CN108694171B - Information pushing method and device - Google Patents
Information pushing method and device Download PDFInfo
- Publication number
- CN108694171B CN108694171B CN201710217063.6A CN201710217063A CN108694171B CN 108694171 B CN108694171 B CN 108694171B CN 201710217063 A CN201710217063 A CN 201710217063A CN 108694171 B CN108694171 B CN 108694171B
- Authority
- CN
- China
- Prior art keywords
- information
- service
- user
- determining
- users
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides an information pushing method and device, relates to the technical field of information processing, and can provide information submission precision and user coverage to a certain extent. The embodiment of the invention can establish an analysis model by analyzing the service number public information and the user characteristics, and select the information to be pushed by matching the model with the user behavior. The embodiment of the invention is suitable for the information pushing process.
Description
[ technical field ] A method for producing a semiconductor device
The present invention relates to the field of information processing technologies, and in particular, to a method and an apparatus for pushing information.
[ background of the invention ]
With the gradual development of internet technology, more and more information is flooded in front of users. In order to enable users to better acquire the information they want. The various service providers can push information, such as advertisement push, game recommendation, and the like, in a targeted manner. The current scheme for information push mainly comprises: recommending according to the popularity or recommending according to the user interest. The popularity recommendation means that popular and high-quality information is ranked and arranged at a position close to the pushing priority according to statistics such as user attention, goodness, attention amount and the like of the current information; the interest recommendation is to obtain interest categories which are interested by the user according to browsing and using history of various information of the user and perform priority recommendation.
According to the popularity recommendation, all kinds of information can only push the information of the appointed category to all people; and according to the careful pushing of the historical behaviors of the users, only part of the users capable of collecting the historical behaviors can be pushed. Therefore, the current information push method has certain limitation in the range of pushing users.
[ summary of the invention ]
In view of this, the embodiment of the present invention provides an information pushing method, which can provide information submission accuracy and user coverage to a certain extent.
In one aspect, an embodiment of the present invention provides an information pushing method, including:
collecting the characteristic information of the hotspot service numbers concerned by the users respectively;
according to the characteristic information, determining characteristic vectors corresponding to registered users and unregistered users related to the specified service respectively;
fitting according to the characteristic vector, and determining a preference degree model corresponding to the specified service;
processing the user data to be processed according to the preference model to obtain a pushing user;
and pushing information related to the specified service to the pushing user.
The foregoing aspect and any possible implementation manner further provide an implementation manner, where the determining, according to the feature information, feature vectors corresponding to registered users and unregistered users related to the specified service respectively includes:
determining a characteristic vector corresponding to each user according to the characteristic information;
screening out users with the use frequency of the specified service meeting specified conditions from the users with the determined feature vectors to serve as registered users;
screening out users not using the specified service from the users with the determined feature vectors to serve as non-registered users;
and determining feature vectors corresponding to registered users and non-registered users related to the specified service from the determined feature vectors.
The foregoing aspect and any possible implementation manner further provide an implementation manner, where the determining, according to the feature information, feature vectors corresponding to respective users includes:
determining original feature vectors corresponding to the users according to the feature information;
determining an adjustment weight according to interaction information between a user and a service number concerned by the user;
and adjusting the original characteristic vector according to the adjusting weight to obtain the characteristic vector corresponding to each user.
The above-mentioned aspect and any possible implementation manner further provide an implementation manner, before the acquiring feature information of the hotspot service numbers concerned by the users, further includes:
collecting characteristic information of a plurality of service numbers;
matching the characteristic information of the plurality of service numbers with a preset keyword to obtain the association degree information corresponding to the characteristic information of the plurality of service numbers;
and processing according to the association degree information to determine the hotspot service number.
The above aspect and any possible implementation manner further provide an implementation manner, where the acquiring the feature information of the plurality of service numbers includes:
acquiring text information of the plurality of service numbers, wherein the text information comprises service number identification information and/or service number description information;
and performing word segmentation processing according to the text information to obtain a word set corresponding to the text information, wherein the word set is used as the characteristic information of the service number.
