CN110297967B - Method, device and equipment for determining interest points and computer readable storage medium - Google Patents
Method, device and equipment for determining interest points and computer readable storage medium Download PDFInfo
- Publication number
- CN110297967B CN110297967B CN201910398136.5A CN201910398136A CN110297967B CN 110297967 B CN110297967 B CN 110297967B CN 201910398136 A CN201910398136 A CN 201910398136A CN 110297967 B CN110297967 B CN 110297967B
- Authority
- CN
- China
- Prior art keywords
- interest
- point
- determining
- points
- interest points
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present disclosure provides a method, an apparatus, a device and a computer-readable storage medium for determining a point of interest, including: mining network resources through a pre-trained mining classifier to obtain interest points; determining an incidence relation between the interest points according to the obtained interest points; and determining the target interest points of the user according to the incidence relation among the interest points. The method, the device, the equipment and the computer readable storage medium provided by the disclosure can extract the interest points in time according to network resources and construct the association between the interest points, so that the target interest points of the user can be determined according to the known user information and the association relation between the interest points, and the content which is possibly interested by the user can be determined quickly under the condition that the iteration speed of the internet information is high.
Description
Technical Field
The present disclosure relates to content push technologies, and in particular, to a method, an apparatus, a device, and a computer-readable storage medium for determining a point of interest.
Background
With the development of the internet, personalized recommended contents are more and more popular for users, and the users can find interesting contents more quickly and browse the interesting contents by recommending the contents to the users in a targeted manner.
In the personalized recommendation process, the mode adopted in the prior art is to determine the interest or potential interest of the user based on the historical data of the user and then push the user.
However, the iteration speed of the network information is high, and if the content is pushed to the user only according to the historical data of the user, the content related to the historical information is easily pushed to the user, and the current newer content cannot be pushed to the user. Therefore, the trial recommendation method cannot accurately determine the content which is likely to be interested by the user currently, and further cannot accurately push the content which is interested by the user.
Disclosure of Invention
The present disclosure provides a method, an apparatus, a device and a computer-readable storage medium for determining a point of interest, so as to solve the problem that, in the prior art, a content that a user is currently interested in cannot be accurately determined, and then the content that the user is interested in cannot be accurately pushed to the user.
A first aspect of the present disclosure is to provide a method for determining a point of interest, including:
mining network resources through a pre-trained mining classifier to obtain interest points;
determining an incidence relation between the interest points according to the obtained interest points;
and determining the target interest points of the user according to the incidence relation among the interest points.
Another aspect of the present disclosure is to provide a point of interest determination apparatus, including:
the mining module is used for mining network resources through a pre-trained mining classifier to obtain interest points;
the association module is used for determining the association relationship among the interest points according to the obtained interest points;
and the determining module is used for determining the target interest points of the user according to the incidence relation among the interest points.
Yet another aspect of the present disclosure is to provide a point of interest determination device, including:
a memory;
a processor; and
a computer program;
wherein the computer program is stored in the memory and configured to be executed by the processor to implement the point of interest determination method as described in the first aspect above.
Yet another aspect of the present disclosure is to provide a computer-readable storage medium having stored thereon a computer program for execution by a processor to implement the point of interest determination method as described in the first aspect above.
The method, the device, the equipment and the computer-readable storage medium for determining the interest points have the technical effects that:
the method, the device, the equipment and the computer readable storage medium for determining the interest points provided by the disclosure comprise the following steps: mining network resources through a pre-trained mining classifier to obtain interest points; determining an incidence relation between the interest points according to the obtained interest points; and determining the target interest points of the user according to the incidence relation among the interest points. The method, the device, the equipment and the computer readable storage medium provided by the disclosure can extract the interest points in time according to network resources and construct the association between the interest points, so that the target interest points of the user can be determined according to the known user information and the association relation between the interest points, and the content which is possibly interested by the user can be determined quickly under the condition that the iteration speed of the internet information is high.
Drawings
FIG. 1 is a flow chart illustrating a method for point of interest determination in accordance with an exemplary embodiment of the present invention;
FIG. 2 is a flow chart illustrating a method for point of interest determination in accordance with another exemplary embodiment of the present invention;
fig. 3 is a block diagram illustrating a point of interest determination apparatus according to an exemplary embodiment of the present invention;
fig. 4 is a block diagram illustrating a point of interest determination apparatus according to another exemplary embodiment of the present invention;
fig. 5 is a block diagram illustrating a point of interest determination device according to an exemplary embodiment of the present invention.
Detailed Description
With the development of network technology, more and more users can acquire required information in the network, for example, watching some hot news. In order to provide users with better service, many network platforms actively push content, such as news that occurs today, and hot content that users are likely to be interested in.
