first account are different, so that before recommending the information issued by the target first account to the account to be recommended, the server may determine the recommendation sequence of the information issued by the target first account, and then recommend the information issued by the target first account to the account to be recommended according to the recommendation sequence. When the server determines the recommendation sequence of the information issued by the target first account, the recommendation sequence of the information can be determined according to the sequence from high to low of the click rate, the collection amount or the review amount and the like; and further, the recommendation score of each piece of information is determined based on the set weight value, the click rate, the collection amount and the comment amount, and the recommendation sequence is determined according to the recommendation scores. Of course, the server may also determine the recommendation order of the information by other methods, which are not described herein.
In another embodiment of the disclosure, after recommending, by the server, the information issued by the target first account to the account to be recommended, an operation behavior of the account to be recommended on the information issued by the target first account is also acquired, and then the target information preference type of the account to be recommended is updated according to the operation behavior of the account to be recommended on the recommended information. For example, the target information preference type of the account to be recommended is gourmet, tourism, and beauty, after the information issued by the target first account is recommended to the account to be recommended, the server acquires that the account to be recommended performs a praise operation on the information of the comedy type, and then the server adds comedy to the target information preference type of the account to be recommended. The embodiment of the disclosure searches the interest type of the user based on the real-time online behavior of the user as feedback information, and then recommends information to the user based on the searched interest type, thereby completing the recommendation function of cold start of the user, not only meeting the interests and hobbies of the user, but also realizing the exploration of the user interest, and being very friendly to the long-term experience of the user.
According to the method provided by the embodiment of the disclosure, because the registration time of the account to be recommended and the registration time of the second account are similar, the information preference type of the second account is more similar to the information preference type of the account to be recommended, and because the information issued by the first account can trigger the account to be recommended to issue information, the information issued by the target first account and the information preference type of the account to be recommended are recommended to the account to be recommended, and the recommended information is more accurate.
Fig. 4 is a block diagram illustrating an information recommendation apparatus according to an example embodiment. Referring to fig. 4, the apparatus includes: a determination module 401, an acquisition module 402 and a recommendation module 403.
The determining module 401 is configured to determine a target account set according to attribute information of an account to be recommended, where at least one piece of target index information with the same dimension as the attribute information of the account to be recommended exists in multidimensional index information of the target account set, the target account set includes a plurality of first accounts, the first accounts are used to trigger issuing information of a second account with at least one dimension in the attribute information being the same as the target index information, and the second account refers to an account with registration time within a preset time range;
an obtaining module 402, configured to obtain, according to attribute information of multiple second accounts, a target second account having at least one dimension in the attribute information that is the same as the dimension of the attribute information of the account to be recommended;
the determining module 401 is further configured to determine the information preference type of the target second account as a target information preference type of the account to be recommended;
the obtaining module 402 is further configured to obtain, from the target account set, a target first account whose type of the issued information matches the target information preference type according to the target information preference type;
the recommending module 403 is configured to recommend the information issued by the target first account to the account to be recommended.
In another embodiment of the present disclosure, the determining module 401 is configured to generate an attribute vector according to at least one dimension in the attribute information of the account to be recommended; generating a set vector of each account set according to the multi-dimensional index information of each account set; calculating the similarity between the attribute vector and the set vector of each account set; and determining the account set with the similarity larger than a preset threshold as a target account set.
In another embodiment of the present disclosure, the apparatus further comprises: and an account number division module.
An obtaining module 402, configured to obtain a plurality of first accounts;
the account number dividing module is used for dividing first account numbers, which are used for triggering second account numbers with attribute information of the same dimensionality, in the multiple first account numbers into the same account number set to obtain multiple account number sets.
In another embodiment of the present disclosure, the obtaining module 401 is further configured to obtain a plurality of first accounts;
the determining module 402 is further configured to determine, according to operation behaviors of the multiple second accounts on information issued by the multiple first accounts, types of information on which operations are performed by the multiple second accounts as information preference types of the multiple second accounts.
In another embodiment of the present disclosure, the obtaining module 401 is further configured to determine, according to information that a plurality of second accounts have performed an operation, a plurality of accounts that issue information; a plurality of first account numbers are obtained from a plurality of account numbers.
In another embodiment of the present disclosure, the obtaining module 401 is further configured to determine a second account with a quantity of published information meeting a target condition; and determining the account concerned by the determined second account as a plurality of first accounts.
In another embodiment of the present disclosure, the determining module 402 is further configured to determine a recommendation order of information issued by the target first account;
the recommending module 403 is configured to recommend the information issued by the target first account to the account to be recommended according to the recommending order.
In another embodiment of the present disclosure, the apparatus further comprises: and updating the module.
And the updating module is used for updating the target information preference type according to the operation behavior of the account to be recommended on the recommended information.