As to the above-mentioned aspects and any possible implementation manner, there is further provided an implementation manner, before the matching the feature information of the plurality of service numbers with a preset keyword, the method further includes:
collecting a first word frequency corresponding to an original keyword from a keyword irrelevant corpus;
collecting a second word frequency corresponding to the original keyword from the keyword related corpus;
calculating according to the first word frequency and the second word frequency to obtain a polysemous word indicating parameter corresponding to the original keyword;
when the polysemous word indicating parameters meet specified conditions, determining the original keywords as preset keywords;
and when the polysemous word indicating parameters do not meet the specified conditions, combining the original keywords with the expanded keywords to obtain preset keywords.
On the other hand, an embodiment of the present invention further provides an information pushing apparatus, where the apparatus includes:
the first acquisition unit is used for acquiring the characteristic information of the hotspot service numbers concerned by the users respectively;
a first determining unit, configured to determine, according to the feature information, feature vectors corresponding to registered users and unregistered users related to a specified service;
the second determining unit is used for fitting according to the characteristic vector and determining a preference degree model corresponding to the specified service;
the first processing unit is used for processing the user data to be processed according to the preference degree model to obtain a pushing user;
and the pushing unit is used for pushing the information related to the specified service to the pushing user.
The above-described aspect and any possible implementation further provide an implementation, where the first determining unit includes:
the first determining module is used for determining the characteristic vector corresponding to each user according to the characteristic information;
the first screening module is used for screening out users with the use frequency of the specified service meeting specified conditions from the users with the determined feature vectors to serve as registered users;
the second screening module is used for screening out users which do not use the specified service from the users with the determined characteristic vectors to serve as non-registered users;
and the second determining module is used for determining the characteristic vectors corresponding to the registered users and the unregistered users related to the specified service from the determined characteristic vectors.
The above-described aspect and any possible implementation further provide an implementation, where the first determining module includes:
the first determining submodule is used for determining the original characteristic vector corresponding to each user according to the characteristic information;
the second determining submodule is used for determining an adjusting weight according to the interactive information between the user and the service number concerned by the user;
and the adjusting submodule is used for adjusting the original characteristic vector according to the adjusting weight value to obtain the characteristic vector corresponding to each user.
The above-described aspects and any possible implementation further provide an implementation, where the apparatus further includes:
the second acquisition unit is used for acquiring the characteristic information of a plurality of service numbers;
the matching unit is used for matching the characteristic information of the plurality of service numbers with a preset keyword to obtain the association degree information corresponding to the characteristic information of the plurality of service numbers;
and the second processing unit is used for processing according to the association degree information and determining the hotspot service number.
The above-described aspects and any possible implementations further provide an implementation, where the second acquisition unit includes:
the acquisition module is used for acquiring the text information of the plurality of service numbers, and the text information comprises service number identification information and/or service number description information;
and the word segmentation module is used for carrying out word segmentation processing according to the text information to obtain a word set corresponding to the text information to be used as the characteristic information of the service number.
The above-described aspects and any possible implementation further provide an implementation, where the apparatus further includes:
the third acquisition unit is used for acquiring a first word frequency corresponding to the original keyword from the keyword irrelevant corpus;
the fourth acquisition unit is used for acquiring a second word frequency corresponding to the original keyword from the keyword related corpus;
the calculation unit is used for calculating according to the first word frequency and the second word frequency to obtain a polysemous word indicating parameter corresponding to the original keyword;
the combination unit is used for determining the original keyword as a preset keyword when the polysemous word indicating parameter meets a specified condition; and when the polysemous word indicating parameters do not meet the specified conditions, combining the original keywords with the expanded keywords to obtain preset keywords.
According to the information pushing method and device provided by the embodiment of the invention, the service number public information and the user characteristics are analyzed to establish an analysis model, and the information to be pushed is selected by matching the user behavior through the model. Compared with the prior art that the push information of the known user can be determined only by analyzing the historical data, the method provided by the embodiment of the invention can push relatively accurate information for both the old user and the newly added user, and the coverage range of the user is larger.
[ description of the drawings ]
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.