In the prior art, a network platform acquires historical data of a user, speculates user interest according to the historical data, selects content which the user is likely to be interested in news content or hot content, and pushes the content to the user.
However, as the network information iteration speed is higher, the accuracy of content which is probably interested currently is estimated to be lower only according to the historical data of the user. According to the scheme provided by the embodiment of the invention, the interest points can be mined in the network data with very high iteration speed, and the incidence relation among the interest points is determined, so that the target interest points of the user can be accurately determined based on the incidence relation, the content which is more consistent with the user can be matched when the content is pushed to the user based on the target interest points, and the pushing efficiency is improved.
Fig. 1 is a flowchart illustrating a method for determining a point of interest according to an exemplary embodiment of the present invention.
As shown in fig. 1, the method for determining a point of interest provided by this embodiment includes:
The method provided by the embodiment can be executed by an electronic device with computing capability, such as a computer. The electronic device may be a background server of a network platform, and the method provided by the embodiment may be packaged in the server, so that the server can execute the method.
The electronic equipment can be connected with the user terminal through a network, so that content pushing is carried out on the user terminal. For example, a client may be installed in the user terminal, the user may operate the client to browse the web content, and the electronic device may push the content to the user terminal through the function of the client.
Specifically, a pre-trained mining classifier may be set in the electronic device. The mining classifier may process the network resources to extract points of interest that may be included therein. The interest point refers to content that may be of interest to the user, such as a star that is currently compared to red, such as an emergency event.
Further, there are some points of interest that already exist, such as those generated months or even years ago, which may be known points of interest. And some interest points are generated or not generated, and because the network information has a very high circulation speed, a plurality of emergencies can be rapidly spread in the network, so that new interest points are generated at any time. Through the pre-trained mining classifier, network resources can be processed, and then currently generated interest points can be obtained in time.
In practical application, the classifier can be trained according to known interest points and corresponding network resources. For example, some points of interest and their corresponding network resources may be selected as samples, which may include some positive and negative samples. The positive sample is the combination of the correct interest point and the network resource, and the negative sample is the combination of the wrong interest point and the network resource.
One part of the sample can be used as a training sample, and the other part of the sample can be used as a test sample. And executing a classifier algorithm on the training samples to generate a classifier. And executing a classifier on the test sample to generate a prediction result. And determining an evaluation index according to the prediction result, and evaluating the performance of the classifier. If the classifier does not reach the standard, the parameters in the classifier can be adjusted, and the training is carried out again.
The network resource may be web page data, and for example, a classifier capable of mining points of interest in the web page data may be trained. The network resource may also be search data, for example, a classifier may be trained that is capable of mining points of interest in the user's search data.
Specifically, network resources in the network can be acquired, the trained classifier is used for calculating the network resources, interest points included in the network resources are extracted, and then the current interest points in the network can be acquired in time.
And step 102, determining the association relationship between the interest points according to the obtained interest points.
The relation between the newly acquired interest point and other interest points can be determined, so that the other interest points of the user can be presumed on the basis of the known interest points of the user according to the relation.
For a plurality of points of interest in the network, the points of interest have a certain relationship, for example, a point of interest a and a point of interest B appear in pairs, and they can be considered to have a certain relationship. As another example, some points of interest include a plurality of sub-points of interest, and the relationship may be determined according to an existing knowledge system, such as a point of interest heading, belonging to a point of interest history. The relationship may also be determined based on network data, such as a television program being a point of interest, and guests participating in the television program may be sub-points of interest of the point of interest.
Specifically, the relationship between the interest points is constructed, an interest point map can be drawn, and each interest point has a corresponding associated interest point.
And 103, determining the target interest points of the user according to the incidence relation among the interest points.
Further, the interest point of a user may be determined first, and according to the association relationship between the interest points, other interest points having a relationship with the interest point may be determined, and these other interest points may be determined as the target interest point of the user. For example, if the point of interest a and the point of interest B have an association relationship, it can be assumed that the user is interested in a and B.
In practical application, some interest points have an inclusion relationship, and in this case, the target interest point of the user can be determined according to the relationship. For example, if the user has a point of interest a, it may be considered that it is also likely to be interested in sub-points of interest a, and these sub-points of interest may be sent to the user, thereby enabling the user to select among them.
The method can also directly inquire the interest points of the user, further accurately determine a known interest point, and then determine the target interest point according to the known interest point and the incidence relation of the interest points.
In the method provided by the embodiment, the target interest point of the user can be determined according to the incidence relation between the interest points, so that the content concerned by the user can be more accurately presumed based on the existing user information or data.
Specifically, the method provided in this embodiment may further include:
and determining push content corresponding to the user according to the target interest point, and sending the push content to the user terminal.