According to the device provided by the embodiment of the disclosure, because the registration time of the account to be recommended and the registration time of the second account are similar, the information preference type of the second account is more similar to the information preference type of the account to be recommended, and because the information issued by the first account can trigger the account to be recommended to issue information, the information issued by the target first account and the information issued by the account to be recommended are recommended to the account to be recommended, and the recommended information is more accurate.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
FIG. 5 illustrates a server for information recommendation, according to an example embodiment. Referring to fig. 5, server 500 includes a processing component 522 that further includes one or more processors and memory resources, represented by memory 532, for storing instructions, such as applications, that are executable by processing component 522. The application programs stored in memory 532 may include one or more modules that each correspond to a set of instructions. Further, the processing component 522 is configured to execute instructions to perform the functions performed by the server in the information recommendation method described above.
The server 500 may also include a power component 526 configured to perform power management for the server 500, a wired or wireless network interface 550 configured to connect the server 500 to a network, and an input/output (I/O) interface 558. The Server 500 may operate based on an operating system, such as Windows Server, stored in the memory 532TM，Mac OS XTM，UnixTM,LinuxTM，FreeBSDTMOr the like.
According to the server provided by the embodiment of the disclosure, because the registration time of the account to be recommended and the registration time of the second account are similar, the information preference type of the second account is more similar to the information preference type of the account to be recommended, and because the information issued by the first account can trigger the account to be recommended to issue information, the information issued by the target first account and the information issued by the account to be recommended are recommended to the account to be recommended, and the recommended information is more accurate.
Embodiments of the present disclosure provide a storage medium that may be a non-transitory computer readable storage medium, which may be, for example, a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like. The instructions in the storage medium, when executed by a processor of the server, enable the server to perform the information recommendation method described in the above embodiments.
According to the storage medium provided by the embodiment of the disclosure, because the registration time of the account to be recommended and the registration time of the second account are similar, the information preference type of the second account is more similar to the information preference type of the account to be recommended, and because the information issued by the first account can trigger the account to be recommended to issue information, the information issued by the target first account and the information issued by the account to be recommended are recommended to the account to be recommended, and the recommended information is more accurate.
The disclosed embodiments provide a computer program product, and instructions in the computer program product, when executed by a processor of a server, enable the server to execute the information recommendation method according to the above embodiments.
According to the computer program product provided by the embodiment of the disclosure, because the registration time of the account to be recommended and the registration time of the second account are similar, the information preference type of the second account is more similar to the information preference type of the account to be recommended, and because the information issued by the first account can trigger the account to be recommended to issue the information, the information issued by the target first account and the information preference type of the account to be recommended are recommended to the account to be recommended, and the recommended information is more accurate.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (10)
1. An information recommendation method, characterized in that the method comprises:
determining a target account set according to attribute information of an account to be recommended, wherein at least one piece of target index information with the same dimension as the attribute information of the account to be recommended exists in multi-dimensional index information of the target account set, the target account set comprises a plurality of first accounts, the first accounts are used for triggering second account release information with at least one dimension being the same as the target index information in the attribute information, and the second accounts refer to accounts with registration time within a preset time range;
acquiring a target second account with at least one dimension in the attribute information being the same as the dimension of the attribute information of the account to be recommended according to the attribute information of the plurality of second accounts;
determining the information preference type of the target second account as the target information preference type of the account to be recommended;
according to the target information preference type, acquiring a target first account of which the type of the issued information is matched with the target information preference type from the target account set;
and recommending the information issued by the target first account to the account to be recommended.
2. The information recommendation method according to claim 1, wherein the determining a target account set according to the attribute information of the account to be recommended comprises:
generating an attribute vector according to at least one dimension in the attribute information of the account to be recommended;
generating a set vector of each account set according to the multi-dimensional index information of each account set;
calculating the similarity between the attribute vector and a set vector of each account set;
and determining the account set with the similarity larger than a preset threshold as the target account set.
3. The information recommendation method according to claim 1, wherein before determining the target account set according to the attribute information of the account to be recommended, the method further comprises:
acquiring a plurality of first account numbers;
and dividing first account numbers, which are used for triggering second account numbers with attribute information of the same dimensionality, in the plurality of first account numbers to release information into the same account number set to obtain a plurality of account number sets.
4. The information recommendation method according to claim 1, wherein before determining the information preference type of the target second account as the target information preference type of the account to be recommended, the method further comprises:
acquiring a plurality of first account numbers;
determining the types of the information of which the plurality of second accounts are operated as the information preference types of the plurality of second accounts according to the operation behaviors of the plurality of second accounts on the information issued by the plurality of first accounts.
5. The information recommendation method according to claim 3 or 4, wherein the obtaining a plurality of first accounts comprises:
determining a plurality of account numbers for issuing information according to the information of the operation implemented by the plurality of second account numbers;
and acquiring a plurality of first account numbers from the plurality of account numbers.
