CN108491540B - Text information pushing method and device and intelligent terminal - Google Patents
Text information pushing method and device and intelligent terminal Download PDFInfo
- Publication number
- CN108491540B CN108491540B CN201810286984.2A CN201810286984A CN108491540B CN 108491540 B CN108491540 B CN 108491540B CN 201810286984 A CN201810286984 A CN 201810286984A CN 108491540 B CN108491540 B CN 108491540B
- Authority
- CN
- China
- Prior art keywords
- interest
- text information
- pushing
- user
- target
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Abstract
The application provides a text information pushing method, a text information pushing device and an intelligent terminal, wherein the method comprises the following steps: calculating the interest correlation degree of each candidate text message and the interest point of the target user; calculating the interest coverage of each candidate text message according to the interest coverage calculation model; scoring the pushing value of each candidate text message, and storing the candidate text message with the highest pushing value as a target text message into a text message pushing set; updating the operation factor of the interest coverage calculation model according to the interest coverage of the target text information which is newly stored in the text information push set; scoring the pushing value of each residual candidate text message in the candidate text message set according to the updated interest coverage calculation model, and storing the candidate text message with the highest pushing value as a target text message in the text message pushing set; and pushing the text information push set to the target user. Therefore, the diversity of information push is improved, and the viscosity of users and products is increased.
Description
Technical Field
The application relates to the technical field of information pushing, in particular to a text information pushing method and device and an intelligent terminal.
Background
Generally, in order to recommend text information meeting the interest points of users from a large amount of text information on the internet, a recommendation system needs to meet different requirements of different users to achieve the effect of thousands of people.
In the related technology, the recommendation system is divided into a recall stage and a sorting stage, the recall stage roughly recalls text information in modes of user preference information and the like, a candidate set of the recall stage is generally in the million level, a score between a user and the text information is calculated for the recalled text information in the sorting stage, and the score can be used as a basis for finally outputting a sorting result.
However, the ranking method in the related art scores only according to the interest preferences of the users, and although the accuracy is relatively high in terms of a single recommendation result, the diversity cannot be guaranteed in terms of the overall recommendation result of each user. Taking a scene of recommending text information as an example of news recommendation, if a target user likes to see news related to "plum", from the perspective of a single news, scores of all news related to "plum" are inevitably higher, and therefore a large amount of news including "plum" is recommended to the user, and diversity is poor.
Content of application
The present application is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, a first objective of the present application is to provide a text information pushing method, which improves the diversity of information pushing and increases the stickiness of users and products.
A second object of the present application is to provide a text information pushing apparatus. A third objective of the present application is to provide an intelligent terminal.
A fourth object of the present application is to propose a non-transitory computer-readable storage medium.
A fifth object of the present application is to propose a computer program product.
In order to achieve the above object, an embodiment of a first aspect of the present application provides a text information pushing method, including the following steps: acquiring a candidate text information set; calculating the interest correlation degree of each candidate text message in the candidate text message set and the interest point of the target user according to a preset user interest point correlation model and a text interest point correlation model; calculating the interest coverage of all target text information in a text information push set when each candidate text information is respectively used as the target text information and added into the text information push set according to a preset interest coverage calculation model; scoring a pushing value of each candidate text message according to the interest correlation degree and the interest coverage degree, and storing the candidate text message with the highest pushing value as a target text message into the text message pushing set; updating the operation factor of the interest coverage calculation model according to the interest coverage of the target text information which is newly stored in the text information pushing set; scoring the pushed value of each remaining candidate text message in the candidate text message set according to the preset user interest point association model, the text interest point association model and the updated interest coverage calculation model, and storing the candidate text message with the highest pushed value as a target text message in a text message pushed set; and pushing the text information push set to the target user.
The text information pushing method can push the text information based on the diversity of the interest points on the basis of ensuring the accuracy of the recommendation result, and can effectively ensure the diversity of the pushing result, so that the personalized experience of the user is met, and the stickiness of the user and the product is improved.
In order to achieve the above object, a second aspect of the present application provides a text information pushing apparatus, including: the acquisition module is used for acquiring a candidate text information set; the calculation module is used for calculating the interest correlation degree of each candidate text message in the candidate text message set and the interest point of the target user according to a preset user interest point correlation model and a text interest point correlation model; the calculation module is further used for calculating the interest coverage of all target text information in a text information push set when each candidate text information is respectively used as the target text information to be added into the text information push set according to a preset interest coverage calculation model; the processing module is used for scoring the pushing value of each candidate text message according to the interest correlation degree and the interest coverage degree, and storing the candidate text message with the highest pushing value as a target text message into a text message pushing set; the updating module is used for updating the operation factors of the interest coverage calculation model according to the interest coverage of the target text information which is newly stored in the text information pushing set; the processing module is further configured to score the pushed value of each remaining candidate text message in the candidate text message set according to the preset user interest point association model, the text interest point association model and the updated interest coverage calculation model, and store the candidate text message with the highest pushed value as a target text message in a text message pushed set; and the pushing module is used for pushing the text information pushing set to the target user.
