CN108363733A - Content recommendation method and device - Google Patents
Content recommendation method and device Download PDFInfo
- Publication number
- CN108363733A CN108363733A CN201810043114.2A CN201810043114A CN108363733A CN 108363733 A CN108363733 A CN 108363733A CN 201810043114 A CN201810043114 A CN 201810043114A CN 108363733 A CN108363733 A CN 108363733A
- Authority
- CN
- China
- Prior art keywords
- information
- content
- locale
- position data
- content information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
-
- 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
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/52—Network services specially adapted for the location of the user terminal
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Abstract
The embodiment that this specification discloses provides a kind of content recommendation method, and this method includes：The position data of user is obtained, and according to position mapping model, determines Locale information corresponding with position data.Then, according at least to the characteristic information of each content information in determining Locale information and content information library, the content recommendation information recommended to the user is determined.
Description
Technical field
This specification disclose multiple embodiments be related to Internet technical field more particularly to a kind of content recommendation method and
Device.
Background technology
With the development of Internet technology, people's content information that more and more continually browse network platform provides.For example,
Merchandise news is browsed in shopping at network platform, either browses hot information in news platform or is managed in financing platform browsing
Wealth information etc..
For different user when using consolidated network platform, the demand of the content information provided it has more or less difference
It is different.On the other hand, the magnanimity growth of information also usually allows user to be difficult to select in the network platform.Currently, in recommended to the user
Hold information due to there is the deficiencies of not accurate enough, not prompt enough, it is difficult to meet the requirement of user.Accordingly, it is desirable to provide a kind of conjunction
The method of reason, to meet a variety of demands of the content information provided in user's browse network platform.
Invention content
Present specification describes a kind of content recommendation method and devices, according in the Locale information of user and content information library
The characteristic information of each content information determines content recommendation information recommended to the user, recommends to user in time more smart
Accurate content information.
In a first aspect, providing a kind of content recommendation method.This method includes：
Obtain the position data of user；
According to position mapping model, Locale information corresponding with the position data, the position mapping model base are determined
Acquisition is trained in the information on services obtained in advance, includes service position data and corresponding service location in the information on services
Data；
According at least to the characteristic information of each content information in the Locale information and content information library, determine to institute
State the content recommendation information of user's recommendation.
In a kind of possible embodiment, the position mapping model is trained by following steps and is obtained：
The service position data for including in the information on services are clustered, multiple class clusters are obtained；
According to service location data corresponding with the service position data in the information on services, the multiple class is determined
The corresponding site tabs of cluster.
In a kind of possible embodiment, determination Locale information corresponding with the position data includes：
Class cluster corresponding with the position data of the user is determined from the multiple class cluster；
According to the site tabs corresponding to the class cluster determined, the Locale information is determined.
In a kind of possible embodiment, the content recommendation information that the determination is recommended to the user, including：
Determine service type information corresponding with the Locale information；
According to the characteristic information of each content information, determines from the content information library and believe with the service type
Relevant content information is ceased, and using the content information as the content recommendation information.
In a kind of possible embodiment, the characteristic information includes the service type letter of the content information
Breath, the determining and relevant content information of service type information from the content information library, including：
The content information corresponding with the service type information is determined from the content information library.
In a kind of possible embodiment, the characteristic information of each content information includes predetermined interior with this
The corresponding Locale information of appearance information, the content recommendation information that the determination is recommended to the user, including：
When between the corresponding Locale information of each content information and Locale information corresponding with the position data
Subordinate relation when meeting preset rules, using the content information as the content recommendation information.
In a kind of possible embodiment, the characteristic information is based on the keyword extracted from the content information
Information and Locale information and determine.
In a kind of possible embodiment, the position data includes Wireless Fidelity WiFi finger print datas and the longitude and latitude number of degrees
At least one of according to.
In a kind of possible embodiment, the Locale information includes in building title, name of firm and commercial circle information
At least one.