Further, after determining the content that may be of interest to the user, the web content may be filtered based on the content, for example, hot news may be filtered, and pushed to the user.
The method provided by the embodiment can accurately determine the target interest point of the user based on the existing data, and further can push the content more matched with the target interest point to the user based on the target interest point.
The method provided by the present embodiment is used for determining the interest point of the user, and is performed by a device provided with the method provided by the present embodiment, and the device is generally implemented in a hardware and/or software manner.
The method for determining the interest point provided by the embodiment comprises the following steps: mining network resources through a pre-trained mining classifier to obtain interest points; determining an incidence relation between the interest points according to the obtained interest points; and determining the target interest points of the user according to the incidence relation among the interest points. The method provided by the embodiment can extract the interest points in time according to the network resources and construct the association between the interest points, so that the target interest points of the user can be determined according to the known user information and the association relationship between the interest points, and the content which the user is likely to be interested in can be determined quickly under the condition that the internet information iteration speed is high.
Fig. 2 is a flowchart illustrating a method for determining a point of interest according to another exemplary embodiment of the present invention.
As shown in fig. 2, the method for determining a point of interest provided in this embodiment includes:
step 201, extracting webpage features in the webpage data through a first mining classifier, and determining interest points according to the webpage features.
The first mining classifier is obtained by training known interest points and corresponding webpage data.
Specifically, known interest points and corresponding web page data thereof may be collected in advance, and these data may be used as training data to train the classifier. The combination of the correct interest points and the webpage data can be used as a positive sample, the combination of the wrong interest points and the webpage data can be used as a negative sample, and the first classifier can be obtained through accurate training based on the two sample data.
Further, the electronic device for training the classifier may be an electronic device for executing the method provided by the embodiment, or may be another electronic device.
And storing the trained first classifier in the electronic equipment for executing the method provided by the embodiment. The electronic equipment can scan webpage data in a network, and the webpage data are calculated through the first classifier to obtain corresponding interest points.
The web page data may be data content included in a web page, such as text content, picture content, link address, and the like.
When the first classifier is used for processing the webpage data, webpage features included in the webpage data can be extracted, and then the webpage features are classified, so that interest points are determined.
The web page features may include at least one of: page characteristics, entry hot spots and entry types. For each webpage feature, one corresponding content can be extracted, and a plurality of corresponding contents can also be extracted.
Step 202, extracting search features in the network search data through a second mining classifier, and determining interest points according to the search features.
The second mining classifier is obtained by training known interest points and corresponding network search data.
The timing sequence of step 201 and step 202 is not limited. Meanwhile, the interest points may be mined in only one of the steps 201 and 202, or in both of the two manners.
Specifically, the second mining classifier is obtained by training known interest points and corresponding network search data thereof.
Furthermore, known interest points and corresponding network search data thereof can be collected in advance, and the data can be used as training data to train the classifier. The combination of the correct interest point and the network search data can be used as a positive sample, the combination of the wrong interest point and the network search data can be used as a negative sample, and the second classifier can be obtained through accurate training based on the two sample data.
The network search data refers to data generated by a user performing information search in a network.
In practical application, the electronic device for training the classifier may be an electronic device for executing the method provided by the embodiment, or may be another electronic device.
And storing the trained second classifier in the electronic equipment for executing the method provided by the embodiment. The electronic equipment can collect search data of a user in a network, and the search data is calculated through the second classifier to obtain corresponding interest points.
When the second classifier processes the network search data, the second classifier can extract the search features included in the network search data and classify the network search features, so that the interest points are determined.
The search features include at least one of: search information, user click information, search timeliness information, and web page content characteristics. For each search feature, one or more corresponding contents may be extracted.
After the interest point is mined based on steps 201 and/or 202, the association relationship between the interest point and other interest points can be determined. The determination may be made in any of steps 203, 205, 206.
Step 203, determining the co-occurrence information among the interest points in the network resource, and determining the association relationship among the interest points according to the co-occurrence information.
In one embodiment, it is contemplated that there may be some relationship between points of interest that are co-occurring. Accordingly, co-occurrence information between points of interest, such as co-occurrence times, ratios of co-occurrence times to individual occurrence times, etc., may be determined in the network resources.
For example, when a television program and an actor name are frequently presented together, the two may be considered to have an association relationship, and at this time, the television program and the actor may be considered to have a non-directional relationship.
Wherein, a weight value can be set for measuring the correlation strength between the two interest points. For example, if the number of co-occurrences of two points of interest exceeds a threshold number, the strength of association between the two points of interest is considered to be high, and therefore, a larger weight value may be set.
Specifically, the weight value may also be obtained by calculation, for example, the number of co-occurrences between the points of interest is used as the weight value.