The text information pushing device provided by the embodiment of the application can push the text information based on the diversity of the interest points on the basis of ensuring the accuracy of the recommendation result, and can effectively ensure the diversity of the pushing result, so that the personalized experience of a user is met, and the stickiness of the user and a product is improved. In order to achieve the above object, an embodiment of a third aspect of the present application provides an intelligent terminal, including: a processor and a memory; wherein the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the text information pushing method described in the above embodiments.
In order to achieve the above object, a fourth aspect of the present application provides a non-transitory computer-readable storage medium, where the program is executed by a processor to implement the text information pushing method as described in the above embodiments.
In order to achieve the above object, a fifth aspect of the present application provides a computer program product, where when executed by an instruction processor, the computer program product executes a text information pushing method as described in the above embodiments.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart of a text information pushing method according to an embodiment of the present application;
FIG. 2 is a diagram illustrating an exemplary scenario for updating an operation factor according to an embodiment of the present application;
fig. 3 is a flowchart of a text information pushing method according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of a text information pushing apparatus according to an embodiment of the present application; and
FIG. 5 is a block diagram of an exemplary computer device implementing embodiments of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.
The text information pushing method, the text information pushing device and the intelligent terminal according to the embodiment of the application are described below with reference to the drawings.
The text information in the embodiment of the application may include various recommended contents that can be implemented in a recommendation system, such as news information, book recommendation information, and the like.
Fig. 1 is a flowchart of a text information pushing method according to an embodiment of the present application. As shown in fig. 1, the method includes:
Specifically, the determination manner of the candidate text information included in the candidate text information set is related to a recall manner of the recommendation system, and in some possible examples, the candidate text information may be recalled according to a geographic location, so that the candidate text information included in the candidate text information set is all matched with the location of the target user, for example, for a user in beijing, news in the candidate text information set obtained is news in beijing; in some possible examples, the text information may be retrieved according to personal preference information of the user, where the personal preference information of the user may be determined according to a historical browsing record of the user, and the like, so that the candidate text information included in the candidate text information set is consistent with the personal preference information of the user, for example, for a user who prefers sports, the text information included in the recommended candidate text information set includes sports, sports surroundings, and the like.
And 102, calculating the interest correlation degree of each candidate text message in the candidate text message set and the interest point of the target user according to a preset user interest point correlation model and a text interest point correlation model.
It can be understood that the user interest point association model is used for describing the interest degree of the user in the interest point, and the text interest point association model describes the satisfaction degree of the text information in the interest point, wherein the user interest point association model and the text interest point association model can be a matrix model, a neural network model and the like according to different application scenes.
Under different application scenes, different modes for constructing the user interest point association model and the text interest point association model are adopted, and as a possible implementation mode, user information is obtained, wherein the user information is related to the interest degree of a user in the interest points and can comprise one or more of search satisfaction feedback information, user portrait information and user social information, and the user portrait information comprises identity information, occupation information, age information, gender information and the like of the user.
Further, the user's interest level in a plurality of interest points is analyzed based on the user information, and in the case of the user portrait information, the user who is in the player status has the highest interest level in sports and interest points around the sports, a higher interest level in sports attention, a lower interest level in makeup interest points, and the like. And establishing a user interest point association model according to the interest degree of the user to the interest points, so that the interest degree of the target user to the interest points can be determined according to the information of the target user through the user interest point association model.
In the present example, a user _ attribution _ weight matrix is established according to the interest degree of the user in the plurality of interest points, wherein the interest degree of the user in the interest points is described in the matrix.
In this embodiment, text information is obtained, interest coverage of the text information to a plurality of interest points is analyzed according to the text information, for example, semantic analysis is performed on the text information, coverage of content and interest points included in the text information is analyzed, for example, coverage of content and interest points included in the text information is analyzed according to the number of keywords which appear in the text information and are related to the interest points, and then a text interest point association model is established according to the interest coverage of the text information to the plurality of interest points, so that the coverage of the text information to the interest points is obtained through an interest coverage calculation model according to information content, information sources and the like of candidate text information.
In the present example, a doc _ attribution _ weight matrix is established according to the interest degree of the user in the plurality of interest points, wherein the coverage degree of the text information on the interest points is described in the matrix.