Fig. 1 is a flowchart of a method for pushing information according to an embodiment of the present invention;
FIG. 2 is a flow chart of another method for pushing information according to an embodiment of the present invention;
FIG. 3 is a flow chart of another method for pushing information according to an embodiment of the present invention;
FIG. 4 is a flow chart of another method for pushing information according to an embodiment of the present invention;
FIG. 5 is a flow chart of another method for pushing information according to an embodiment of the present invention;
FIG. 6 is a flow chart of another method for pushing information according to an embodiment of the present invention;
fig. 7 is a block diagram illustrating an information pushing apparatus according to an embodiment of the present invention;
fig. 8 is a block diagram of another information pushing apparatus according to an embodiment of the present invention;
fig. 9 is a block diagram of another information pushing apparatus according to an embodiment of the present invention;
fig. 10 is a block diagram of another information pushing apparatus provided in the embodiment of the present invention;
fig. 11 is a block diagram of another information pushing apparatus according to an embodiment of the present invention;
fig. 12 is a block diagram of another information pushing apparatus according to an embodiment of the present invention.
[ detailed description ] embodiments
For better understanding of the technical solutions of the present invention, the following detailed descriptions of the embodiments of the present invention are provided with reference to the accompanying drawings.
It should be understood that the described embodiments are only some embodiments of the invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the prior art, the information pushing mode based on heat cannot accurately reach the information concerned by each user, so that the information pushing can only cover part of crowds, and the information pushing mode based on the historical behaviors of the users is only practical for users with more historical behaviors and cannot cover the users entering for the first time and the users with less historical behaviors. In order to solve the problems in the prior art, embodiments of the present invention provide an information push method for modeling based on user behavior characteristics, which can be implemented by a service number management server, and simultaneously, in order to acquire accurate service number related information and push the information to a user, the flow provided by embodiments of the present invention also relates to communication with a third-party server operating a service number and communication with a user terminal.
The main flow is shown in fig. 1 and comprises:
101. and collecting the characteristic information of the hotspot service numbers concerned by the users respectively.
The service number provides an external open account of the network service for the enterprise to the user, various information such as videos, characters, network links and the like can be added to the account according to the fact that the enterprise actually is under the account, and the user can access the service number to obtain various information under the account.
In the embodiment of the invention, the hotspot service number is a service number with a certain user coverage rate or related to certain specific service, and has the properties of larger access amount, higher user attention, correlation with specified service and the like.
In the embodiment of the present invention, the feature information is information derived from information published to the public by a service number, for example, text information extracted from various information such as a service number name, a service number description, and public information according to a specified manner, and may be represented as an article, a paragraph, a phrase, a single word, and the like.
102. And determining the characteristic vectors corresponding to the registered users and the unregistered users related to the specified service according to the characteristic information.
The designated service refers to services operated by various service providers, including games, social contact, commodity sales, and the like, and can be embodied in the user side through a web means or an application program and the like.
The registered user is a user who pays attention to a certain service number and uses a specific service related to the service number in the original data for establishing the user model. This is understood to mean that the user of the application is simultaneously interested in the service number associated with the application, or the purchaser of the goods is simultaneously interested in the service number associated with the goods, and so on.
And the non-registered user means a user who does not use a specified service related to a service number although paying attention to the service number in the raw data for building the user model. In this case, it is understood that a user pays attention to a service number related to an application but does not use the application or register an account with the application, or a user pays attention to a service number related to a product but does not purchase or use the product, and so on.
The feature vectors of the above various users can be obtained in this way: integrating various text information corresponding to the service number concerned by the user, and expressing an integration result into a vector form through a text processing method, such as NLP methods of a word bag model, an LDA theme model, a word vector model and the like.
103. And fitting according to the characteristic vector, and determining a preference degree model corresponding to the specified service.
The preference model can be established in a machine learning manner.
104. And processing the user data to be processed according to the preference model to obtain a push user.
The processing process can be completed by means of linear models, or tree classifiers such as decision trees, or vector machine decisions, and the like.
105. And pushing information related to the specified service to the pushing user.
Additionally, step 101 may be implemented by the service number management server by calling information from a third party server. Step 105 can be implemented by a way that the service number management server directly pushes to the user terminal, and the remaining steps 102 to 104 can be implemented by only the service number management server, and of course, if the specific process of executing steps 102 to 104 involves a process that needs to interact with a third-party server or the user terminal, the embodiment of the present invention does not limit that all processes must be completed at the local end of the service number management server, and a part of processes can also be implemented by other terminals or servers.