Second aspect provides a kind of content recommendation device.The device includes：
Acquiring unit, the position data for obtaining user；
Determination unit, for according to position mapping model, determining Locale information corresponding with the position data, institute's rheme
It sets mapping model and trains acquisition based on the information on services obtained in advance, comprising service position data and right in the information on services
The service location data answered；
Processing unit, for the feature according at least to each content information in the Locale information and content information library
Information determines the content recommendation information recommended to the user.
In a kind of possible embodiment, the position mapping model is trained by the determination unit by following steps
It obtains：
The service position data for including in the information on services are clustered, multiple class clusters are obtained；
According to service location data corresponding with the service position data in the information on services, the multiple class is determined
The corresponding site tabs of cluster.
In a kind of possible embodiment, the determination unit specifically includes：
First determination subelement, for determining class corresponding with the position data of the user from the multiple class cluster
Cluster；
Second determination subelement determines the Locale information for the site tabs corresponding to the class cluster determined.
In a kind of possible embodiment, the processing unit specifically includes：
First processing subelement, for determining service type information corresponding with the Locale information；
Second processing subelement, for the characteristic information according to each content information, from the content information library
The determining and relevant content information of service type information, and using the content information as the content recommendation information.
In a kind of possible embodiment, the characteristic information in the second processing subelement includes the content information
The service type information, the second processing subelement is specifically used for：
The content information corresponding with the service type information is determined from the content information library.
In a kind of possible embodiment, the characteristic information for each content information that the processing unit includes includes
Predetermined Locale information corresponding with the content information, the processing unit are specifically used for：
When between the corresponding Locale information of each content information and Locale information corresponding with the position data
Subordinate relation when meeting preset rules, using the content information as the content recommendation information.
In a kind of possible embodiment, the characteristic information that the processing unit includes is based on from the content information
In the key word information that extracts and Locale information and determine.
In a kind of possible embodiment, the position data that the acquiring unit obtains includes Wireless Fidelity WiFi fingerprints
At least one of data and longitude and latitude degrees of data.
In a kind of possible embodiment, the Locale information that the determination unit determines includes building title, trade company's name
At least one of title and commercial circle information.
The third aspect provides a kind of computer readable storage medium, is stored thereon with computer program.When the calculating
When machine program executes in a computer, computer is enabled to execute the method that any embodiment provides in above-mentioned first aspect.
Fourth aspect provides a kind of computing device, including memory and processor.Being stored in the memory can hold
Line code when the processor executes the executable code, realizes any embodiment offer in above-mentioned first aspect
Method.
A kind of content recommendation method and device that this specification provides, by obtaining the position data of user, and according to position
Mapping model is set, determines Locale information corresponding with position data.Then, according at least to the Locale information and content information
The characteristic information of each content information in library determines the content recommendation information recommended to the user, in time to user
Recommend more accurately content information.
Description of the drawings
In order to illustrate more clearly of the technical solution for multiple embodiments that this specification discloses, embodiment will be described below
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description be only this specification disclose
Multiple embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 is a kind of application scenarios schematic diagram for content recommendation method that one embodiment that this specification discloses provides；
Fig. 2 is a kind of flow chart for content recommendation method that one embodiment that this specification discloses provides；
Fig. 3 be this specification disclose one embodiment provide it is a kind of by service position data clusters at multiple class clusters
Schematic diagram；
Fig. 4 is the flow chart that a kind of characteristic information that one embodiment that this specification discloses provides determines method；
Fig. 5 is a kind of schematic diagram for content recommendation device that one embodiment that this specification discloses provides.
Specific implementation mode
Below in conjunction with the accompanying drawings, the multiple embodiments disclosed this specification are described.
Fig. 1 is a kind of application scenarios schematic diagram for content recommendation method that one embodiment that this specification discloses provides.