Specifically, the interest relevance of each candidate text information in the candidate text information set and the interest point of the target user is calculated according to a preset user interest point relevance model and a preset text interest point relevance model, wherein the manner of obtaining the interest relevance can be different in different application scenes, the user interest point relevance model and the text interest point relevance model are taken as matrixes continuously, and the interest relevance is a product value of the user interest point relevance matrix and the text interest point relevance matrix.
It should be emphasized that the interest correlation between each candidate text message and the interest point of the target user in this embodiment represents the correlation between each candidate text message and the interest point of the target user in the interest point dimension, so as to ensure that the candidate text message finally pushed to the target user is relatively consistent with the interest point of the user, and ensure the accuracy of the pushed text message.
And 103, calculating the interest coverage of all target text information in the text information push set when each candidate text information is respectively used as the target text information and added into the text information push set according to a preset interest coverage calculation model.
The interest coverage degree represents the coverage degree of all target text information in the current text information pushing set to each interest point, the higher the interest coverage degree is, the more interest points are covered in all the target text information in the current text information pushing set, the more the coverage frequency of each interest point is, the lower the interest coverage degree is, the fewer the interest points are covered in all the target text information in the current text information pushing set, the fewer the coverage frequency of each interest point is, or the more the coverage frequency of individual interest points is, and the like.
It can be understood that, if the target text information finally pushed to the target user is determined only according to the interest correlation between each candidate text information in the candidate text information set and the interest point of the target user, it is obvious that the pushed target text information is too concentrated on the interest point with higher interest degree of the user, and the diversity of the pushed target text information is not enough, so in order to ensure the diversity of text information pushing, in the embodiment of the present application, a preset interest coverage calculation model is introduced, and when each candidate text information is calculated by the interest coverage calculation model to be added into the text information push set as the target text information, the interest coverage of all target text information in the text information push set is calculated, wherein all target text information in the current text information push set includes the previously determined target text information and the candidate text information to be scored, and considering the candidate text information in the candidate text information set according to the diversity dimension of the interest points.
For example, assuming that the current candidate text information set includes candidate text information 1 and 2, and the target text information determined in the text information pushing set is 3 and 4, the interest coverage of the target text information 1, 3, and 4 after the candidate text information 1 is taken as the target text information and added into the text information pushing set, and the interest coverage of the target text information 2, 3, and 4 after the candidate text information 2 is taken as the target text information and added into the text information pushing set are calculated.
And 104, scoring the pushing value of each candidate text message according to the interest relevance and the interest coverage, and storing the candidate text message with the highest pushing value as the target text message into a text message pushing set.
Specifically, the score of the pushed value is carried out on each candidate text message according to the interest relevance and the interest coverage, and then the candidate text message with the highest pushed value is stored into the text message pushing set as the target text message, so that the target text message added into the text message pushing set at present is ensured to refer to the diversity dimensionality of the interest points and the dimensionality of the interest points of the target user.
According to different application scenes, the push value scoring mode of each candidate text message according to the interest relevance and the interest coverage is different, and the following examples are given:
the first example:
in this example, the preset interest coverage calculation model, the user interest point association model, and the text interest point association model are all matrices, where the interest coverage calculation model is a T matrix, the user interest point association model is a user _ entry _ weight matrix, and the text interest point association model is a doc _ entry _ weight matrix, and then a product value of the three matrices may be scored as a push value, that is, the push value user _ doc _ score is user _ entry _ weight T _ doc _ entry _ weight.
The second example is:
in this example, the scoring mechanism of the interest relevance and the scoring mechanism of the interest coverage may be different, for example, if the scoring mechanism of the interest relevance is a percentage system and the scoring of the interest coverage is a ten-degree system, the scoring of the interest relevance and the scoring of the interest coverage may be performed after the interest relevance and the interest coverage are respectively calculated, and then the scoring of the interest relevance and the scoring of the interest coverage are respectively normalized, and the sum of the scores of the two is used as the scoring of the push value.
Of course, in this embodiment, in order to meet the requirements in different application scenarios, weighted values corresponding to the interest relevance and the interest coverage may be given, and after multiplying the interest relevance and the interest coverage by the corresponding weighted values, the sum of the corresponding multiplied values is used as the score of the push value.
And 105, updating the operation factor of the interest coverage calculation model according to the interest coverage of the target text information which is newly stored in the text information pushing set.
And 106, scoring the pushing value of each residual candidate text message in the candidate text message set according to a preset user interest point association model, a text interest point association model and an updated interest coverage calculation model, and storing the candidate text message with the highest pushing value as a target text message into the text message pushing set.