Further, after obtaining an interest point, the relationship between the interest point and other interest points may be determined, where the other interest points may be obtained by mining based on the method of the present embodiment, or may be obtained by other methods.
Step 204, determining the associated users corresponding to the social network users in the social network interest points, and associating the associated interest points of the associated users.
In another embodiment, the point of interest may be a social network user, such as user N. At this time, the associated interest point of this interest point may also be determined through the social network.
There are many network users in the network, some of which may themselves be points of interest, such as some actors, singers, etc. The network users have social relations in the social network, so that the association relations can be determined according to the associated users of the social network users belonging to the interest points in the social network. For example, a user who has a mutual interest with a social network user belonging to a point of interest may be regarded as a related user of the point of interest, that is, may also be regarded as a point of interest, and the two points of interest may be considered to have a relationship therebetween.
When the associated user is determined, the associated user can be determined according to the interaction information between the two users. For example, if two users interact frequently, where one user is considered a point of interest, the other user may be the associated point of interest for that point of interest.
In particular, there are some interest points that have a top-to-bottom relationship, for example, a plurality of other interest points such as football, basketball, etc. may be included in the interest point sports. Therefore, the interest point association relationship with the point-to-point relationship can be constructed based on the upper and lower relationships between the interest points.
Further, a to-be-determined notion of a position of interest may be determined.
In one embodiment, a concept hierarchy may be constructed in advance based on an existing knowledge hierarchy in the network resources. The concept system can include the corresponding upper names and lower names of each noun.
After an interest point is determined, if the interest point does not belong to the concept system, a to-be-determined upper concept of the interest point can be determined in the concept system.
In another embodiment, a concept to be positioned may also be determined according to the content of the web page corresponding to the point of interest. Or determining the concept to be positioned according to the search data corresponding to the interest point.
And step 206, determining whether the pending concept is a real concept of interest point according to the third classifier.
In practical application, a third classifier can be further arranged in the electronic equipment and used for determining whether the to-be-determined upper concept is accurate or not.
Wherein the third classifier may be pre-trained. For example, the determined concept system may be used as training data, and the third classifier may be trained according to search distribution data of known interest points and their corresponding superordinate concepts, and combinations thereof.
Specifically, the trained third classifier is arranged in the electronic device and used for determining whether the to-be-determined concept is accurate or not. The web page content, search data and the like corresponding to the interest point and the determined to-be-determined upper concept can be input into the third classifier so as to be confirmed.
Further, in step 205, a plurality of pending concepts may be determined for one point of interest, and in this case, this step may further confirm each pending concept. A point of interest may have multiple real superordinates, for example, dream of Red mansions may have the superordinates "literature", and may also have the superordinates "history", "Qing dynasty", etc.
In actual application, if it is determined that the to-be-determined concept is the real concept of the interest point, it is determined that the to-be-determined concept and the interest point have a pointing association relationship, for example, the interest point points to the real concept of the interest point.
If it is determined that each to-be-determined concept of the interest point is not the real concept, the interest point may not have the concept. For example, at the beginning of establishing the point-of-interest relationship graph, there may be less information included therein, and at this time, it may not be possible to determine the true superordinate concept in the existing data.
For the interest points without the upper concept, when the new interest points are added into the map, the new interest points are identified according to the newly added content, and whether the new interest points are the upper concept of the interest points is determined.
Specifically, the method provided in this embodiment may further determine the target interest point of the user according to the determined relationship between the interest points.
Further, historical interest points of the user, namely contents which the user is interested in once, can be determined according to historical data of the user. This determination may be made by methods known in the art.
In practical application, the network resource iteration speed is high, and it is possible that the historical interest points of the user do not match the current interest points. Therefore, if contents are pushed to the user only according to the historical interest points, the contents which are not interested in the contents are easily pushed to the user.
And step 209, determining a target interest point having an association with the historical interest point according to the association relationship between the interest points.
And determining target interest points having association with the historical interest points based on the constructed association relationship between the interest points.
For example, if an undirected association relationship is constructed, interest points having a relationship with historical interest points are directly obtained, and the interest points can be directly used as target interest points; some interest points with stronger relevance can be screened as target interest points according to the weight values between the interest points and the historical interest points, for example, a preset number of interest points with higher weight values are screened as the target interest points.
For another example, if a directed association relationship is constructed, a top interest point of the historical interest points may be obtained and may be used as a target interest point. The interest of the user can be more accurately determined. For example, the interest points that belong to the top interest point and have undirected association with the historical interest points may be used as the target interest points in combination with undirected association. For example, the historical point of interest is A1If the upper interest point is A, other interest points belonging to A, such as A, can be obtained2、A3Let A be2And A1Having a directionless associative relationship, A3And A1If the undirected correlation relationship is not available, the A is considered to be2Is a target point of interest.