The executive agent of the recommendation method can be server.When user by terminal (e.g., terminal can be mobile phone, tablet computer,
Wearable smart machine etc.) log in commending contents platform (e.g., Alipay application platform) when, may be used this specification disclosure
The recommendation method for the business function that multiple embodiments provide, server obtain the position data (e.g., longitude and latitude degrees of data) of user, and
According to position mapping model (e.g., position mapping model may include position data and the mapping relations of Locale information), determine with
The corresponding Locale information of the position data (e.g., 4 S auto shop).Then, according to each in the Locale information and content information library
The characteristic information (e.g., characteristic information may include service type information) of a content information, determines that content recommended to the user pushes away
Recommend information (e.g., Motor Maintenance knowledge etc.).
The recommendation method for the business function that multiple embodiments that this specification discloses provide, by the positional number for obtaining user
According to, and according to position mapping model, determine Locale information corresponding with position data.Then, according at least to the Locale information, with
And in content information library each content information characteristic information, determine the content recommendation information recommended to the user, thus and
When recommend more accurately content information to user.
Fig. 2 is a kind of flow chart for content recommendation method that one embodiment that this specification discloses provides.The method
Executive agent can be the equipment with processing capacity：Server either system or device, for example, clothes as shown in Figure 1
Business device.As shown in Fig. 2, the method specifically includes：
Step S210 obtains the position data of user.
Specifically, position data can be by location based service (Location Based Service, abbreviation
LBS) the data acquired from the terminal of user.LBS includes multiple positioning modes, such as global positioning system (Global
Positioning System, abbreviation GPS) positioning, base station location, Wireless Fidelity (Wireless Fidelity, abbreviation WiFi)
Positioning etc..
In one embodiment, position data may include the longitude and latitude degrees of data obtained by GPS positioning or base station location.
For example, the longitude and latitude degrees of data that position data includes is：39 ° 54 ' 25.70 of north latitude " and east longitude 116 ° 23 ' 28.49 ".
In another embodiment, position data may include the WiFi finger print datas positioned by WiFi.One
In a example, WiFi finger print datas may include address (e.g., the media visit of WiFi access points (Access Point, abbreviation AP)
Ask the address control (Media Access Control, abbreviation MAC)) and corresponding signal strength etc..
Step S220 determines Locale information corresponding with the position data of user according to position mapping model.
Specifically, Locale information and position data the difference is that, Locale information, which is user, can perceive, have
There is the location information of explicit semantic meaning.In one embodiment, Locale information may include building title (e.g., New Zhongguan Building),
Name of firm (e.g., the toyshops Ha Mu) and commercial circle information (e.g., in three inner village romote antiquity) etc..In another embodiment, place is believed
Ceasing can also include and place (e.g., trade company) relevant service type information (e.g., cuisines, clothes, beauty, film etc.).
Conversion from position data to Locale information can be executed according to the position mapping model being obtained ahead of time.At one
In embodiment, above-mentioned position mapping model includes the mapping relations between GPS coordinate data and site tabs, and the mapping relations are pre-
It first passes through artificial acquisition and obtains.
In another embodiment, above-mentioned position mapping model can be based on the service for including in the information on services obtained in advance
Position data and corresponding service location data and train acquisition.Service corresponding with information on services may include payment services
(e.g., being paid to order using payment application) and positioning service are (e.g., when social platform sends status information to present bit
Set and positioned) etc..
Wherein, information on services can obtain in several ways.In one embodiment, service corresponding with information on services
May include payment services, corresponding information on services may include payment information.Payment information is obtained, may include：Server
When detecting that user is submitted order or paid to order by terminal, payment information can be acquired, specifically includes and passes through
The position data of acquisition terminal is as the service position data in payment information, and acquisition and service bit from the order of submission
Set the corresponding service location information of data.In one example, user uses the payment application having connected in the terminal of WiFi
When (e.g., Alipay) carries out delivery operation, the server of payment application can pass through the position of WiFi positioning acquisitions terminal at this time
Data are set, the service location data (e.g., McDonald Xueyuan Road shop) that (e.g., WiFi finger print datas) and payment information include.