It should be understood that, considering the diversity of the text information push, it is not a simple consideration of the single pushed target text information, but a comprehensive consideration of all recommended target text information, and therefore, the interest coverage calculation model in the present application calculates the interest coverage, assuming that after each candidate text information is added to the corresponding text information push set, the coverage degree of each interest point is calculated according to all currently selected target text information.
For example, if all the currently selected target text messages have a high coverage degree for the points of interest a and B, and no coverage is performed for the points of interest C and D, then in the process of determining the next target text message, the coverage degree of interest of the text message push set is considered for one candidate text message 1 containing the point of interest A, B, C and another candidate text message 2 containing the points of interest C and D, respectively, after the candidate text message push set is added, at this time, although the candidate text message 1 covers more points of interest, since the covered points of interest A, B are already included in the selected target text message in a large amount, the coverage degree of interest obtained after the candidate text message 1 is added is low, although the candidate text message 2 covers less points of interest, since the covered points of interest C and D are not included in the selected target text message, therefore, the interest coverage obtained after the candidate text information 2 is added is high, and the candidate text information 2 may bring positive influence on the diversity of the text information.
Obviously, since the user interest point association model is determined according to the coverage degree of all currently selected target text information on each interest point, after the latest target text information is determined, the selected target text information is added with new target text information, and at this time, the operation factor of the interest coverage calculation model needs to be updated according to the interest coverage of the target text information newly stored in the text information push set, so as to improve the practicability and flexibility of the recommendation system.
And then, scoring the pushing value of each residual candidate text information in the candidate text information set according to a preset user interest point association model, a text interest point association model and an updated interest coverage calculation model, and storing the candidate text information with the highest pushing value as target text information into a text information pushing set, so that after each target text information is determined, the association model of the interest point of the user is updated according to the newly determined target text information, and the diversity contribution of the next determined target text information to the text information pushed to the user is ensured.
It should be noted that, according to different application scenarios, the interest coverage calculation model may be embodied in a form that any current text information that can be reflected is pushed to be centralized, and coverage of all target text information to the interest point:
as one possible implementation, the interest coverage calculation model is a T matrix, where the formula of the T matrix is T ═ ealpha*countIn some possible embodiments, the value of alpha is-0.32, the count is the number of times that the target text information stored in the text information pushing set is generalized to each interest point, and it can be known from a formula of the interest coverage calculation model that, as the count value increases, the exponential decay of the T matrix represents the degree of diversity deterioration, that is, when the number of times that the target text information stored in the text information pushing set is displayed to each interest point is higher, and the target text information is newly added, the interest coverage may not be improved, so that the diversity is worse.
In this embodiment, the count in the T matrix may be updated according to the interest coverage of the target text information newly stored in the text information push set.
Of course, in practical applications, considering that a user has a certain forgetting property for an interest point, for example, for a target user a, the interest point of a target text message 1 browsed by the user includes an interest point a, when the user browses a target text message 2 also including the interest point a, the interest point a browsed in the target text message 1 may be forgotten, at this time, when calculating the interest point coverage of the target text message 2, for more precise determination, such forgetting of the interest point should be considered, so when updating the operation factor of the interest coverage calculation model according to the interest coverage of the target text message newly stored in the text message push set, the method further includes: and obtaining an interest point forgetting factor, wherein the interest point forgetting factor is used for expressing the forgetting degree of the user to the browsed interest points, and updating the operation factor in the interest coverage calculation model according to the interest point forgetting factor and the interest coverage.
For the sake of clarity, the operation factor is still taken as the count, and in this example, the count can be expressed by the following formula:
count＝count*forgetting_factor+Δcount
in this example, the forgetting _ factor may be 0.98, which represents a forgetting degree of the user to see the interest point, and Δ count represents a number of times that the new target text information is converted into each interest point, where Δ count may be calculated by the following formula:
wherein i represents the ith interest point, j represents the interest point contained in the newly added target text information, and sim (i, j) represents the cosine similarity between the ith interest point and the jth interest point in the newly added target text information. For example, the vector of the number of presentations of each point of interest in the current text information push set is represented as { pony: 1, small treasure: 0, small song 0}, and the newly added target text information includes two interest points, namely a small horse and a small treasure, wherein Δ count in the implementation is calculated as shown in fig. 2, and a vector of the updated count is expressed as { small horse: 2, small treasure: 1, 0.7 Song minority, and the updated count considers the display times of the newly added target text on the interest points and influences on the diversity of the whole text information push set.
