CN110297967A - Point of interest determines method, apparatus, equipment and computer readable storage medium - Google Patents
Point of interest determines method, apparatus, equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN110297967A CN110297967A CN201910398136.5A CN201910398136A CN110297967A CN 110297967 A CN110297967 A CN 110297967A CN 201910398136 A CN201910398136 A CN 201910398136A CN 110297967 A CN110297967 A CN 110297967A
- Authority
- CN
- China
- Prior art keywords
- interest
- point
- user
- incidence relation
- excavation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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 disclosure provides a kind of point of interest and determines method, apparatus, equipment and computer readable storage medium, comprising: by excavation classifier trained in advance, excavates to Internet resources, obtains point of interest;The incidence relation between point of interest is determined according to the point of interest of acquisition;The target point of interest of user is determined according to the incidence relation between point of interest.Method, apparatus, equipment and the computer readable storage medium that the disclosure provides can extract in time point of interest according to Internet resources, and construct the association between point of interest, so as to according to known user information, and the incidence relation between point of interest determines the target point of interest of user, thus again in the faster situation of internet information iteration speed, it is faster to determine that user is possible to interested content.
Description
Technical field
This disclosure relates to which content push technology more particularly to a kind of point of interest determine that method, apparatus, equipment and computer can
Read storage medium.
Background technique
With the development of internet, to user carry out personalized recommendation content it is more more and more universal, by targetedly to
User's recommendation enables to user more efficiently to find content of interest, and browses in user.
It is to determine that user's is emerging based on user's history data by the way of in the prior art during personalized recommendation
Interesting or potential interest, then be pushed to the user.
But the iteration speed of the network information is very fast, if the historical data according only to user pushes content to user, is easy
Will content push relevant to historical information to user, and can not be by currently newer content push to user.Therefore, this to taste
The mode that examination property is recommended, can not accurately determine that user is currently possible to interested content, and then can not be accurately to user
Push its interested content.
Summary of the invention
The disclosure provides a kind of point of interest and determines method, apparatus, equipment and computer readable storage medium, existing to solve
The content of user's current interest can not be accurately determined in technology, and then it is interested interior accurately can not to push its to user
The problem of appearance.
The first aspect of the disclosure is to provide a kind of point of interest and determines method, comprising:
By excavation classifier trained in advance, Internet resources are excavated, obtain point of interest;
The incidence relation between point of interest is determined according to the point of interest of acquisition;
The target point of interest of user is determined according to the incidence relation between the point of interest.
Another aspect of the disclosure is to provide a kind of point of interest determining device, comprising:
Module is excavated, for being excavated to Internet resources by excavation classifier trained in advance, obtains point of interest;
Relating module determines the incidence relation between point of interest for the point of interest according to acquisition;
Determining module, for determining the target point of interest of user according to the incidence relation between the point of interest.
The another aspect of the disclosure is to provide a kind of point of interest and determines equipment, comprising:
Memory;
Processor;And
Computer program;
Wherein, the computer program stores in the memory, and is configured to be executed by the processor to realize
Point of interest as described in above-mentioned first aspect determines method.
The another aspect of the disclosure is to provide a kind of computer readable storage medium, is stored thereon with computer program,
The computer program is executed by processor to realize that the point of interest as described in above-mentioned first aspect determines method.
The point of interest that the disclosure provides determines having the technical effect that for method, apparatus, equipment and computer readable storage medium
The point of interest that the disclosure provides determines method, apparatus, equipment and computer readable storage medium, comprising: by pre-
First trained excavation classifier, excavates Internet resources, obtains point of interest;It is determined between point of interest according to the point of interest of acquisition
Incidence relation;The target point of interest of user is determined according to the incidence relation between point of interest.The method, apparatus of disclosure offer,
Equipment and computer readable storage medium can extract in time point of interest according to Internet resources, and construct the pass between point of interest
Connection, so as to determine the target point of interest of user according to the incidence relation between known user information and point of interest, from
And again in the faster situation of internet information iteration speed, it is faster to determine that user is possible to interested content.
Detailed description of the invention
Fig. 1 determines the flow chart of method for the point of interest shown in an exemplary embodiment of the invention;
Fig. 2 determines the flow chart of method for the point of interest shown in another exemplary embodiment of the present invention;
Fig. 3 is the structure chart of the point of interest determining device shown in an exemplary embodiment of the invention;
Fig. 4 is the structure chart of the point of interest determining device shown in another exemplary embodiment of the present invention;
Fig. 5 determines the structure chart of equipment for the point of interest shown in an exemplary embodiment of the invention.
Specific embodiment
With the development of network technology, more and more users can obtain the information of needs in a network, for example, viewing one
A little hot news.In order to provide the user with more quality services, many network platforms all actively can push content, example to user
Such as the news that today occurs, user is possible to the Hot Contents being concerned about.
In the prior art, the network platform obtains the historical data of user, and according to historical data supposition user interest, and
In news content or Hot Contents, selects user and be possible to interested content, and be pushed to user.