In another embodiment, service corresponding with information on services may include positioning service, corresponding information on services
It may include location services information.Location services information is obtained, may include：Server is detecting that user used by terminal
When positioning service, location services information can be obtained, specifically includes the position data of acquisition terminal as in location services information
Service position data, and obtain the location information that user is selected when using positioning service or creates.In one example,
User is using supporting the terminal of GPS positioning to log in job applications (e.g., nail nail) and carry out working punching operation, at this point it is possible to pass through
The service location data that the position data (e.g., longitude and latitude degrees of data) of GPS positioning acquisition terminal and information of checking card include (e.g., add
State gourmet restaurant).In another example, it includes place that user, which is issued using the terminal for being connected with Cellular Networks in social platform,
The state of information, at this point it is possible to selected by the position data (e.g., longitude and latitude degrees of data) of base station location acquisition terminal and user
The service location data (e.g., one or two coffee-house) of (or establishment).
Further, based on the information on services obtained in advance, position mapping model can be trained by following steps and be obtained：
The service position data for including in information on services are clustered, multiple class clusters are obtained.Then, according in information on services with service
The corresponding service location data of position data determine the corresponding site tabs of multiple class clusters.In this way, can establish multiple class clusters with
Mapping relations between multiple site tabs.
In one embodiment, clustering algorithm may be used to cluster service position data, and obtain multiple class clusters.
Wherein clustering algorithm can be GEOHASH algorithms or DBSCAN algorithms, be not limited thereto.
In one example, service position data include multiple longitude and latitude degrees of data, and GEOHASH algorithms may be used at this time will
The data conversion determines the grid number of each longitude and latitude degrees of data at GEOHASH grids.For example, service position data include
100 latitude and longitude coordinates, after this 100 latitude and longitude coordinates is converted into GEOHASH grids using GEOHASH algorithms, such as Fig. 3 institutes
Show, every 20 latitude and longitude coordinates grid number having the same, obtained 5 grids number is respectively：WX4G01、WX4H02、
WX4I03、WX4J04、WX4K05.In this way, above 100 latitude and longitude coordinates are clustered into 5 class clusters.
In another example, service position data include multiple WiFi finger print datas, and DBSCAN calculations may be used at this time
Method clusters the data, obtains multiple class clusters with different numbers.DBSCAN algorithms are a kind of density clusterings
Algorithm.Different from division and hierarchy clustering method, cluster is defined as the maximum set of the connected point of density by it, can be with foot
Enough highdensity region divisions are cluster, and the cluster of arbitrary shape can be found in the spatial database of noise.Specifically,
In DBSCAN algorithms, all location points are labeled as core point, boundary point or noise spot first, delete noise spot therein.So
It is that distance assigns a line between all core points within parameter preset afterwards, the core point of every group of connection forms a cluster,
Each boundary point is assigned in the cluster for the core point that one is associated, thus completes the cluster of location point.
It is understood that the precision for the class cluster that cluster obtains by adjusting the parameter of clustering algorithm, can be controlled (e.g.,
The range size of class cluster).For example, can be by controlling the bits of coded encoded to service position data in GEOHASH algorithms
Number controls the precision of class cluster.More specifically, the number of encoding bits of service position data are more, obtained class cluster range is more accurate.Again
For example, the precision of class cluster can be controlled by the size of radius of neighbourhood ε in control DBSCAN algorithms.More specifically, the neck of input
The value of domain radius ε is smaller, and obtained class cluster range is more accurate.
In one embodiment, to cluster obtain multiple class clusters, can according in information on services with service position data
Corresponding service location data determine the site tabs corresponding to each class cluster in multiple class clusters.
In one example, service location data corresponding with service position data include the title of service location (e.g.,
Happy supermarket), then the corresponding site tabs of class cluster where the name of this service location can be referred to as to the service position data.
In another example, can correspond to multiple service position data in single class cluster, and with multiple service position numbers
It may be different according to the range (e.g., trade company or commercial circle) that corresponding multiple service location data are identified.Correspondingly, single class cluster can
With with multiple site tabs, e.g., trade company's label and commercial circle label etc..For example, the site tabs of some class cluster may include quotient
Family label (e.g., western Ma Hua) and commercial circle label (e.g., Da Zhongsi) etc..