Specifically, the determined target text information in the text information pushing set is determined according to the interest correlation degree of each candidate text information and the interest point of the target user and the interest coverage degree of each candidate text information, and the determining mode considers the correlation between the interest point of the target user and the interest point dimension of the candidate text information on one hand and the diversity of the interest points of the whole candidate text set on the other hand, so that the text information in the text information pushing set comprises the target text information which accords with the interest point of the target user and the target text information which possibly does not accord with the interest point of the target user very much, therefore, the pushed text information pushing set not only meets the recommendation of the target text information which is more interesting for the target user, but also meets the target text information which can accord with the interest point of the target user very much on the basis of other target text information which does not accord with the interest point of the target user very much, other interest points of the target user are mined, so that the click rate of the target user on the text information and the user experience of text information pushing are improved, and the stickiness of the user and the product is increased.
In the actual execution process, the timing of pushing the text information push set to the target user may be when the current candidate text information set is empty, that is, all candidate text information in the candidate text information set is traversed, and all candidate text information is screened out item by item, and the timing of pushing the text information push set to the target user may also be when the number of screened target information texts meets a preset number, for example, for a recommendation system that recommends three pieces of text information at a time, when the number of screened target information texts is 3, the text information push set is pushed to the target user.
In addition, the form of pushing the text information push set to the target user also includes, but is not limited to, the following ways according to different application scenarios:
the first example:
and generating a text information recommendation list according to the stored sequence of the target text information in the text information pushing set, and pushing the text information recommendation list to the target user, so that the satisfaction degree of the target user on the text information pushing is improved.
The second example is:
and marking reading values of the target text information in the text information pushing set according to the stored sequence, wherein the reading values can be represented by numbers and can also be represented by grades, and the reading values of the target text information stored in the text information pushing set are higher, so that reading sequence indication is provided for a user according to the reading values, and browsing experience of the target user is improved.
In order to make the description of the text information pushing method of the embodiment of the present application clearer, the following description is made with reference to an embodiment in a specific application scenario:
in this example, the score of the push value is represented by user _ doc _ score in the above embodiment, the interest coverage calculation model is a T matrix, and the user interest point association model and the text interest point association model are a user _ attribution _ weight matrix and a doc _ attribution _ weight matrix, respectively.
Specifically, as shown in fig. 3, a user _ attribution _ weight matrix is established in advance, a candidate text information set is preliminarily determined according to the user _ attribution _ weight matrix, if the candidate text information set is not empty, a scoring user _ doc _ score of a recommendation value is calculated, the candidate text information with the highest scoring is stored in a text information push set, after a T matrix is updated, if the candidate text information set is not empty, the scoring user _ doc _ score of the recommendation values of the remaining candidate text information is continuously calculated according to the updated T matrix, the candidate text information with the highest scoring is added into the text information push set, and therefore, until the candidate text information set is empty, a text information recommendation list is generated according to a storage sequence by target text information in the text information push set, and the text information recommendation list is pushed to a target user.
In summary, the text information pushing method in the embodiment of the application can push the text information based on the diversity of the points of interest on the basis of ensuring the accuracy of the recommendation result, and can effectively ensure the diversity of the pushing result, thereby satisfying the personalized experience of the user and improving the stickiness of the user and the product.
In order to implement the above embodiments, the present application further provides a text information pushing apparatus.
Fig. 4 is a schematic structural diagram of a text information pushing apparatus according to an embodiment of the present application.
As shown in fig. 4, the text information pushing apparatus includes: an acquisition module 100, a calculation module 200, a processing module 300, an update module 400, and a push module 500.
The obtaining module 100 is configured to obtain a candidate text information set.
A calculating module 200, configured to calculate, according to a preset user interest point association model and a text interest point association model, an interest correlation between each candidate text information in the candidate text information set and an interest point of the target user.
In an embodiment of the present application, the calculating module 200 is further configured to calculate, according to a preset interest coverage calculating model, interest coverage of all target text information in a text information push set when each candidate text information is respectively added to the text information push set as the target text information.
And the processing module 300 is configured to score a push value of each candidate text message according to the interest relevance and the interest coverage, and store the candidate text message with the highest push value as the target text message into a text message push set.
And the updating module 400 is configured to update the operation factor of the interest coverage calculation model according to the interest coverage of the target text information newly stored in the text information push set.
In an embodiment of the present application, the processing module 300 is further configured to score a pushing value of each remaining candidate text information in the candidate text information set according to a preset user interest point association model, a text interest point association model, and an updated interest coverage calculation model, and store the candidate text information with the highest pushing value as the target text information into the text information pushing set.
The pushing module 500 is configured to push a text information push set to a target user.
It should be noted that the explanation of the embodiment of the text information pushing method is also applicable to the text information pushing apparatus of the embodiment, and details are not repeated here.