According to the information pushing method provided by the embodiment of the invention, the service number public information and the user characteristics are analyzed to establish an analysis model, and the information to be pushed is selected by matching the user behavior through the model. Compared with the prior art that the push information of the known user can be determined only by analyzing the historical data, the method provided by the embodiment of the invention can push relatively accurate information for both the old user and the newly added user, and the coverage range of the user is larger.
Further, the embodiment of the present invention provides the following procedure for implementing, in step 102, determining, according to the feature information, feature vectors corresponding to registered users and unregistered users related to a specified service, where as shown in fig. 2, the procedure is executed in a service number management server, and includes:
201. and determining the characteristic vector corresponding to each user according to the characteristic information.
202. And screening out users with the use frequency of the specified service meeting specified conditions from the users with the determined feature vectors as registered users.
The determination of whether the user is a registered user or a non-registered user can be confirmed by the frequency of use of a service related to a certain service number (e.g., advertisement viewing or click of pushed advertisement). Screening out users with the use frequency meeting specified conditions for the specified service from the concerned users as registered users; and screening out users which do not use the specified service from the concerned users as non-registered users.
203. And screening out users not using the specified service from the users with the determined feature vectors as non-registered users.
204. And determining feature vectors corresponding to registered users and non-registered users related to the specified service from the determined feature vectors.
Since the feature vectors can be determined for all users in step 201, the step 204 only needs to be performed for the two types of people, namely, the registered users and the unregistered users.
Further, in order to improve the effectiveness of the screening result of the specified service number, for the step 201 of determining the implementation of the feature vector corresponding to each user according to the feature information, the following procedure is provided, as shown in fig. 3, executed in the service number management server, and includes:
301. and determining the original characteristic vectors corresponding to the users according to the characteristic information.
The original feature vector is an initial feature vector directly obtained after integration of various information corresponding to the service number concerned by the user.
302. And determining an adjustment weight according to the interactive information between the user and the service number concerned by the user.
The interactive information can be understood as the number of messages left, the number of praise or the frequency of the user in the service number, and the information can reflect the attention degree of the user to the service number, so that the service number with higher attention degree can better reflect the preference degree of the user, and the corresponding weight value can be set to be larger. Otherwise, it can be set smaller.
In addition, the weight value can also comprise various information such as service number quality, service number credit, service number fan number and the like for configuration.
303. And adjusting the original characteristic vector according to the adjusting weight to obtain the characteristic vector corresponding to each user.
Further, considering that not all the information provided by the service number is related to the information to be pushed, some service numbers related to the specified service may be selectively screened out as a basis for pushing the information, and therefore, before step 101 is executed and the feature information of the hotspot service number concerned by the user is collected, as shown in fig. 4, executed in the service number management server, the embodiment of the present invention further includes the following procedures:
401. characteristic information of a plurality of service numbers is collected.
402. And matching the characteristic information of the plurality of service numbers with a preset keyword to obtain the association degree information corresponding to the characteristic information of the plurality of service numbers.
The preset keywords are information used for user selection, or related information of information pushed to the user. For example, to select a user who uses a certain product, the preset keywords may be related to the characteristics of the product or to the characteristics of the user. Or, for example, to push an advertisement for a certain product to a user, the preset keywords may be related to the characteristics of the product or to potential consumers using the product.
403. And processing according to the association degree information to determine the hotspot service number.
The processing mode can be configured to select a certain number of service numbers with the relevance degree ranked higher than the highest ranking. Other processing means may of course be provided.
In addition, considering that not all the information provided by the service numbers has a larger audience, some service numbers with high attention or larger user browsing amount can be selectively screened out to serve as the basis for information pushing, and the corresponding hotspot service number processing flow can be used for collecting the user attention information of a plurality of service numbers; and selecting a service number with the user attention exceeding the attention threshold from the plurality of service numbers as a hotspot service number based on the user attention information.
The user attention information may include various information that may reflect the user's service condition to the service number, such as the number of user subscriptions, the amount of attention, the amount of browsing, and so on.
The attention threshold is a condition for the user to filter the hotspot service number, and may be configured as a user subscription number, an attention amount, a browsing amount, and the like, and specific parameter values may be set according to actual needs, which is not limited in the embodiment of the present invention.