In this step, determines Locale information corresponding with the position data of user, may include：It is obtained from cluster multiple
Class cluster corresponding with the position data of user is determined in class cluster.Then, the site tabs corresponding to the class cluster determined, really
Determine Locale information.
In one embodiment, position mapping model includes the positional number of the central point of each class cluster in multiple class clusters
According to and mapping relations between multiple class clusters and multiple site tabs.In this way, determined from multiple class clusters for obtaining of cluster with
The corresponding class cluster of position data of user may include：Between the position data and the position data of each central point that calculate user
Distance, using the class cluster corresponding to the minimum value of calculated distance as class cluster corresponding with the position data of user.
Further, the site tabs corresponding to the class cluster determined, determine Locale information, may include：According to
Mapping relations between multiple class clusters and multiple site tabs, and class cluster corresponding with user position data that is determining, really
Fixed site tabs corresponding with such cluster, and then determine Locale information.
In one embodiment, the site tabs determined include trade company's label (e.g., western Ma Hua), can be determined therefrom that
Go out the Locale information for including name of firm (e.g., western Ma Hua).
After determining Locale information corresponding with the position data of user in step S220, then, in step S230, until
Few characteristic information according to each content information in Locale information and content information library determines that content recommended to the user pushes away
Recommend information.
Specifically, it is determined that content recommendation information recommended to the user, may include：Determine service corresponding with Locale information
Classification information, and according to the characteristic information of each content information, determination is related to service type information from content information library
Content information, and using the content information as content recommendation information.
In one embodiment, the determining and relevant content information of service type information from content information library, can wrap
It includes：According to the characteristic information of each content information, the degree of correlation of each content information and service type information is calculated, and according to meter
The degree of correlation of calculating determines and the relevant content information of service type information.
In one example, can by the degree of correlation within a predetermined range (e.g., preset range can be more than content 0.6)
Information is determined as and the relevant content information of service type information.In another example, content can be believed according to the degree of correlation
Breath carries out ranking, and content information of the ranking in preset range (e.g., preset range can be TOP V) is determined as and is taken
The business relevant content information of classification information.
In another embodiment, characteristic information may include the service type information of content information.From content information library
Middle determination and the relevant content information of service type information may include：It is determined and service type information from content information library
Corresponding content information.
In one example, in step S220 determine Locale information be " 4 S auto shop ", accordingly, it may be determined that go out with
The corresponding service type information of the Locale information is " automobile ".Correspondingly, it can determine that service type is believed from content information library
Breath is the content information of " automobile ", and using this content information as content recommendation information recommended to the user.
Alternatively, may include predetermined place corresponding with the content information in the characteristic information of each content information
Information.In this case, it determines content recommendation information recommended to the user, may include：When with each content information pair
When subordinate relation between the Locale information and Locale information corresponding with position data answered meets preset rules, which is believed
Breath is used as content recommendation information.
In one embodiment, preset rules can according to the attribute of business, (e.g., the business be to place range by business root
Required precision) determine.For example, when business is to push coupons to user, preset rules may include requirement with it is interior
It is identical to hold the corresponding Locale information of recommendation information Locale information corresponding with position data.When business is attached to user's recommendation
When close similar place, preset rules may include requirement Locale information corresponding with content recommendation information and position data pair
The Locale information part answered is identical.
In one example, preset rules include requiring Locale information corresponding with content recommendation information and position data pair
The Locale information answered is identical, and Locale information corresponding with position data includes " old nice fragrant pot ", " San Litun ".Accordingly
Ground, can will content information corresponding with Locale information " old nice fragrant pot " (trade company), " San Litun " (commercial circle) (e.g., trade company is excellent
Favour activity, trade company's public platform information, trade company's evaluation information) it is used as content recommendation information.