But as network information iteration speed is very fast, only in accordance with user's history data-speculative, it currently may be interested
Content accuracy rate it is lower.Scheme provided in an embodiment of the present invention can be excavated in the very fast network data of iteration speed
Point of interest, and determine the incidence relation between point of interest, so as to be based on this incidence relation, accurately determine the target of user
Point of interest can be matched to the content being more consistent with user, mention when so that pushing content to user based on the target point of interest
High pushing efficiency.
Fig. 1 determines the flow chart of method for the point of interest shown in an exemplary embodiment of the invention.
As shown in Figure 1, point of interest provided in this embodiment determines that method includes:
Step 101, by excavation classifier trained in advance, Internet resources are excavated, obtain point of interest.
Method provided in this embodiment can be by having the execution of the electronic equipment of computing capability, such as computer.The electronics
Equipment can be the background server of a network platform, method provided in this embodiment can be encapsulated in the server, be made
It obtains server and is able to carry out this method.
Wherein, which can be with user terminal by network connection, to carry out content push to user terminal.
For example, a client can be installed in the user terminal, user can operate the Client browse Web content, and electronic equipment can
It is realized with the function by the client to user terminal and pushes content.
Specifically, the excavation classifier that training finishes in advance can be arranged in the electronic device.The excavation classifier can be with
Internet resources are handled, the point of interest that wherein may include is extracted.Point of interest refers to the possible interested content of user, example
Such as an instantly redder star, such as an emergency event etc..
Further, some points of interest are already existing, such as the interest even just generated several years ago in some months
Point, these can be used as known point of interest.It is generating or is not generating also there are also some points of interest, due to the network information
Velocity of liquid assets is very fast, and many emergency events can propagate rapidly in a network, therefore, at every moment can all have new point of interest to produce
It is raw.By excavation classifier trained in advance, Internet resources can be handled, so timely obtain currently generate it is emerging
Interesting point.
When practical application, first classifier can be trained according to known point of interest and its corresponding Internet resources.
It here may include some positive samples and negative for example, some points of interest and its corresponding Internet resources can be selected as sample
Sample.Positive sample is the combination of correct point of interest and Internet resources, and negative sample is the point of interest of mistake and the group of Internet resources
It closes.
Can also be using a part of sample as training sample, another part sample is as test sample.On training sample
Classifier algorithm is executed, classifier is generated.Classifier is executed in test sample, generates prediction result.It is true according to prediction result
Determine evaluation index, assesses the performance of classifier.If classifier is not up to standard, parameter in adjustable classifier is re-started
Training.
Wherein, Internet resources can be web data, for example, can train one can excavate interest in web data
The classifier of point.Internet resources can also be search data, for example, can train one can dig in the search data of user
Dig the classifier of point of interest.
Specifically, the Internet resources in available network, and it is calculated using the classifier that training finishes, it mentions
Take including point of interest, and then point of interest current in network can be obtained in time.
Step 102, the incidence relation between point of interest is determined according to the point of interest of acquisition.
For the point of interest newly obtained, its relationship between other points of interest can be determined, so as to according to this
Relationship, on the basis of the known point of interest of a user, thus it is speculated that other points of interest of the user.
Wherein, for numerous points of interest in network, there is certain relationship, for example, emerging between these points of interest
Interest point A and point of interest B occurs in pairs, it may be considered that the two has certain relationship.For another example it includes more for having some points of interest
A sub- point of interest can determine this relationship, such as the point of interest Qing Dynasty according to existing knowledge hierarchy, belong to point of interest history.
This relationship can also be determined according to network data, for example a TV programme are a points of interest, then participate in the TV programme
Welcome guest can be the sub- point of interest of the point of interest.
Specifically, the relationship between building point of interest, can draw point of interest map, for each point of interest, all have
Its corresponding association point of interest.
Step 103, the target point of interest of user is determined according to the incidence relation between point of interest.
Further, it can first determine the point of interest of a user, and according to the incidence relation between point of interest, determine and be somebody's turn to do
Point of interest has other related points of interest, and these other points of interest can be determined as to the target point of interest of user.For example,
Point of interest A and point of interest B has incidence relation, then when speculating that user is interested in A, can speculate that it is also interested in B.
When practical application, there are some points of interest that there is inclusion relation, at this point it is possible to determine the mesh of user according to this relationship
Mark point of interest.Such as user has a point of interest A, then it is considered that it is also possible to feel emerging to the sub- point of interest of point of interest A
Interest can send this little point of interest to user, to allow users to be selected wherein.
Wherein, can be with the point of interest of its care of direct access inquiry user, and then accurately determine point of interest known to one, then root
According to the known point of interest and the incidence relation of point of interest, target point of interest is determined.
In method provided in this embodiment, the target point of interest of user can be determined according to the incidence relation between point of interest,
So as to be based on existing user information or data, the content that user is concerned about more accurately is speculated.
Specifically, method provided in this embodiment, can also include:
Push content corresponding to the user is determined according to target point of interest, and sends push content to user terminal.