Further, the embodiment of the present invention provides the following process for implementing the step 401 and collecting the feature information of a plurality of service numbers, as shown in fig. 5, where the process is executed in a service number management server, and includes:
501. and acquiring text information of the plurality of service numbers, wherein the text information comprises service number identification information and/or service number description information.
The text information may include service number identification information, service number description information, and the like, such as a service number name, a service number content introduction, and the like. The information can be directly obtained through the content disclosed to the public by the service number.
502. And performing word segmentation processing according to the text information to obtain a word set corresponding to the text information, wherein the word set is used as the characteristic information of the service number.
The word set can be obtained by performing reverse indexing on the text information. Inverted indexing is an indexing method in which a search process looks up records based on attribute values. In the method, each entry in the index table includes an attribute value and an address of each record having the attribute value. Since the attribute value is not determined by the record but the position of the record is determined by the attribute value, it is called an inverted index. In the embodiment of the invention, each word or word combination obtained by word segmentation processing is used for searching the attributive text information, and then the service number is determined.
In addition, since embodiments of the present invention relate to text information searching, recognition may involve a certain amount of semantic equipment and keyword matching. In order to improve the information pushing and service number matching accuracy, the embodiment of the present invention further identifies the ambiguous word condition to determine a reasonable preset keyword, and the implementation flow thereof is as shown in fig. 6, and the step 402 is executed to match the feature information of the plurality of service numbers with the preset keyword to obtain the association degree information corresponding to each of the feature information of the plurality of service numbers, where the association degree information includes: before the matching of the feature information of the service numbers with the preset keywords, the method is executed in a service number management server, and the method further comprises the following steps:
601. and acquiring a first word frequency corresponding to the original keyword from the keyword irrelevant corpus.
The original keywords generally refer to keywords that are initially set for matching, but these keywords may have meanings different from the meanings of the original in different contexts. For example, if the original keyword is apple, but the search intent is apple products, then the concept of apple as a fruit will make the term apple a ambiguous word as defined in the embodiments of the present invention. Such terms must be adjusted.
Keyword independent corpora generally refer to corpora in which the meaning expressed by the original keyword differs by a certain amount from the semantics used in searching for matches.
In an embodiment of the invention, the first word frequency is recordable FaThe keyword irrelevant corpora is the a corpus. The calculation method is as follows:
602. and collecting a second word frequency corresponding to the original keyword from the keyword related linguistic data.
Corresponding to keyword independent corpora, keyword dependent corpora generally refer to corpora in which the meaning expressed by the original keyword is substantially the same as the semantic meaning used in search matching.
Accordingly, the second word frequency is recordable FbThe keyword irrelevant corpora are b corpora. The calculation method is as follows:
603. and calculating according to the first word frequency and the second word frequency to obtain a polysemous word indicating parameter corresponding to the original keyword. When the polysemous word indication parameter meets the specified condition, executing step 604; otherwise, step 605 is executed.
The polysemous word indicating parameter can be expressed as
When the value is not less than a certain value, it can be understood that there is no ambiguity, that is, a specified condition is satisfied. When the value is larger than a certain value, it can be understood that there is ambiguity, that is, the specified condition is not satisfied.
604. And determining the original keyword as a preset keyword.
605. And combining the original keywords and the expanded keywords to obtain preset keywords.
Extended keywords are typically used in combination with the original keywords to ensure that unnecessary ambiguities are not generated. For example, if the original keyword is apple, but the search intent is apple products, then the concept of apple as a fruit will make the term apple a ambiguous word as defined in the embodiments of the present invention. At this time, if the extended keywords such as "company", "computer", "mobile phone" are added, unnecessary ambiguity can be avoided.
Based on the foregoing method flow, an embodiment of the present invention provides an information pushing apparatus, which is shown in fig. 7 and includes:
the first collecting unit 71 is configured to collect feature information of a hotspot service number that a user pays attention to.
A first determining unit 72, configured to determine, according to the feature information, feature vectors corresponding to registered users and unregistered users related to the specified service.
And a second determining unit 73, configured to perform fitting according to the feature vector, and determine a preference model corresponding to the specified service.
And the first processing unit 74 is configured to process the user data to be processed according to the preference model, so as to obtain the pushing user.
A pushing unit 75, configured to push information related to the specified service to the pushing user.
Optionally, as shown in fig. 8, the first determining unit 72 includes:
a first determining module 721, configured to determine, according to the feature information, a feature vector corresponding to each user.