To sum up, the text information pushing device of the embodiment of the application can push the text information based on the diversity of the points of interest on the basis of ensuring the accuracy of the recommendation result, and can effectively ensure the diversity of the pushing result, thereby satisfying the personalized experience of the user and improving the viscosity of the user and the product.
In order to implement the above embodiment, the present application further provides an intelligent terminal, including: a processor, wherein the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory, for implementing the text information pushing method described in the above embodiments. The intelligent terminal can be a hardware device with various operating systems, such as a mobile phone, a tablet computer, a personal digital assistant and wearable equipment, and the wearable equipment can be an intelligent bracelet, an intelligent watch, intelligent glasses and the like.
In order to implement the above embodiments, the present application also proposes a non-transitory computer-readable storage medium, in which instructions are executed by a processor to enable execution of the text information pushing method shown in the above embodiments.
In order to implement the foregoing embodiments, the present application also proposes a computer program product, which when executed by an instruction processor in the computer program product, executes the text information pushing method shown in the foregoing embodiments.
FIG. 5 illustrates a block diagram of an exemplary computer device suitable for use in implementing embodiments of the present application. The computer device 12 shown in fig. 5 is only an example and should not bring any limitation to the function and scope of use of the embodiments of the present application.
As shown in FIG. 5, computer device 12 is in the form of a general purpose computing device. The components of computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that couples various system components including the system memory 28 and the processing unit 16.
Computer device 12 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.
A program/utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 42 generally perform the functions and/or methodologies of the embodiments described herein.
The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), with one or more devices that enable a user to interact with the computer system/server 12, and/or with any devices (e.g., network card, modem, etc.) that enable the computer system/server 12 to communicate with one or more other computing devices. Such communication may be through an input/output (I/O) interface 22. Moreover, computer device 12 may also communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public Network such as the Internet) via Network adapter 20. As shown, network adapter 20 communicates with the other modules of computer device 12 via bus 18. It should be understood that although not shown in the figures, other hardware and/or software modules may be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
The processing unit 16 executes various functional applications and data processing, for example, implementing the methods mentioned in the foregoing embodiments, by executing programs stored in the system memory 28.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (9)
1. A text information pushing method is characterized by comprising the following steps:
acquiring a candidate text information set;
calculating interest correlation degrees of each candidate text message in the candidate text message set and an interest point of a target user according to a preset user interest point correlation model and a preset text interest point correlation model, wherein the user interest point correlation model is used for describing the interest degree of the user on the interest point, and the text interest point correlation model is used for describing the satisfaction degree of the text message on the interest point;
calculating interest coverage of all target text information in a text information push set when each candidate text information is respectively used as the target text information and added into the text information push set according to a preset interest coverage calculation model, wherein the interest coverage represents the coverage degree of all the target text information in the current text information push set on each interest point;
scoring a pushing value of each candidate text message according to the interest correlation degree and the interest coverage degree, and storing the candidate text message with the highest pushing value as a target text message into the text message pushing set;
updating the operation factor of the interest coverage calculation model according to the interest coverage of the target text information which is newly stored in the text information pushing set;
scoring the pushed value of each remaining candidate text message in the candidate text message set according to the preset user interest point association model, the text interest point association model and the updated interest coverage calculation model, and storing the candidate text message with the highest pushed value as a target text message in a text message pushed set;
and pushing the text information push set to the target user.
2. The method of claim 1, wherein the interest coverage calculation model is a T matrix, wherein the formula of the T matrix is:
T＝ealpha*count
wherein, alpha is an attenuation coefficient, and count is the generalized display times of the target text information stored in the text information pushing set to each interest point;
the updating of the operation factor of the interest coverage calculation model according to the interest coverage of the target text information which is newly stored in the text information push set comprises the following steps:
and updating the count in the T matrix according to the interest coverage of the target text information which is newly stored in the text information push set.
3. The method of claim 1, wherein updating the operational factor of the interest coverage calculation model based on the interest coverage of the target text information most recently stored in the text information push set further comprises:
obtaining an interest point forgetting factor;
and updating the operation factor in the interest coverage calculation model according to the interest point forgetting factor and the interest coverage.
4. The method of claim 1, further comprising, before said calculating interest relevance of each candidate text message in the candidate text message set to the interest point of the target user according to a preset user interest point relevance model and a text interest point relevance model:
acquiring user information, and analyzing the interest degree of a user in a plurality of interest points according to the user information, wherein the user information comprises one or more of search satisfaction feedback information, user portrait information and user social information;
establishing a user interest point association model according to the interest degree of the user to a plurality of interest points;
acquiring text information, and analyzing interest coverage of the text information on a plurality of interest points according to the text information;
and establishing the text interest point association model according to the interest coverage of the text information to a plurality of interest points.