In another example, preset rules may include requirement Locale information corresponding with content recommendation information and position
The corresponding Locale information part of data is identical, and Locale information corresponding with position data includes " excellent clothing library " (trade company), " in three
Village " (commercial circle).It correspondingly, can will content information (e.g., the clothes in three commercial circles Li Tun corresponding with Locale information " San Litun "
Shop, dining room etc.) it is used as content recommendation information.
After determining content recommendation information recommended to the user through the above steps, can also include：Into user's transmission
Hold recommendation information.When applying App accordingly for example, working as user and opening, or opening recommendation channel, or when the position for detecting user
When changing, the newest location information of user is obtained, determines therefrom that content recommendation information recommended to the user, and send out to user
Give the content recommendation information.
It should be noted that in step S220, can also include：Determine service type letter corresponding with Locale information
Breath.In one embodiment, can also include service class corresponding with place in Locale information other than the title including place
Other information correspondingly can directly determine service type information corresponding with Locale information accordingly.
It in another embodiment, can be after determining Locale information, according to the multiple Locale informations and multiple clothes to prestore
The correspondence for classification information of being engaged in, determines the corresponding service type information of the Locale information.In one example, the field determined
Institute's label includes trade company's label (e.g., western Ma Hua), can determine therefrom that out including name of firm (e.g., western Ma Hua) and
Service type information (e.g., cuisines) corresponding with name of firm.
The characteristic information referred in step S230 can be based on the key word information and/or position extracted from content information
Confidence is ceased and is determined, method as shown in Figure 4 specifically may be used and be determined：
Step S410, pre-processes content information.
Specifically, pretreatment may include structured analysis, word segmentation processing, remove stop words processing, part-of-speech tagging
(postag) etc..Wherein, structured analysis may include the analysis to paragraph structure in content information, for example, judging content
Title in information and text；Word segmentation processing may include one-gram word (unigram), binary participle (bigram), ternary point
Word (trigram) etc.；It may include according to stop words (e.g., the nothing in preset deactivated vocabulary removal content information to remove stop words
The function word of practical significance：This, that)；Part-of-speech tagging refer to the part of speech of the word in content information (e.g., noun, adverbial word,
Adjective etc.) it is labeled.
Step S420 extracts key word information according to pretreated content information.
Specifically, (e.g., in title or can be located at according to position of the obtained word of pretreatment in content information
In text), mark part of speech, using TextRank algorithm or TF-IDF (Term Frequency-inverse Document
Frequency) algorithm weights identify key word information.
Step S430 extracts location information according to pretreated content information.
Specifically, location information can be extracted according to the location tags that content information includes, and can be wrapped in location information
Include Locale information.In one embodiment, the location tags can be content information the producer when issuing the content information,
For the location tags that it is pasted, e.g., Beijing, Hangzhou etc..
Alternatively, the method that name Entity recognition (Named Entity Recognition, abbreviation NER) may be used, knows
Place name, mechanism name in other content information etc..For example, the entitled Wutai Mountain in ground in content information can be identified, mechanism is entitled
Haidian District Department of Civil Affairs (belongs to Locale information).
Or location information library can be previously stored in server, it can be accordingly from pretreated content information
In extract location information with the information matches in the location information library.
Step S440, by the key word information tentatively extracted and location information input Feature Selection Model trained in advance
In, determine characteristic information.
Specifically, Feature Selection Model can be to carry out off-line training, and the word of output to extensive content corpus data
Embedded (Word Embedding) model is based on bidirectional circulating neural network (Recurrent Neural Networks, abbreviation
RNN NER models).
In addition, the determination of features described above information can content information generation after predetermined time (e.g., 5min or
It is carried out in 10min).
From the foregoing, it will be observed that in the content recommendation method that multiple embodiments that this specification discloses provide, by obtaining user
Position data, and according to position mapping model, determine Locale information corresponding with position data.Then, according at least to this
The characteristic information of each content information in institute's information and content information library determines the commending contents letter recommended to the user
Breath, to recommend more accurately content information to user in time.