The first filtering module 722 is configured to filter out, from the users whose feature vectors have been determined, users whose usage frequency of the specified service meets a specified condition, as registered users.
And a second filtering module 723, configured to filter out, from the users with determined feature vectors, users who do not use the specified service as non-registered users.
A second determining module 724, configured to determine, from the determined feature vectors, feature vectors corresponding to registered users and unregistered users related to the specified service, respectively.
Optionally, as shown in fig. 9, the first determining module 721 includes:
the first determining submodule 7211 is configured to determine, according to the feature information, original feature vectors corresponding to the users, respectively.
The second determining sub-module 7212 is configured to determine an adjustment weight according to interaction information between the user and the service number concerned by the user.
The adjusting submodule 7213 is configured to adjust the original feature vector according to the adjustment weight value, so as to obtain a feature vector corresponding to each user.
Optionally, as shown in fig. 10, the apparatus further includes:
and a second collecting unit 76 for collecting characteristic information of a plurality of service numbers.
A matching unit 77, configured to match the feature information of the service numbers with a preset keyword, so as to obtain association degree information corresponding to the feature information of the service numbers.
And the second processing unit 78 is configured to perform processing according to the association degree information to determine a hotspot service number.
Optionally, as shown in fig. 11, the second collecting unit 76 includes:
the collection module 761 is configured to collect text information of the plurality of service numbers, where the text information includes service number identification information and/or service number description information.
A word segmentation module 762, configured to perform word segmentation processing according to the text information to obtain a word set corresponding to the text information, where the word set is used as feature information of a service number.
Optionally, as shown in fig. 12, the apparatus further includes:
the third collecting unit 79 is configured to collect the first word frequency corresponding to the original keyword from the keyword irrelevant corpus.
The fourth collecting unit 80 is configured to collect the second word frequency corresponding to the original keyword from the keyword related corpus.
And the calculating unit 81 is configured to calculate according to the first word frequency and the second word frequency to obtain a polysemous word indicating parameter corresponding to the original keyword.
A combining unit 82, configured to determine the original keyword as a preset keyword when the ambiguous word indicating parameter satisfies a specified condition. And when the polysemous word indicating parameters do not meet the specified conditions, combining the original keywords with the expanded keywords to obtain preset keywords.
The information pushing device provided by the embodiment of the invention establishes an analysis model by analyzing the service number public information and the user characteristics, and selects the information to be pushed by matching the model with the user behavior. Compared with the prior art that the push information of the known user can be determined only by analyzing the historical data, the method provided by the embodiment of the invention can push relatively accurate information for both the old user and the newly added user, and the coverage range of the user is larger.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions in actual implementation, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) or a Processor (Processor) to execute some steps of the methods according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (11)
1. A method of information push, the method comprising:
collecting characteristic information of a plurality of service numbers;
matching the characteristic information of the plurality of service numbers with a preset keyword to obtain the association degree information corresponding to the characteristic information of the plurality of service numbers;
processing according to the association degree information to determine a hotspot service number;
collecting the characteristic information of the hotspot service numbers concerned by the users respectively;
integrating the characteristic information corresponding to the service number concerned by each user according to the characteristic information, expressing the integrated result into a vector form by a text processing method, and determining the characteristic vectors corresponding to the registered users and the unregistered users related to the specified service;
fitting according to the characteristic vector, and determining a preference degree model corresponding to the specified service;
processing the user data to be processed according to the preference model to obtain a pushing user;
and pushing information related to the specified service to the pushing user.
2. The method of claim 1, wherein determining feature vectors corresponding to registered users and unregistered users associated with the specified service according to the feature information comprises:
determining a characteristic vector corresponding to each user according to the characteristic information;
screening out users with the use frequency of the specified service meeting specified conditions from the users with the determined feature vectors to serve as registered users;
screening out users not using the specified service from the users with the determined feature vectors to serve as non-registered users;
and determining feature vectors corresponding to registered users and non-registered users related to the specified service from the determined feature vectors.
3. The method according to claim 2, wherein the determining, according to the feature information, a feature vector corresponding to each user comprises:
determining original feature vectors corresponding to the users according to the feature information;
determining an adjustment weight according to interaction information between a user and a service number concerned by the user;
and adjusting the original characteristic vector according to the adjusting weight to obtain the characteristic vector corresponding to each user.