Further, after the content that user may be concerned about has been determined, Web content can be carried out based on this content
Screening, such as hot news is screened, and is pushed to the user.
Method provided in this embodiment can accurately determine the target point of interest of user, Jin Erneng based on data with existing
Enough pushed and its more matched content based on the target point of interest to user.
Method provided in this embodiment is used to determine the point of interest of user, and this method is by being provided with side provided in this embodiment
The equipment of method executes, which realizes usually in a manner of hardware and/or software.
Point of interest provided in this embodiment determines method, comprising: by excavation classifier trained in advance, to Internet resources
It is excavated, obtains point of interest;The incidence relation between point of interest is determined according to the point of interest of acquisition;According to the association between point of interest
Relationship determines the target point of interest of user.Method provided in this embodiment can extract in time point of interest according to Internet resources, and
The association between point of interest is constructed, so as to determine according to the incidence relation between known user information and point of interest
The target point of interest of user, thus again in the faster situation of internet information iteration speed, it is faster to determine that user is possible to feel
The content of interest.
Fig. 2 determines the flow chart of method for the point of interest shown in another exemplary embodiment of the present invention.
As shown in Fig. 2, point of interest provided in this embodiment determines method, comprising:
Step 201, the web page characteristics in web data are extracted by the first excavation classifier, and is determined according to web page characteristics
Point of interest.
Wherein, the first excavation classifier is obtained by known point of interest and its training of corresponding web data.
Specifically, known point of interest and its corresponding web data can be collected in advance, it can be using these data as instruction
Practice data, and then classifier is trained.Can be using the combination of correct point of interest and web data as positive sample, it will be wrong
The combination of point of interest accidentally and web data can be trained accurately based on both sample datas as negative sample and obtain first
Classifier.
Further, the electronic equipment of training classifier can be the electronic equipment for executing method provided in this embodiment,
It is also possible to other electronic equipments.
The first classifier that training finishes is stored in the electronic equipment for executing method provided in this embodiment.The electronics
Equipment can scan the web data in network, and be calculated by the first classifier these web datas, be corresponded to
Point of interest.
Web data can be the data content for including in Webpage, such as word content, image content, chained address
Deng.
First classifier can extract the web page characteristics that web data includes when handling web data, then to webpage spy
Sign is classified, so that it is determined that point of interest.
Web page characteristics may include following at least one: page feature, entry hot spot, entry type.For every kind of webpage
Feature can extract a corresponding content, can also extract multiple corresponding contents.
Step 202, the search characteristics in web search data are extracted by the second excavation classifier, and according to search characteristics
Determine point of interest.
Wherein, the second excavation classifier is obtained by known point of interest and its training of corresponding web search data.
The timing of step 201 and step 202 is with no restrictions.Meanwhile it can be only with one of step 201 and 202 side
Formula excavates point of interest, can also excavate point of interest using both modes.
Specifically, the second excavation classifier is obtained by known point of interest and its training of corresponding web search data
's.
Further, known point of interest and its corresponding web search data can be collected in advance, it can be by these data
It is trained as training data, and then to classifier.Can using the combination of correct point of interest and web search data as
Positive sample can be quasi- based on both sample datas using the combination of the point of interest of mistake and web search data as negative sample
True training obtains the second classifier.
Web search data refer to that user carries out the data of information search generation in a network.
When practical application, the electronic equipment of training classifier, which can be, to be executed the electronics of method provided in this embodiment and sets
It is standby, it is also possible to other electronic equipments.
The second classifier that training finishes is stored in the electronic equipment for executing method provided in this embodiment.The electronics
Equipment can acquire the search data that user carries out in a network, and be counted by the second classifier to these search data
It calculates, obtains corresponding point of interest.
Second classifier can extract the search characteristics that web search data include when handling web search data, then
Classify to web search feature, so that it is determined that point of interest.
Search characteristics comprise at least one of the following: search information, user's click information, search timeliness information, in webpage
Hold feature.For every kind of search characteristics, a corresponding content can be extracted, multiple corresponding contents can also be extracted.
It is excavated based on step 201 and/or 202 to after point of interest, can be determined between the point of interest and other points of interest
Incidence relation.It can specifically be determined by any mode in step 203,205,206.
Step 203, the co-occurrence information between point of interest is determined in Internet resources, and is determined between point of interest according to co-occurrence information
Incidence relation.
In one embodiment, it is believed that between the point of interest occurred jointly, it is understood that there may be certain incidence relation.Cause
This, can determine the co-occurrence information between point of interest in Internet resources, for example, co-occurrence number, co-occurrence number and independent occurrence out
Several ratio etc..
For example, when a TV programme and an actor names often occur jointly, it is believed that the two has association
Relationship, at this point it is possible to think that this TV programme has contacting for non-directive property with this performer.
Wherein, weighted value can also be set, for measuring the strength of association between two points of interest.For example, if two interest
Point co-occurrence number exceeds a threshold number, then it is assumed that the strength of association of the two is higher, therefore, a biggish weight can be set
Value.