Accordingly with content recommendation method, multiple embodiments that this specification discloses also provide a kind of content recommendation device,
As shown in figure 5, the device 500 includes：
Acquiring unit 510, the position data for obtaining user；
Determination unit 520, for according to position mapping model, determining Locale information corresponding with position data, the position
Mapping model trains acquisition based on the information on services obtained in advance, comprising service position data and corresponding in the information on services
Service location data；
Processing unit 530 is believed for the feature according at least to each content information in Locale information and content information library
Breath, determines content recommendation information recommended to the user.
In a kind of possible embodiment, position mapping model is trained by following steps by determination unit 520 and is obtained：
The service position data for including in information on services are clustered, multiple class clusters are obtained；
According to service location data corresponding with service position data in information on services, the corresponding place of multiple class clusters is determined
Label.
In a kind of possible embodiment, determination unit 520 specifically includes：
First determination subelement 521, for determining class cluster corresponding with the position data of user from multiple class clusters；
Second determination subelement 522 determines Locale information for the site tabs corresponding to the class cluster determined.
In a kind of possible embodiment, processing unit 530 specifically includes：
First processing subelement 531, for determining service type information corresponding with Locale information；
Second processing subelement 532, for according to the characteristic information of each content information, determined from content information library with
The relevant content information of service type information, and using content information as content recommendation information.
In a kind of possible embodiment, the characteristic information in second processing subelement 532 includes the clothes of content information
Business classification information, second processing subelement 532 are specifically used for：
Content information corresponding with service type information is determined from content information library.
In a kind of possible embodiment, the characteristic information for each content information that processing unit 530 includes includes
Predetermined Locale information corresponding with the content information, processing unit 530 are specifically used for：
When the subordinate between the corresponding Locale information of each content information and Locale information corresponding with position data is closed
When system meets preset rules, using the content information as content recommendation information.
In a kind of possible embodiment, the characteristic information that processing unit 530 includes is based on carrying from content information
Key word information and the Locale information of taking-up and determine.
In a kind of possible embodiment, which is characterized in that the position data that acquiring unit 510 obtains includes wireless protects
True at least one of WiFi finger print datas and longitude and latitude degrees of data.
In a kind of possible embodiment, the Locale information that determination unit 520 determines includes building title, name of firm
At least one of with commercial circle information.
From the foregoing, it will be observed that in the content recommendation device that multiple embodiments that this specification discloses provide, acquiring unit 510 obtains
The position data at family is taken, determination unit 520 determines Locale information corresponding with position data according to position mapping model.Place
Characteristic information of the unit 530 according at least to each content information in the Locale information and content information library is managed, is determined to described
The content recommendation information that user recommends, to recommend more accurately content information to user in time.
It will be appreciated that in said one or multiple examples, this specification discloses more those skilled in the art
A embodiment described function can be realized with hardware, software, firmware or their arbitrary combination.When using software realization
When, these functions can be stored in computer-readable medium or be referred to as the one or more on computer-readable medium
It enables or code is transmitted.
Above-described specific implementation mode to the purpose of multiple embodiments of this specification disclosure, technical solution and has
Beneficial effect has been further described, it should be understood that the foregoing is merely multiple embodiments that this specification discloses
Specific implementation mode, be not used to limit this specification disclose multiple embodiments protection domain, it is all in this explanation
On the basis of the technical solution for multiple embodiments that book discloses, any modification, equivalent substitution, improvement and etc. done should all wrap
It includes within the protection domain for multiple embodiments that this specification discloses.
Claims (18)
1. a kind of content recommendation method, which is characterized in that including：
Obtain the position data of user；
According to position mapping model, determine that Locale information corresponding with the position data, the position mapping model are based on pre-
The information on services that first obtains and train acquisition, include service position data and corresponding service location number in the information on services
According to；
According at least to the characteristic information of each content information in the Locale information and content information library, determine to the use
The content recommendation information that family is recommended.
2. according to the method described in claim 1, it is characterized in that, the position mapping model is obtained by following steps training
：
The service position data for including in the information on services are clustered, multiple class clusters are obtained；
According to service location data corresponding with the service position data in the information on services, the multiple class cluster pair is determined
The site tabs answered.