Step 209 is a method for determining the target point of interest by the electronic device, and may also be a method for determining the target point of interest by interacting with the user.
If the directed association relationship between the interest points is established, the electronic device may interact with the user terminal based on the directed association relationship, so as to determine the target interest point.
Specifically, a first interest point may be determined first, for example, the first interest point may be a historical interest point as described above, the first interest point may be determined according to the current browsing content of the user, or the first interest point may be determined randomly.
Further, the electronic device may send a query message to the user terminal for querying whether the user is interested in the first point of interest.
In actual application, the user may operate the user terminal to reply, for example, to reply with yes or no, like or dislike information, and the like.
And step 211, receiving the interest result returned by the user terminal.
After the user operates the user terminal to reply, the electronic device can receive an interest result fed back by the user terminal. The result may specifically be yes or no.
Step 212, determining an actual interest point according to the interest result, and determining a target interest point according to the orientation relation of the actual interest point in the association relation.
Specifically, the actual interest point may be determined according to the result returned by the user, for example, if the user returns yes, the first interest point may be considered as the actual interest point of the user, and if the user returns no, step 210 may be continuously performed to query the user.
Further, after the actual interest point is determined, the target interest point can be determined according to the directional relation of the actual interest point in the directional association relation. For example, if the user is interested in football, the electronic device may determine that the user is interested in sports, and then take sports as a target point of interest.
In practical application, if targeted content push can be performed on a user, a more aggressive target interest point needs to be determined, so that a child interest point belonging to an actual interest point can be determined in the directed inter-interest-point association relationship, and/or a parent interest point of the actual interest point can be determined in the directed inter-interest-point association relationship. For example, if the actual point of interest is sports, sub-points of interest for sports, such as basketball, football, gymnastics, swimming, etc., may be obtained. If the actual interest point is basketball, the corresponding parent interest point can be obtained, such as sports.
For example, if the actual interest point only has a parent interest point, the parent interest point may be obtained, and if the actual interest point only has a child interest point, the child interest point may be obtained. The parent point of interest here is a generic concept of its children.
And sending inquiry information comprising the child interest points and/or the parent interest points to the user terminal of the user according to the child interest points and/or the parent interest points.
Specifically, after the interest point associated with the actual interest point is obtained, the user may be further queried whether the interest point is interested, and therefore, query information including the determined interest point may be sent to the user terminal.
After seeing the inquiry information, the user can operate the user terminal to feed back to the electronic equipment.
And receiving a second interest result returned by the user terminal.
After the user operates the user terminal to determine whether the current interest point is interested, the user terminal can send the current interest point to the electronic equipment, and then receive a second interest result.
Assuming that the second interest result is that the user is interested in the currently determined interest point, the second interest result can be used as a target interest point. The process of determining the target interest point may be ended, and the steps of determining other child interest points belonging to the actual interest point and/or determining other parent interest points of the actual interest point in the directed inter-interest point association relationship may be further performed.
And if the user is determined not to be interested in the child interest points and/or the parent interest points according to the second interest result, the step of determining other child interest points belonging to the actual interest points in the association relationship and/or determining other parent interest points of the actual interest points in the association relationship is continued.
If the user is not interested in the currently determined interest point, other interest points associated with the actual interest point can be continuously determined, and interaction is carried out with the user based on the determined interest points. Through the mode of interacting with the user, the target interest point can be determined more directly and accurately.
For example, the interaction process between the system where the electronic device is located and the user may be:
the system comprises the following steps: do you like sports news?
The user: (iii) liked;
the system comprises the following steps: do you like NBA?
The user: is not of interest;
the system comprises the following steps: do you like football?
The user: liking;
the system comprises the following steps: do you like beckmham?
The user: liking;
the system comprises the following steps: we will make recommendations to you based on your interest.
Fig. 3 is a block diagram illustrating a point of interest determination apparatus according to an exemplary embodiment of the present invention.
As shown in fig. 3, the apparatus for determining a point of interest provided in this embodiment includes:
the mining module 31 is used for mining network resources through a pre-trained mining classifier to obtain interest points;
the association module 32 is configured to determine an association relationship between the interest points according to the obtained interest points;
and the determining module 33 is configured to determine the target interest point of the user according to the association relationship between the interest points.
The device for determining the interest points provided by the embodiment comprises a mining module, a searching module and a searching module, wherein the mining module is used for mining network resources through a pre-trained mining classifier to obtain the interest points; the association module is used for determining the association relationship among the interest points according to the obtained interest points; and the determining module is used for determining the target interest points of the user according to the incidence relation among the interest points. The device provided by the embodiment can extract the interest points in time according to the network resources and construct the association between the interest points, so that the target interest points of the user can be determined according to the known user information and the association relationship between the interest points, and the content which the user is likely to be interested in can be determined quickly under the condition that the internet information iteration speed is high.