Specifically, weighted value can also be obtained by calculation, for example, using the co-occurrence number between point of interest as weighted value.
Further, after obtaining a point of interest, its relationship between other points of interest can be determined, herein other are emerging
Interest point can be what the method based on the present embodiment was excavated, be also possible to obtain by other means.
Step 204, in social networks point of interest, the corresponding association user of social network user is determined, and association is used
The association point of interest at family.
In another embodiment, point of interest is likely to be social network user, such as user N.At this point, can also lead to
Cross the association point of interest that social networks determines this point of interest.
There are many network users in a network, some network users itself are also likely to be point of interest, such as some performers,
Singer etc..And these network users of society have social networks therefore can be according to the social activity for belonging to point of interest in social networks
Association user of the network user in social networks, determines its incidence relation.For example, can by with the social network that belongs to point of interest
The user that network user mutually pays close attention to, the association user as the point of interest, it can it is also used as to a point of interest, and is thought
There is incidence relation between the two points of interest.
Wherein, when determining association user, association user can also be determined according to the interactive information between two users.
For example, if two user interactions are frequent, one of user is considered as point of interest, then another user can be the interest
The association point of interest of point.
Step 205, the upperseat concept undetermined of point of interest is determined according to Internet resources.
Specifically, having between some points of interest, there is the relationships of upper bottom, for example, may include more in point of interest sport
A others point of interest, such as football, basketball etc..And hence it is also possible to be built with finger based on the hyponymy between point of interest
To the point of interest incidence relation of relationship.
Further, the upperseat concept undetermined of a point of interest can be first determined.
In one embodiment, a concept system can be constructed previously according to knowledge hierarchy existing in Internet resources.
It may include the corresponding upper name of each noun and the next noun in the concept system.
After determining a point of interest, if the point of interest is not belonging to this concept system, it can be determined in concept system
The upperseat concept undetermined of one point of interest.
In another embodiment, can also the web page contents according to corresponding to the point of interest, determine that one is undetermined upper
Concept.Or according to the corresponding search data of the point of interest, determine upperseat concept undetermined.
Step 206, according to third classifier determine upperseat concept undetermined whether be point of interest true upperseat concept.
When practical application, third classifier can also be set in the electronic device, for whether determining upperseat concept undetermined
Accurately.
Wherein, third classifier can be what training in advance obtained.For example, can be using determining concept system as training
Data, can also according to known point of interest and its corresponding upperseat concept, with and combinations thereof search distributed data training third
Classifier.
Specifically, in the electronic device by the setting of trained third classifier, for determining that the upperseat concept undetermined is
It is no accurate.Can be by the corresponding web page contents of point of interest, search data etc., and the upperseat concept undetermined input third determined is divided
In class device, so that it be made to be confirmed.
Further, in step 205, multiple upperseat concepts undetermined can also be determined for a point of interest, at this point, this
Step can also confirm each concept undetermined.One point of interest can have multiple true upperseat concepts, such as
Dream of the Red Mansion can have upperseat concept " literature ", can also have upperseat concept " history ", " Qing Dynasty " etc..
Step 207, if so, determining that true upperseat concept and point of interest have is directed toward incidence relation
When practical application, however, it is determined that the upperseat concept undetermined is the true upperseat concept of point of interest, it is determined that the two has
It is directed toward incidence relation, for example, point of interest is directed toward the true upperseat concept belonging to it.
Wherein, if each of confirmation point of interest upperseat concept undetermined is not its true upperseat concept, it may be considered that should
Point of interest does not have upperseat concept also.For example, establish initial stage in point of interest relation map, may including information it is less,
At this time, it is possible to true upperseat concept can not be determined in existing data.
It, can be when new point of interest be added in map, further according to newly added interior for the point of interest of not upperseat concept
Appearance identified, determine if be these points of interest upperseat concept.
Step 208, the historical interest point of user is determined according to user's history data.
Specifically, method provided in this embodiment can also determine the target of user according to relationship between determining point of interest
Point of interest.
Further, the historical interest point of user can be determined according to the historical data of user, i.e., user is once interested
Content.This method of determination can use method in the prior art.
When practical application, Internet resources iteration speed is quickly, it is possible to the historical interest point and current point of interest of user
It is not consistent.Therefore, it if pushing content to user according only to historical interest point, is easy to push its uninterested content to user.
Step 209, there is associated target point of interest with historical interest point according to the incidence relation between point of interest is determining.
Wherein it is possible to which incidence relation between the point of interest based on building, determines wherein with historical interest point with associated
Target point of interest.
For example, if constructing undirected incidence relation, directly acquires and have related point of interest with historical interest point, it can be with
Directly using these points of interest as target point of interest;Can also according to the weighted value between these points of interest and historical interest point,
The stronger point of interest of some relevances is screened as target point of interest, such as the screening higher preset quantity point of interest of weighted value
As target point of interest.