6. The information recommendation method according to claim 5, wherein the obtaining a plurality of first account numbers from the plurality of account numbers comprises:
determining a second account number of which the number of issued information meets the target condition;
and determining the account concerned by the determined second account as the plurality of first accounts.
7. The information recommendation method according to any one of claims 1 to 4, wherein before recommending the information issued by the target first account to the account to be recommended, the method further includes:
determining a recommendation sequence of information issued by the target first account;
the recommending the information issued by the target first account to the account to be recommended includes:
and recommending the information issued by the target first account to the account to be recommended according to the recommendation sequence.
8. An information recommendation apparatus, characterized in that the apparatus comprises:
the system comprises a determining module, a recommending module and a recommending module, wherein the determining module is used for determining a target account set according to attribute information of an account to be recommended, at least one piece of target index information with the same dimension as the attribute information of the account to be recommended exists in multi-dimensional index information of the target account set, the target account set comprises a plurality of first accounts, the first accounts are used for triggering second account release information with at least one dimension in the attribute information being the same as the target index information, and the second accounts refer to accounts with the registration time within a preset time range;
the acquisition module is used for acquiring a target second account with at least one dimension in the attribute information being the same as the dimension of the attribute information of the account to be recommended according to the attribute information of the plurality of second accounts;
the determining module is further configured to determine an information preference type of the target second account as a target information preference type of the account to be recommended;
the acquisition module is further used for acquiring a target first account of which the type of the issued information is matched with the target information preference type from the target account set according to the target information preference type;
and the recommending module is used for recommending the information issued by the target first account to the account to be recommended.
9. A server, comprising:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the information recommendation method of any one of claims 1 to 7.
10. A storage medium, characterized in that instructions in the storage medium, when executed by a processor of a server, enable the server to perform the information recommendation method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011364852.0A CN114564653A (en) | 2020-11-27 | 2020-11-27 | Information recommendation method and device, server and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011364852.0A CN114564653A (en) | 2020-11-27 | 2020-11-27 | Information recommendation method and device, server and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114564653A true CN114564653A (en) | 2022-05-31 |
Family
ID=81712599
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011364852.0A Pending CN114564653A (en) | 2020-11-27 | 2020-11-27 | Information recommendation method and device, server and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114564653A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116662659A (en) * | 2023-05-31 | 2023-08-29 | 福建莫界文化发展有限公司 | Media content intelligent recommendation system based on artificial intelligence |
-
2020
- 2020-11-27 CN CN202011364852.0A patent/CN114564653A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116662659A (en) * | 2023-05-31 | 2023-08-29 | 福建莫界文化发展有限公司 | Media content intelligent recommendation system based on artificial intelligence |
| CN116662659B (en) * | 2023-05-31 | 2024-05-14 | 广州泡芙传媒有限公司 | Media content intelligent recommendation system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022041979A1 (en) | Information recommendation model training method and related device | |
| US10642882B2 (en) | Media content discovery and character organization techniques | |
| CN107609152B (en) | Method and apparatus for expanding query expressions | |
| TWI636416B (en) | Method and system for multi-phase ranking for content personalization | |
| CN105426528B (en) | A kind of retrieval ordering method and system of commodity data | |
| WO2020048084A1 (en) | Resource recommendation method and apparatus, computer device, and computer-readable storage medium | |
| CN110543598B (en) | Information recommendation method and device and terminal | |
| US10180979B2 (en) | System and method for generating suggestions by a search engine in response to search queries | |
| US20220405607A1 (en) | Method for obtaining user portrait and related apparatus | |
| Qian et al. | Social media based event summarization by user–text–image co-clustering | |
| KR20160057475A (en) | System and method for actively obtaining social data | |
| CN106294500B (en) | Content item pushing method, device and system | |
| JP6767342B2 (en) | Search device, search method and search program | |
| CN111651678B (en) | Personalized recommendation method based on knowledge graph | |
| CN110069713B (en) | Personalized recommendation method based on user context perception | |
| Zhao et al. | On video recommendation over social network | |
| CN114564653A (en) | Information recommendation method and device, server and storage medium | |
| Zhang et al. | Discovering consumers’ purchase intentions based on mobile search behaviors | |
| Sun | Music individualization recommendation system based on big data analysis | |
| CN113254513B (en) | Sequencing model generation method, sequencing device and electronic equipment | |
| Amali et al. | Semantic video recommendation system based on video viewers impression from emotion detection | |
| Kim et al. | TrendsSummary: a platform for retrieving and summarizing trendy multimedia contents | |
| CN113495985A (en) | Article recommendation method and device, short video recommendation method and device and server | |
| Liu et al. | Item life cycle based collaborative filtering | |
| US20240152512A1 (en) | Machine learning for dynamic information retrieval in a cold start setting |
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 |