The specific principle and implementation of the device for determining a point of interest provided by this embodiment are similar to those of the embodiment shown in fig. 1, and are not described herein again.
Fig. 4 is a block diagram illustrating a point of interest determination apparatus according to another exemplary embodiment of the present invention.
As shown in fig. 4, on the basis of the foregoing embodiment, in the interest point determining apparatus provided in this embodiment, the mining module 31 includes a first mining unit 311, configured to:
extracting webpage features in webpage data through a first mining classifier, and determining the interest points according to the webpage features;
the first mining classifier is obtained by training known interest points and corresponding webpage data.
Optionally, the excavation module 31 includes a second excavation unit 312, configured to:
extracting search features in network search data through a second mining classifier, and determining the interest points according to the search features;
and the second mining classifier is obtained by training known interest points and network search data corresponding to the known interest points.
Optionally, the associating module 32 includes a first associating unit 321, configured to:
and determining co-occurrence information among the interest points in the network resources, and determining the association relationship among the interest points according to the co-occurrence information.
Optionally, the interest point is a social network user;
the association module 32 comprises a second association unit 322 for:
and determining an associated user corresponding to the social network user in the interest points of the social network, and associating the interest points of the associated user.
Optionally, the associating module 32 includes a third associating unit 323, configured to:
determining the undetermined upper concept of the interest point according to network resources;
determining whether the undetermined upper concept is a real upper concept of the interest point according to a third classifier;
if yes, determining that the real upper concept has a pointing association relation with the interest point.
Optionally, the determining module 33 includes a first determining unit 331 configured to:
determining historical interest points of the user according to historical data of the user;
and determining target interest points which are related to the historical interest points according to the association relationship among the interest points.
Optionally, the determining module 33 includes a second determining unit 332, configured to:
sending query information including a first point of interest to a user terminal of the user;
receiving an interest result returned by the user terminal;
and determining an actual interest point according to the interest result, and determining the target interest point according to the orientation relation of the actual interest point in the association relation.
Optionally, the second determining unit 332 is specifically configured to:
determining a child interest point belonging to the actual interest point in the incidence relation, and/or determining a parent interest point of the actual interest point in the incidence relation;
sending inquiry information comprising the child interest points and/or the parent interest points to a user terminal of the user according to the child interest points and/or the parent interest points;
receiving a second interest result returned by the user terminal;
if the user is determined not to be interested in the child interest points and/or the parent interest points according to the second interest result, the step of determining other child interest points belonging to the actual interest points in the association relationship and/or determining other parent interest points of the actual interest points in the association relationship is continued.
The specific principle and implementation of the device for determining a point of interest provided in this embodiment are similar to those of the embodiment shown in fig. 2, and are not described herein again.
Fig. 5 is a block diagram illustrating a point of interest determination device according to an exemplary embodiment of the present invention.
As shown in fig. 5, the interest point determining apparatus provided in this embodiment includes:
a memory 51;
a processor 52; and
a computer program;
wherein the computer program is stored in the memory 51 and configured to be executed by the processor 52 to implement any of the point of interest determination methods as described above.
The present embodiments also provide a computer-readable storage medium, having stored thereon a computer program,
the computer program is executed by a processor to implement any of the point of interest determination methods described above.
Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. A method for point of interest determination, comprising:
mining network resources through a pre-trained mining classifier to obtain interest points, wherein the interest points are contents in which users are interested;
determining an incidence relation between the interest points according to the obtained interest points;
sending query information including a first point of interest to a user terminal of the user;
receiving a first interest result returned by the user terminal;
determining an actual interest point according to the first interest result, and determining a first child interest point and/or a first parent interest point of the actual interest point in the incidence relation;
sending query information including the first child point of interest and/or the first parent point of interest to the user terminal;
determining whether the user is interested in the first child interest point and/or the first parent interest point according to a second interest result returned by the user terminal;
if the interest is found, the first child interest point and/or the first parent interest point are/is used as a target interest point;
if the interest is not in the interest, continuing to determine other child interest points belonging to the actual interest point in the incidence relation, and/or determining other parent interest points of the actual interest point in the incidence relation;
if the interest points do not belong to a pre-constructed concept system, determining the association relationship between the interest points according to the obtained interest points comprises the following steps:
determining a pending upper concept of the interest point in the concept system; the concept system comprises upper names and lower names corresponding to a plurality of nouns, and is constructed according to an existing knowledge system in network resources;
determining whether the undetermined upper concept is a real upper concept of the interest point according to a third classifier;
if so, determining that the real upper concept has a pointing association relation with the interest point;
if not, determining that the interest point is at the initial stage of establishing the interest point relation map, and the interest point does not have a superordinate concept; and when a new interest point is added into the relation graph, continuously determining whether the new interest point is a superior concept of the interest point.