For another example if constructing oriented incidence relation, the available upper point of interest with historical interest point can be incited somebody to action
It is as a target point of interest.It can also more accurately determine the interest of user.Such as can be in conjunction with undirected incidence relation, it will
Belong to the upper point of interest, and has the point of interest of undirected incidence relation as target point of interest with historical interest point.For example,
Historical interest point is A1, upper point of interest is A, then available other belong to the point of interest such as A of A2、A3, it is assumed that A2With A1Tool
There are undirected incidence relation, A3With A1Do not have undirected incidence relation, then it is assumed that A2It is a target point of interest.
Step 209 is a kind of mode that target point of interest is determined by electronic equipment, further, it is also possible to by carrying out with user
Interactive mode determines target point of interest.
Step 210, sending to the user terminal of user includes the first point of interest inquiry message.
Wherein, if oriented incidence relation between constructing point of interest, electronic equipment based on the oriented relationship and can be used
Family terminal interacts, so that it is determined that target point of interest.
Specifically, first point of interest can first be determined, for example, it may be historical interest point as described above, may be used also
To determine the first point of interest according to the current browsing content of user, the first point of interest can also be determined at random.
Further, electronic equipment can send an inquiry message to user terminal, for asking the user whether to first
Point of interest is interested.
When practical application, user can operate user terminal and reply, such as reply yes/no, like or do not like
Information.
Step 211, the interest result that user terminal returns is received.
Wherein, after user's operation user terminal is replied, electronic equipment can receive the interest of user terminal feedback
As a result.As a result it specifically can be yes/no.
Step 212, an actual interest point is determined according to interest result, and in incidence relation, according to actual interest point
Points relationship determines target point of interest.
Specifically, can according to user reply result determine actual interest point, if for example, user reply, can
It is right if user's reply whether, can continue to execute step 210 to think that the first point of interest makes the actual interest point of user
User inquires.
It further, can be in oriented incidence relation, according to the finger of actual interest point after determining actual interest point
Target point of interest is determined to relationship.For example, electronic equipment can determine that user is emerging to sport sense if user is interested in football
Interest, and then using sport as a target point of interest.
When practical application, if targeted content push can be carried out to user, it is thus necessary to determine that the target more intensified
Point of interest, and hence it is also possible between oriented point of interest in incidence relation, the determining sub- point of interest for belonging to actual interest point, and/
Or between oriented point of interest in incidence relation, father's point of interest of actual interest point is determined.For example, if actual interest point is body
It educates, then the sub- point of interest of available sport, such as basketball, football, gymnastics, swimming.It, can be with if actual interest point is basketball
Obtain its corresponding father's point of interest, such as sport.
Wherein, the case where being also based on actual interest point, selection obtain sub- point of interest or father's point of interest, for example, if
Actual interest point only has father's point of interest, then available father's point of interest can if actual interest point only has sub- point of interest
To obtain the sub- point of interest.Father's point of interest herein is the upperseat concept of its sub- point of interest.
Being sent according to sub- point of interest and/or father's point of interest to the user terminal of user includes sub- point of interest and/or father's interest
The inquiry message of point.
Specifically, obtaining with after the associated point of interest of actual interest point, can further ask the user whether to this
Point of interest is interested, therefore, can send the inquiry message including determining point of interest to user terminal.
After user sees inquiry message, user terminal can be operated, is fed back to electronic equipment.
Receive the second interest result that user terminal returns.
User's operation user terminal be confirmed whether it is interested in current point of interest after, user terminal can be sent it to
Electronic equipment, and then receive the second interest result.
Assuming that the second interest result is, user is interested in currently determining point of interest, then can be as a mesh
Mark point of interest.The process that can terminate determining target point of interest can also continue to execute other for determining and belonging to actual interest point
Sub- point of interest, and/or between oriented point of interest in incidence relation, the step of determining other father's points of interest of actual interest point.
Assuming that determining that user loses interest in sub- point of interest and/or father's point of interest according to the second interest result, then continue
Other the sub- points of interest for belonging to actual interest point are determined in incidence relation, and/or in incidence relation, determine actual interest point
The step of other father's points of interest.
If user loses interest in currently determining point of interest, can continue to determine that other are associated with actual interest point
Point of interest, and interacted based on determining point of interest with user.It, can be more direct, quasi- by way of being interacted with user
Really set the goal point of interest really.
For example, system where electronic equipment and the interactive process of user may is that
System: are you good, you like sports news?
User: like;
Do system: you like NBA?
User: lose interest in;
Do system: you like football?
User: like;
Do system: you like Beckham?
User: like;
System: we can recommend according to your interest with you.
Fig. 3 is the structure chart of the point of interest determining device shown in an exemplary embodiment of the invention.
As shown in figure 3, point of interest determining device provided in this embodiment, comprising:
Module 31 is excavated, for being excavated to Internet resources by excavation classifier trained in advance, obtains interest
Point;
Relating module 32 determines the incidence relation between point of interest for the point of interest according to acquisition;
Determining module 33, for determining the target point of interest of user according to the incidence relation between the point of interest.