5. The method of claim 1, wherein said pushing the push set of textual information to the target user comprises:
judging whether the recommended number of the target text information in the text information pushing set reaches a preset number or not;
and if the preset number is reached, generating a text information recommendation list according to the sequence of the target text information stored, and pushing the text information recommendation list to the target user.
6. The method of claim 1, wherein said pushing the push set of textual information to the target user comprises:
judging whether the candidate text information set is empty or not;
and if the target text information is empty, pushing the target text information in the text information pushing set to the target user.
7. A text information pushing apparatus, comprising:
the acquisition module is used for acquiring a candidate text information set;
the calculation module is used for calculating the interest correlation degree of each candidate text information in the candidate text information set and the interest point of the target user according to a preset user interest point correlation model and a preset text interest point correlation model, wherein the user interest point correlation model is used for describing the interest degree of the user on the interest point, and the text interest point correlation model is used for describing the satisfaction degree of the text information on the interest point;
the calculation module is further configured to calculate, according to a preset interest coverage calculation model, interest coverage of all target text information in a text information push set when each candidate text information is respectively added to the text information push set as the target text information, where the interest coverage represents coverage of all target text information in a current text information push set to each interest point;
the processing module is used for scoring the pushing value of each candidate text message according to the interest correlation degree and the interest coverage degree, and storing the candidate text message with the highest pushing value as a target text message into a text message pushing set;
the updating module is used for updating the operation factors of the interest coverage calculation model according to the interest coverage of the target text information which is newly stored in the text information pushing set;
the processing module is further configured to score the pushed value of each remaining candidate text message in the candidate text message set according to the preset user interest point association model, the text interest point association model and the updated interest coverage calculation model, and store the candidate text message with the highest pushed value as a target text message in a text message pushed set;
and the pushing module is used for pushing the text information pushing set to the target user.
8. An intelligent terminal is characterized by comprising a processor and a memory;
wherein the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory, for implementing the text information pushing method according to any one of claims 1 to 6.
9. A non-transitory computer-readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the text information pushing method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810286984.2A CN108491540B (en) | 2018-03-30 | 2018-03-30 | Text information pushing method and device and intelligent terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810286984.2A CN108491540B (en) | 2018-03-30 | 2018-03-30 | Text information pushing method and device and intelligent terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108491540A CN108491540A (en) | 2018-09-04 |
| CN108491540B true CN108491540B (en) | 2022-02-22 |
Family
ID=63317786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810286984.2A Active CN108491540B (en) | 2018-03-30 | 2018-03-30 | Text information pushing method and device and intelligent terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108491540B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110134827B (en) * | 2019-03-28 | 2021-07-09 | 北京达佳互联信息技术有限公司 | Method and device for determining recommended video, electronic equipment and storage medium |
| CN110866106A (en) * | 2019-10-10 | 2020-03-06 | 重庆金融资产交易所有限责任公司 | Text recommendation method and related equipment |
| CN112163399A (en) * | 2020-10-12 | 2021-01-01 | 北京字跳网络技术有限公司 | Online document pushing method and device, electronic equipment and computer readable medium |
| CN112256970A (en) * | 2020-10-28 | 2021-01-22 | 四川金熊猫新媒体有限公司 | News text pushing method, device, equipment and storage medium |
| CN112633977A (en) * | 2020-12-22 | 2021-04-09 | 苏州斐波那契信息技术有限公司 | User behavior based scoring method, device computer equipment and storage medium |
| CN113553421B (en) * | 2021-06-22 | 2023-05-05 | 北京百度网讯科技有限公司 | Comment text generation method and device, electronic equipment and storage medium |
| CN117708439A (en) * | 2024-02-06 | 2024-03-15 | 每日互动股份有限公司 | Target text pushing method, device, medium and equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102360364A (en) * | 2011-09-30 | 2012-02-22 | 奇智软件（北京）有限公司 | Automatic application recommendation method and device |