2. The method of claim 1, wherein mining network resources to obtain points of interest through a pre-trained mining classifier comprises:
extracting webpage features in webpage data through a first mining classifier, and determining the interest points according to the webpage features;
the first mining classifier is obtained by training known interest points and corresponding webpage data.
3. The method of claim 1, wherein mining network resources to obtain points of interest through a pre-trained mining classifier comprises:
extracting search features in network search data through a second mining classifier, and determining the interest points according to the search features;
and the second mining classifier is obtained by training known interest points and network search data corresponding to the known interest points.
4. An apparatus for point of interest determination, comprising:
the mining module is used for mining network resources through a pre-trained mining classifier to obtain interest points, wherein the interest points are contents in which users are interested;
the association module is used for determining the association relationship among the interest points according to the obtained interest points;
the determining module is used for determining the target interest points of the user according to the incidence relation among the interest points;
the determining module comprises a second determining unit configured to: sending first interest point inquiry information to a user terminal of a user; determining an actual interest point according to an interest result returned by the user terminal; determining a first child interest point and/or a first parent interest point of the actual interest points in the incidence relation; sending query information including the first child point of interest and/or the first parent point of interest to the user terminal; determining whether the user is interested in the first child interest point and/or the first parent interest point according to a second interest result returned by the user terminal; if the interest is found, the first child interest point and/or the first parent interest point are/is used as a target interest point; if the interest is not in the interest, continuing to determine other child interest points belonging to the actual interest point in the incidence relation, and/or determining other parent interest points of the actual interest point in the incidence relation;
the association module comprises a third association unit, and if the interest point does not belong to a pre-constructed concept system, the third association unit is used for: determining a pending upper concept of the interest point in the concept system; the concept system comprises upper names and lower names corresponding to a plurality of nouns, and is constructed according to an existing knowledge system in network resources; determining whether the undetermined upper concept is a real upper concept of the interest point according to a third classifier; if so, determining that the real upper concept has a pointing association relation with the interest point; if not, determining that the interest point is at the initial stage of establishing the interest point relation map, and the interest point does not have a superordinate concept; and when a new interest point is added into the relation graph, continuously determining whether the new interest point is a superior concept of the interest point.
5. The apparatus of claim 4, wherein the excavation module comprises a first excavation unit to:
extracting webpage features in webpage data through a first mining classifier, and determining the interest points according to the webpage features;
the first mining classifier is obtained by training known interest points and corresponding webpage data.
6. The apparatus of claim 4, wherein the excavation module comprises a second excavation unit to:
extracting search features in network search data through a second mining classifier, and determining the interest points according to the search features;
and the second mining classifier is obtained by training known interest points and network search data corresponding to the known interest points.
7. A point of interest determination device, comprising:
a memory;
a processor; and
a computer program;
wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method of any of claims 1-3.
8. A computer-readable storage medium, having stored thereon a computer program,
the computer program is executed by a processor to implement the method according to any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910398136.5A CN110297967B (en) | 2019-05-14 | 2019-05-14 | Method, device and equipment for determining interest points and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910398136.5A CN110297967B (en) | 2019-05-14 | 2019-05-14 | Method, device and equipment for determining interest points and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110297967A CN110297967A (en) | 2019-10-01 |
| CN110297967B true CN110297967B (en) | 2022-04-12 |
Family
ID=68026849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910398136.5A Active CN110297967B (en) | 2019-05-14 | 2019-05-14 | Method, device and equipment for determining interest points and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110297967B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110726418B (en) * | 2019-10-10 | 2021-08-03 | 北京百度网讯科技有限公司 | Method, device and equipment for determining interest point region and storage medium |