Point of interest determining device provided in this embodiment, including module is excavated, for passing through excavation classification trained in advance
Device excavates Internet resources, obtains point of interest;Relating module determines between point of interest for the point of interest according to acquisition
Incidence relation;Determining module, for determining the target point of interest of user according to the incidence relation between point of interest.The present embodiment provides
Device point of interest can be extracted in time according to Internet resources, and the association between point of interest is constructed, so as to known to
User information and point of interest between incidence relation determine the target point of interest of user, thus internet information iteration again
It is faster to determine that user is possible to interested content in the case where fast speed.
The concrete principle and implementation of point of interest determining device provided in this embodiment with embodiment class shown in FIG. 1
Seemingly, details are not described herein again.
Fig. 4 is the structure chart of the point of interest determining device shown in another exemplary embodiment of the present invention.
As shown in figure 4, on the basis of the above embodiments, point of interest determining device provided in this embodiment, the excavation
Module 31, including the first excavation unit 311, are used for:
The web page characteristics in web data are extracted by the first excavation classifier, and according to web page characteristics determination
Point of interest;
Wherein, the first excavation classifier is obtained by known point of interest and its training of corresponding web data.
Optionally, the excavation module 31 includes the second excavation unit 312, is used for:
The search characteristics in web search data are extracted by the second excavation classifier, and are determined according to described search feature
The point of interest;
Wherein, the second excavation classifier is obtained by known point of interest and its training of corresponding web search data
's.
Optionally, the relating module 32 includes the first associative cell 321, is used for:
Co-occurrence information in the Internet resources between the determination point of interest, and according to co-occurrence information determination
Incidence relation between point of interest.
Optionally, the point of interest is social network user;
The relating module 32 includes the second associative cell 322, is used for:
In social networks point of interest, the corresponding association user of the social network user is determined, and the association is used
The association point of interest at family.
Optionally, the relating module 32 includes third associative cell 323, is used for:
The upperseat concept undetermined of the point of interest is determined according to Internet resources;
According to third classifier determine the upperseat concept undetermined whether be the point of interest true upperseat concept;
If so, determining that the true upperseat concept and the point of interest have is directed toward incidence relation.
Optionally, the determining module 33 includes that the first determination unit 331 is used for:
The historical interest point of the user is determined according to user's history data;
There is associated target point of interest with the historical interest point according to the incidence relation between the point of interest is determining.
Optionally, the determining module 33 includes the second determination unit 332, is used for:
Send to the user terminal of the user includes the first point of interest inquiry message;
Receive the interest result that the user terminal returns;
An actual interest point is determined according to the interest result, and in the incidence relation, according to the actual interest
The points relationship of point determines the target point of interest.
Optionally, second determination unit 332 is specifically used for:
The sub- point of interest for belonging to the actual interest point is determined in the incidence relation, and/or in the incidence relation
In, determine father's point of interest of the actual interest point;
Being sent according to the sub- point of interest and/or father's point of interest to the user terminal of the user includes that the son is emerging
The inquiry message of interest point and/or father's point of interest;
Receive the second interest result that the user terminal returns;
If determining that the user does not feel the sub- point of interest and/or father's point of interest according to the second interest result
Interest then continues to determine other the sub- points of interest for belonging to the actual interest point in the incidence relation, and/or in the pass
In connection relationship, the step of determining other father's points of interest of the actual interest point.
The concrete principle and implementation of point of interest determining device provided in this embodiment with embodiment class shown in Fig. 2
Seemingly, details are not described herein again.
Fig. 5 determines the structure chart of equipment for the point of interest shown in an exemplary embodiment of the invention.
As shown in figure 5, point of interest provided in this embodiment determines that equipment includes:
Memory 51;
Processor 52;And
Computer program;
Wherein, the computer program is stored in the memory 51, and be configured to by the processor 52 execute with
Realize that any point of interest as described above determines method.
The present embodiment also provides a kind of computer readable storage medium, is stored thereon with computer program,
The computer program is executed by processor to realize that any point of interest as described above determines method.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence;And storage medium above-mentioned include: ROM, RAM, magnetic disk or
The various media that can store program code such as person's CD.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations;To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement;And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (20)
1. a kind of point of interest determines method characterized by comprising
By excavation classifier trained in advance, Internet resources are excavated, obtain point of interest;
The incidence relation between point of interest is determined according to the point of interest of acquisition;
The target point of interest of user is determined according to the incidence relation between the point of interest.
2. the method according to claim 1, wherein described by excavation classifier trained in advance, to network
Resource is excavated, and point of interest is obtained, comprising:
The web page characteristics in web data are extracted by the first excavation classifier, and determine the interest according to the web page characteristics
Point;
Wherein, the first excavation classifier is obtained by known point of interest and its training of corresponding web data.
3. the method according to claim 1, wherein described by excavation classifier trained in advance, to network
Resource is excavated, and point of interest is obtained, comprising:
The search characteristics in web search data are extracted by the second excavation classifier, and according to the determination of described search feature
Point of interest;
Wherein, the second excavation classifier is obtained by known point of interest and its training of corresponding web search data.