4. The method of claim 1, wherein collecting the feature information of the plurality of service numbers comprises:
acquiring text information of the plurality of service numbers, wherein the text information comprises service number identification information and/or service number description information;
and performing word segmentation processing according to the text information to obtain a word set corresponding to the text information, wherein the word set is used as the characteristic information of the service number.
5. The method according to claim 1, wherein before the matching the feature information of the plurality of service numbers with the preset keyword, the method further comprises:
collecting a first word frequency corresponding to an original keyword from a keyword irrelevant corpus;
collecting a second word frequency corresponding to the original keyword from the keyword related corpus;
calculating according to the first word frequency and the second word frequency to obtain a polysemous word indicating parameter corresponding to the original keyword;
when the polysemous word indicating parameters meet specified conditions, determining the original keywords as preset keywords;
and when the polysemous word indicating parameters do not meet the specified conditions, combining the original keywords with the expanded keywords to obtain preset keywords.
6. An information pushing apparatus, the apparatus comprising:
the second acquisition unit is used for acquiring the characteristic information of a plurality of service numbers;
the matching unit is used for matching the characteristic information of the plurality of service numbers with a preset keyword to obtain the association degree information corresponding to the characteristic information of the plurality of service numbers;
the second processing unit is used for processing according to the association degree information and determining a hotspot service number;
the first acquisition unit is used for acquiring the characteristic information of the hotspot service numbers concerned by the users respectively;
the first determining unit is used for integrating the characteristic information corresponding to the service number concerned by each user according to the characteristic information, expressing the integration result into a vector form by a text processing method, and determining the characteristic vectors corresponding to the registered users and the unregistered users related to the specified service;
the second determining unit is used for fitting according to the characteristic vector and determining a preference degree model corresponding to the specified service;
the first processing unit is used for processing the user data to be processed according to the preference degree model to obtain a pushing user;
and the pushing unit is used for pushing the information related to the specified service to the pushing user.
7. The apparatus according to claim 6, wherein the first determining unit comprises:
the first determining module is used for determining the characteristic vector corresponding to each user according to the characteristic information;
the first screening module is used for screening out users with the use frequency of the specified service meeting specified conditions from the users with the determined feature vectors to serve as registered users;
the second screening module is used for screening out users which do not use the specified service from the users with the determined characteristic vectors to serve as non-registered users;
and the second determining module is used for determining the characteristic vectors corresponding to the registered users and the unregistered users related to the specified service from the determined characteristic vectors.
8. The apparatus of claim 7, wherein the first determining module comprises:
the first determining submodule is used for determining the original characteristic vector corresponding to each user according to the characteristic information;
the second determining submodule is used for determining an adjusting weight according to the interactive information between the user and the service number concerned by the user;
and the adjusting submodule is used for adjusting the original characteristic vector according to the adjusting weight value to obtain the characteristic vector corresponding to each user.
9. The apparatus of claim 6, wherein the second acquisition unit comprises:
the acquisition module is used for acquiring the text information of the plurality of service numbers, and the text information comprises service number identification information and/or service number description information;
and the word segmentation module is used for carrying out word segmentation processing according to the text information to obtain a word set corresponding to the text information to be used as the characteristic information of the service number.
10. The apparatus of claim 6, further comprising:
the third acquisition unit is used for acquiring a first word frequency corresponding to the original keyword from the keyword irrelevant corpus;
the fourth acquisition unit is used for acquiring a second word frequency corresponding to the original keyword from the keyword related corpus;
the calculation unit is used for calculating according to the first word frequency and the second word frequency to obtain a polysemous word indicating parameter corresponding to the original keyword;
the combination unit is used for determining the original keyword as a preset keyword when the polysemous word indicating parameter meets a specified condition; and when the polysemous word indicating parameters do not meet the specified conditions, combining the original keywords with the expanded keywords to obtain preset keywords.