| CN111160471B (en) * | 2019-12-30 | 2023-04-07 | 腾讯云计算（北京）有限责任公司 | Interest point data processing method and device, electronic equipment and storage medium |
| CN115002675A (en) * | 2022-05-23 | 2022-09-02 | 北京字节跳动科技有限公司 | Data matching method and device, readable medium and electronic equipment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102542489A (en) * | 2011-12-27 | 2012-07-04 | 纽海信息技术（上海）有限公司 | Recommendation method based on user interest association |
| CN102857471A (en) * | 2011-06-27 | 2013-01-02 | 深圳深讯和科技有限公司 | Multimedia interacting method and system |
| CN103166918A (en) * | 2011-12-12 | 2013-06-19 | 阿里巴巴集团控股有限公司 | Data recommendation method and device |
| CN105069125A (en) * | 2015-08-13 | 2015-11-18 | 上海斐讯数据通信技术有限公司 | Social network recommending method and social network recommending system |
| CN105512334A (en) * | 2015-12-29 | 2016-04-20 | 成都陌云科技有限公司 | Data mining method based on search words |
| CN105528459A (en) * | 2016-01-08 | 2016-04-27 | 腾讯科技（深圳）有限公司 | Information processing method, server and terminal |
| CN108108465A (en) * | 2017-12-29 | 2018-06-01 | 北京奇宝科技有限公司 | A kind of method and apparatus for pushing recommendation |
| CN108287864A (en) * | 2017-12-06 | 2018-07-17 | 深圳市腾讯计算机系统有限公司 | A kind of interest group division methods, device, medium and computing device |
| CN108345702A (en) * | 2018-04-10 | 2018-07-31 | 北京百度网讯科技有限公司 | Entity recommends method and apparatus |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103914536B (en) * | 2014-03-31 | 2017-11-07 | 北京百度网讯科技有限公司 | A kind of point of interest for electronic map recommends method and system |
| CN108875007B (en) * | 2018-06-15 | 2019-12-17 | 腾讯科技（深圳）有限公司 | method and device for determining interest point, storage medium and electronic device |
-
2019
- 2019-05-14 CN CN201910398136.5A patent/CN110297967B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102857471A (en) * | 2011-06-27 | 2013-01-02 | 深圳深讯和科技有限公司 | Multimedia interacting method and system |
| CN103166918A (en) * | 2011-12-12 | 2013-06-19 | 阿里巴巴集团控股有限公司 | Data recommendation method and device |
| CN102542489A (en) * | 2011-12-27 | 2012-07-04 | 纽海信息技术（上海）有限公司 | Recommendation method based on user interest association |
| CN105069125A (en) * | 2015-08-13 | 2015-11-18 | 上海斐讯数据通信技术有限公司 | Social network recommending method and social network recommending system |
| CN105512334A (en) * | 2015-12-29 | 2016-04-20 | 成都陌云科技有限公司 | Data mining method based on search words |
| CN105528459A (en) * | 2016-01-08 | 2016-04-27 | 腾讯科技（深圳）有限公司 | Information processing method, server and terminal |
| CN108287864A (en) * | 2017-12-06 | 2018-07-17 | 深圳市腾讯计算机系统有限公司 | A kind of interest group division methods, device, medium and computing device |
| CN108108465A (en) * | 2017-12-29 | 2018-06-01 | 北京奇宝科技有限公司 | A kind of method and apparatus for pushing recommendation |
| CN108345702A (en) * | 2018-04-10 | 2018-07-31 | 北京百度网讯科技有限公司 | Entity recommends method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110297967A (en) | 2019-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6438135B2 (en) | Data mining method and apparatus based on social platform | |
| CN106326391B (en) | Multimedia resource recommendation method and device | |
| US20150242497A1 (en) | User interest recommending method and apparatus | |
| CN110297967B (en) | Method, device and equipment for determining interest points and computer readable storage medium | |
| Qian et al. | Social media based event summarization by user–text–image co-clustering | |
| CN110909182A (en) | Multimedia resource searching method and device, computer equipment and storage medium | |
| US10019419B2 (en) | Method, server, browser, and system for recommending text information | |
| Dewan et al. | Facebook Inspector (FbI): Towards automatic real-time detection of malicious content on Facebook | |
| CN107592572B (en) | Video recommendation method, device and equipment | |
| US20230004608A1 (en) | Method for content recommendation and device | |
| KR20150046431A (en) | Auto-learning system and method for derive effective marketing | |
| CN108401005B (en) | Expression recommendation method and device | |
| KR101929649B1 (en) | System and method for recommendation of open chat room through chat log keyword extraction | |
| JP6434954B2 (en) | Information processing apparatus, information processing method, and program | |
| CN103955480A (en) | Method and equipment for determining target object information corresponding to user | |
| US11010687B2 (en) | Detecting abusive language using character N-gram features | |
| CN114547439A (en) | Service optimization method based on big data and artificial intelligence and electronic commerce AI system | |
| CN106570003B (en) | Data pushing method and device | |
| CN112182351A (en) | News recommendation method and device based on multi-feature fusion | |
| CN111523034B (en) | Application processing method, device, equipment and medium | |
| CN111597444B (en) | Searching method, searching device, server and storage medium | |
| CN114547440A (en) | User portrait mining method based on internet big data and artificial intelligence cloud system | |
| US10242106B2 (en) | Enhance search assist system's freshness by extracting phrases from news articles | |
| US11409750B2 (en) | Information search method and server | |
| CN113244618A (en) | Car logo replacing method and device applied to map client and electronic equipment |
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 |