4. the method according to claim 1, wherein the point of interest according to acquisition determines between point of interest
Incidence relation, comprising:
The co-occurrence information between the point of interest is determined in the Internet resources, and the interest is determined according to the co-occurrence information
Incidence relation between point.
5. the method according to claim 1, wherein the point of interest is social network user;
The point of interest according to acquisition determines the incidence relation between point of interest, comprising:
In social networks point of interest, the corresponding association user of the social network user is determined, and by the association user
It is associated with point of interest.
6. the method according to claim 1, wherein the point of interest according to acquisition determines between point of interest
Incidence relation, comprising:
The upperseat concept undetermined of the point of interest is determined according to Internet resources;
According to third classifier determine the upperseat concept undetermined whether be the point of interest true upperseat concept;
If so, determining that the true upperseat concept and the point of interest have is directed toward incidence relation.
7. according to the described in any item methods of claim 4-6, which is characterized in that the association according between the point of interest is closed
It is the target point of interest for determining user, comprising:
The historical interest point of the user is determined according to user's history data;
There is associated target point of interest with the historical interest point according to the incidence relation between the point of interest is determining.
8. according to the method described in claim 6, it is characterized in that, the incidence relation according between the point of interest is determined and is used
The target point of interest at family, comprising:
Send to the user terminal of the user includes the first point of interest inquiry message;
Receive the interest result that the user terminal returns;
An actual interest point is determined according to the interest result, and in the incidence relation, according to the actual interest point
Points relationship determines the target point of interest.
9. emerging according to the reality according to the method described in claim 8, it is characterized in that, described in the incidence relation
The points relationship of interest point determines the target point of interest, comprising:
The sub- point of interest for belonging to the actual interest point is determined in the incidence relation, and/or in the incidence relation, really
Father's point of interest of the fixed actual interest point;
Being sent according to the sub- point of interest and/or father's point of interest to the user terminal of the user includes the sub- point of interest
And/or the inquiry message of father's point of interest;
Receive the second interest result that the user terminal returns;
If determining that the user does not feel emerging to the sub- point of interest and/or father's point of interest according to the second interest result
Interest then continues to determine other the sub- points of interest for belonging to the actual interest point in the incidence relation, and/or in the association
In relationship, the step of determining other father's points of interest of the actual interest point.
10. a kind of point of interest determining device characterized by comprising
Module is excavated, for being excavated to Internet resources by excavation classifier trained in advance, obtains point of interest;
Relating module determines the incidence relation between point of interest for the point of interest according to acquisition;
Determining module, for determining the target point of interest of user according to the incidence relation between the point of interest.
11. device according to claim 10, which is characterized in that the excavation module, including the first excavation unit are used
In:
The web page characteristics in web data are extracted by the first excavation classifier, and determine the interest according to the web page characteristics
Point;
Wherein, the first excavation classifier is obtained by known point of interest and its training of corresponding web data.
12. device according to claim 10, which is characterized in that the excavation module includes the second excavation unit, is used for:
The search characteristics in web search data are extracted by the second excavation classifier, and according to the determination of described search feature
Point of interest;
Wherein, the second excavation classifier is obtained by known point of interest and its training of corresponding web search data.
13. device according to claim 10, which is characterized in that the relating module includes the first associative cell, is used for:
The co-occurrence information between the point of interest is determined in the Internet resources, and the interest is determined according to the co-occurrence information
Incidence relation between point.
14. device according to claim 10, which is characterized in that the point of interest is social network user;
The relating module includes the second associative cell, is used for:
In social networks point of interest, the corresponding association user of the social network user is determined, and by the association user
It is associated with point of interest.
15. device according to claim 10, which is characterized in that the relating module includes third associative cell, is used for:
The upperseat concept undetermined of the point of interest is determined according to Internet resources;
According to third classifier determine the upperseat concept undetermined whether be the point of interest true upperseat concept;
If so, determining that the true upperseat concept and the point of interest have is directed toward incidence relation.
16. the described in any item devices of 3-15 according to claim 1, which is characterized in that the determining module includes first determining
Unit is used for:
The historical interest point of the user is determined according to user's history data;
There is associated target point of interest with the historical interest point according to the incidence relation between the point of interest is determining.
17. device according to claim 15, which is characterized in that the determining module includes the second determination unit, is used for:
Send to the user terminal of the user includes the first point of interest inquiry message;
Receive the interest result that the user terminal returns;
An actual interest point is determined according to the interest result, and in the incidence relation, according to the actual interest point
Points relationship determines the target point of interest.
18. device according to claim 17, which is characterized in that second determination unit is specifically used for:
The sub- point of interest for belonging to the actual interest point is determined in the incidence relation, and/or in the incidence relation, really
Father's point of interest of the fixed actual interest point;
Being sent according to the sub- point of interest and/or father's point of interest to the user terminal of the user includes the sub- point of interest
And/or the inquiry message of father's point of interest;
Receive the second interest result that the user terminal returns;
If determining that the user does not feel emerging to the sub- point of interest and/or father's point of interest according to the second interest result
Interest then continues to determine other the sub- points of interest for belonging to the actual interest point in the incidence relation, and/or in the association
In relationship, the step of determining other father's points of interest of the actual interest point.