11. A computer-readable storage medium having stored thereon software functional units comprising instructions for causing a computer device or processor to perform the method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710217063.6A CN108694171B (en) | 2017-04-05 | 2017-04-05 | Information pushing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710217063.6A CN108694171B (en) | 2017-04-05 | 2017-04-05 | Information pushing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108694171A CN108694171A (en) | 2018-10-23 |
| CN108694171B true CN108694171B (en) | 2021-10-19 |
Family
ID=63842574
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710217063.6A Active CN108694171B (en) | 2017-04-05 | 2017-04-05 | Information pushing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108694171B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111598638B (en) * | 2019-02-21 | 2023-11-07 | 北京沃东天骏信息技术有限公司 | Click rate determination method, device and equipment |
| CN110598472B (en) * | 2019-09-18 | 2022-03-22 | 广州虎牙科技有限公司 | Equipment identification method, device, server and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184618A (en) * | 2015-10-20 | 2015-12-23 | 广州唯品会信息科技有限公司 | Commodity individual recommendation method for new users and system |
| CN105787061A (en) * | 2016-02-29 | 2016-07-20 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Information pushing method |
| CN106294410A (en) * | 2015-05-22 | 2017-01-04 | 苏宁云商集团股份有限公司 | A kind of determination method of personalized information push time and determine system |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012204894A (en) * | 2011-03-24 | 2012-10-22 | Toshiba Corp | Information recommendation device |
| CN103176982B (en) * | 2011-12-20 | 2016-04-27 | 中国移动通信集团浙江有限公司 | The method and system that a kind of e-book is recommended |
| CN103885987B (en) * | 2012-12-21 | 2018-04-10 | 中国移动通信集团公司 | A kind of music recommends method and system |
| CN106250378B (en) * | 2015-06-08 | 2020-06-02 | 腾讯科技（深圳）有限公司 | Public identification classification method and device |
| CN105893398A (en) * | 2015-10-12 | 2016-08-24 | 乐视网信息技术（北京）股份有限公司 | Multimedia content recommendation method and device, user terminal equipment and server |
-
2017
- 2017-04-05 CN CN201710217063.6A patent/CN108694171B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106294410A (en) * | 2015-05-22 | 2017-01-04 | 苏宁云商集团股份有限公司 | A kind of determination method of personalized information push time and determine system |
| CN105184618A (en) * | 2015-10-20 | 2015-12-23 | 广州唯品会信息科技有限公司 | Commodity individual recommendation method for new users and system |
| CN105787061A (en) * | 2016-02-29 | 2016-07-20 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Information pushing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108694171A (en) | 2018-10-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108241667B (en) | Method and apparatus for pushed information | |
| WO2020048084A1 (en) | Resource recommendation method and apparatus, computer device, and computer-readable storage medium | |
| CN108694647B (en) | Method and device for mining merchant recommendation reason and electronic equipment | |
| WO2019056661A1 (en) | Search term pushing method and device, and terminal | |
| CN103714130B (en) | Video recommendation system and method | |
| CN107341268B (en) | Hot searching ranking method and system | |
| CN106339507B (en) | Streaming Media information push method and device | |
| CN108960945A (en) | Method of Commodity Recommendation and device | |
| CN111444304B (en) | Search ordering method and device | |
| CN108614832B (en) | Method and device for realizing user personalized commodity search | |
| CN110111167A (en) | A kind of method and apparatus of determining recommended | |
| JP5831204B2 (en) | Information providing system, information providing method, and program | |
| CN109241451B (en) | Content combination recommendation method and device and readable storage medium | |
| KR20100029581A (en) | Recommended search terms providing system and method for each user and computer readable medium processing the method | |
| CN110766486A (en) | Method and device for determining item category | |
| KR101596632B1 (en) | Method for recommanding product using big data image | |
| CN110516033A (en) | A kind of method and apparatus calculating user preference | |
| CN112182351A (en) | News recommendation method and device based on multi-feature fusion | |
| CN110489531B (en) | Method and device for determining high-frequency problem | |
| KR101712291B1 (en) | System for recommending a user-customized famous place based on opinion mining and Method of the Same | |
| CN108694171B (en) | Information pushing method and device | |
| CN107133811A (en) | The recognition methods of targeted customer a kind of and device | |
| CN115168700A (en) | Information flow recommendation method, system and medium based on pre-training algorithm | |
| CN111666513A (en) | Page processing method and device, electronic equipment and readable storage medium | |
| KR101621735B1 (en) | Recommended search word providing method and system |
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 |