| CN103440341A (en) * | 2013-09-09 | 2013-12-11 | 广州品唯软件有限公司 | Information recommendation method and device |
| CN105045858A (en) * | 2015-07-10 | 2015-11-11 | 湖南科技大学 | Voting based taxi passenger-carrying point recommendation method |
| CN105404680A (en) * | 2015-11-25 | 2016-03-16 | 百度在线网络技术（北京）有限公司 | Searching recommendation method and apparatus |
| CN105787055A (en) * | 2016-02-26 | 2016-07-20 | 合网络技术（北京）有限公司 | Information recommendation method and device |
| CN106407364A (en) * | 2016-09-08 | 2017-02-15 | 北京百度网讯科技有限公司 | Information recommendation method and apparatus based on artificial intelligence |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030204588A1 (en) * | 2002-04-30 | 2003-10-30 | International Business Machines Corporation | System for monitoring process performance and generating diagnostic recommendations |
| US7363543B2 (en) * | 2002-04-30 | 2008-04-22 | International Business Machines Corporation | Method and apparatus for generating diagnostic recommendations for enhancing process performance |
-
2018
- 2018-03-30 CN CN201810286984.2A patent/CN108491540B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102360364A (en) * | 2011-09-30 | 2012-02-22 | 奇智软件（北京）有限公司 | Automatic application recommendation method and device |
| CN103440341A (en) * | 2013-09-09 | 2013-12-11 | 广州品唯软件有限公司 | Information recommendation method and device |
| CN105045858A (en) * | 2015-07-10 | 2015-11-11 | 湖南科技大学 | Voting based taxi passenger-carrying point recommendation method |
| CN105404680A (en) * | 2015-11-25 | 2016-03-16 | 百度在线网络技术（北京）有限公司 | Searching recommendation method and apparatus |
| CN105787055A (en) * | 2016-02-26 | 2016-07-20 | 合网络技术（北京）有限公司 | Information recommendation method and device |
| CN106407364A (en) * | 2016-09-08 | 2017-02-15 | 北京百度网讯科技有限公司 | Information recommendation method and apparatus based on artificial intelligence |
Non-Patent Citations (5)
| Title |
|---|
| Improving a News Recommendation System in Adapting to Interests of a User with Storage of a Constant Size;Akito Nishitarumizu 等;《2010 12th International Asia-Pacific Web Conference》;20100601;第109-115页 * |
| 一种增强现实分场景推送情景感知服务的方法;林一 等;《软件学报》;20160815;第27卷(第8期);第2115-2134页 * |
| 基于抽象状态机的位置服务推荐模型的研究;蒋晶晶;《中国优秀硕士学位论文全文数据库 信息科技辑》;20140115;I138-2392 * |
| 基于网络阅读行为兴趣度模型的网摘推荐;沈阳;《情报杂志》;20070218(第2期);第68-69、73页 * |
| 社会化媒体中若干时空相关的推荐问题研究;阴红志;《中国博士学位论文全文数据库 信息科技辑》;20141115;I138-49 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108491540A (en) | 2018-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108491540B (en) | Text information pushing method and device and intelligent terminal | |
| CN111061946B (en) | Method, device, electronic equipment and storage medium for recommending scenerized content | |
| CN107330023B (en) | Text content recommendation method and device based on attention points | |
| CN107577807B (en) | Method and device for pushing information | |
| US20150278350A1 (en) | Recommendation System With Dual Collaborative Filter Usage Matrix | |
| US20160132601A1 (en) | Hybrid Explanations In Collaborative Filter Based Recommendation System | |
| CN112307344B (en) | Object recommendation model, object recommendation method and device and electronic equipment | |
| CN110334356B (en) | Article quality determining method, article screening method and corresponding device | |
| CN111782947B (en) | Search content display method and device, electronic equipment and storage medium | |
| CN111144952A (en) | Advertisement recommendation method, device, server and storage medium based on user interests | |
| CN113592535B (en) | Advertisement recommendation method and device, electronic equipment and storage medium | |
| CN109255070B (en) | Recommendation information processing method and device, computer equipment and storage medium | |
| CN112445970A (en) | Information recommendation method and device, electronic equipment and storage medium | |
| WO2015148420A1 (en) | User inactivity aware recommendation system | |
| CN112328889A (en) | Method and device for determining recommended search terms, readable medium and electronic equipment | |
| CN112532507A (en) | Method and device for presenting expression image and method and device for sending expression image | |
| CN108563648B (en) | Data display method and device, storage medium and electronic device | |
| CN109472028B (en) | Method and device for generating information | |
| CN110992127A (en) | Article recommendation method and device | |
| US20150170035A1 (en) | Real time personalization and categorization of entities | |
| CN110083687A (en) | A kind of information conversion method, equipment and storage medium | |
| CN112802454B (en) | Method and device for recommending awakening words, terminal equipment and storage medium | |
| CN114265948A (en) | Image pushing method and device | |
| CN113221572A (en) | Information processing method, device, equipment and medium | |
| CN113626638A (en) | Short video recommendation processing method and device, intelligent terminal and storage medium |
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 |