19. a kind of point of interest determines equipment characterized by comprising
Memory;
Processor;And
Computer program;
Wherein, the computer program stores in the memory, and is configured to be executed by the processor to realize such as power
Benefit requires any method of 1-9.
20. a kind of computer readable storage medium, which is characterized in that it is stored thereon with computer program,
The computer program is executed by processor to realize the method as described in claim 1-9 is any.
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 true CN110297967A (en) | 2019-10-01 |
CN110297967B 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) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110726418A (en) * | 2019-10-10 | 2020-01-24 | 北京百度网讯科技有限公司 | Method, device and equipment for determining interest point region and storage medium |
CN111160471A (en) * | 2019-12-30 | 2020-05-15 | 腾讯云计算(北京)有限责任公司 | Method and device for processing point of interest data, electronic equipment and storage medium |
WO2023226819A1 (en) * | 2022-05-23 | 2023-11-30 | 北京字节跳动科技有限公司 | Data matching method and apparatus, readable medium, and electronic device |
Citations (11)
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 |
CN103914536A (en) * | 2014-03-31 | 2014-07-09 | 北京百度网讯科技有限公司 | Interest point recommending method and system for electronic maps |
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 |
CN108875007A (en) * | 2018-06-15 | 2018-11-23 | 腾讯科技(深圳)有限公司 | The determination method and apparatus of point of interest, storage medium, electronic device |
-
2019
- 2019-05-14 CN CN201910398136.5A patent/CN110297967B/en active Active
Patent Citations (11)
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 |
CN103914536A (en) * | 2014-03-31 | 2014-07-09 | 北京百度网讯科技有限公司 | Interest point recommending method and system for electronic maps |
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 |
CN108875007A (en) * | 2018-06-15 | 2018-11-23 | 腾讯科技(深圳)有限公司 | The determination method and apparatus of point of interest, storage medium, electronic device |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110726418A (en) * | 2019-10-10 | 2020-01-24 | 北京百度网讯科技有限公司 | Method, device and equipment for determining interest point region and storage medium |
US11740102B2 (en) | 2019-10-10 | 2023-08-29 | Beijing Baidu Netcom Science Technology Co., Ltd. | Method, apparatus, device and storage medium for determining point of interest area |
CN111160471A (en) * | 2019-12-30 | 2020-05-15 | 腾讯云计算(北京)有限责任公司 | Method and device for processing point of interest data, electronic equipment 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 |
WO2023226819A1 (en) * | 2022-05-23 | 2023-11-30 | 北京字节跳动科技有限公司 | Data matching method and apparatus, readable medium, and electronic device |
Also Published As
Publication number | Publication date |
---|---|
CN110297967B (en) | 2022-04-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220351016A1 (en) | Presentation module for webinterface production and deployment system | |
US20220215032A1 (en) | Ai-based recommendation method and apparatus, electronic device, and storage medium | |
CN110580500B (en) | Character interaction-oriented network weight generation few-sample image classification method | |
US10713574B2 (en) | Cognitive distributed network | |
US9934514B2 (en) | Network aware product rollout in online social networks | |
US11263255B2 (en) | Content carousel in a social media timeline | |
CN110297967A (en) | Point of interest determines method, apparatus, equipment and computer readable storage medium | |
CN110597962A (en) | Search result display method, device, medium and electronic equipment | |
CN113569129A (en) | Click rate prediction model processing method, content recommendation method, device and equipment | |
CN112231570B (en) | Recommendation system support attack detection method, device, equipment and storage medium | |
CN107944026A (en) | A kind of method, apparatus, server and the storage medium of atlas personalized recommendation | |
US20170124200A1 (en) | Content recommendation | |
KR102284862B1 (en) | Method for providing video content for programming education | |
CN113742567B (en) | Recommendation method and device for multimedia resources, electronic equipment and storage medium | |
CN107111607A (en) | The system and method detected for language | |
CN113297486B (en) | Click rate prediction method and related device | |
CN111507471B (en) | Model training method, device, equipment and storage medium | |
CN114510592B (en) | Image classification method, device, electronic equipment and storage medium | |
CN113821677A (en) | Method, device and equipment for generating cover image and storage medium | |
WO2021064411A1 (en) | Visualisation system and methods | |
CN112308016A (en) | Expression image acquisition method and device, electronic equipment and storage medium | |
CN111046300A (en) | Method and device for determining crowd attributes of users | |
CN117993520B (en) | Data processing method, device, equipment and medium based on resource class | |
KR102647904B1 (en) | Method, system, and computer program for classify place review images based on deep learning | |
Ma et al. | Inferring social relationship in mobile social networks using tempo-spatial information |
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 |