3. according to the method described in claim 2, it is characterized in that, determination Locale information corresponding with the position data
Including：
Class cluster corresponding with the position data of the user is determined from the multiple class cluster；
According to the site tabs corresponding to the class cluster determined, the Locale information is determined.
4. according to the method described in claim 1, it is characterized in that, the commending contents letter that the determination is recommended to the user
Breath, including：
Determine service type information corresponding with the Locale information；
According to the characteristic information of each content information, determined and the service type information phase from the content information library
The content information of pass, and using the content information as the content recommendation information.
5. according to the method described in claim 4, it is characterized in that, the characteristic information includes the clothes of the content information
Business classification information, the determining and relevant content information of service type information from the content information library, including：
The content information corresponding with the service type information is determined from the content information library.
6. according to the method described in claim 1, it is characterized in that, the characteristic information of each content information includes true in advance
Fixed Locale information corresponding with the content information, the content recommendation information that the determination is recommended to the user, including：
When between the corresponding Locale information of each content information and Locale information corresponding with the position data from
When category relationship meets preset rules, using the content information as the content recommendation information.
7. according to claim 1-6 any one of them methods, which is characterized in that the characteristic information is based on believing from the content
The key word information and the Locale information that are extracted in breath and determine.
8. according to the method described in any one of claim 1-6, which is characterized in that the position data includes Wireless Fidelity
At least one of WiFi finger print datas and longitude and latitude degrees of data.
9. according to claim 1-6 any one of them methods, which is characterized in that the Locale information includes building title, quotient
Name in an account book at least one of claims with commercial circle information.
10. a kind of content recommendation device, which is characterized in that including：
Acquiring unit, the position data for obtaining user；
Determination unit, for according to position mapping model, determining Locale information corresponding with the position data, the position is reflected
It penetrates model and trains acquisition based on the information on services obtained in advance, comprising service position data and corresponding in the information on services
Service location data；
Processing unit is used for the characteristic information according at least to each content information in the Locale information and content information library,
Determine the content recommendation information recommended to the user.
11. device according to claim 10, which is characterized in that the position mapping model is passed through by the determination unit
Following steps training obtains：
The service position data for including in the information on services are clustered, multiple class clusters are obtained；
According to service location data corresponding with the service position data in the information on services, the multiple class cluster pair is determined
The site tabs answered.
12. according to the devices described in claim 11, which is characterized in that the determination unit specifically includes：
First determination subelement, for determining class cluster corresponding with the position data of the user from the multiple class cluster；
Second determination subelement determines the Locale information for the site tabs corresponding to the class cluster determined.
13. device according to claim 10, which is characterized in that the processing unit specifically includes：
First processing subelement, for determining service type information corresponding with the Locale information；
Second processing subelement is determined for the characteristic information according to each content information from the content information library
With the relevant content information of service type information, and using the content information as the content recommendation information.
14. device according to claim 13, which is characterized in that the characteristic information in the second processing subelement includes
The service type information of the content information, the second processing subelement are specifically used for：
The content information corresponding with the service type information is determined from the content information library.
15. device according to claim 10, which is characterized in that each content information that the processing unit includes
Characteristic information includes predetermined Locale information corresponding with the content information, and the processing unit is specifically used for：
When between the corresponding Locale information of each content information and Locale information corresponding with the position data from
When category relationship meets preset rules, using the content information as the content recommendation information.
16. according to claim 10-15 any one of them devices, which is characterized in that the feature that the processing unit includes
Information is determined based on key word information and the Locale information extracted from the content information.
17. according to the device described in any one of claim 10-15, which is characterized in that the position that the acquiring unit obtains
Data include at least one of Wireless Fidelity WiFi finger print datas and longitude and latitude degrees of data.
18. according to claim 10-15 any one of them devices, which is characterized in that the place letter that the determination unit determines
Breath includes at least one of building title, name of firm and commercial circle information.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810043114.2A CN108363733A (en) | 2018-01-17 | 2018-01-17 | Content recommendation method and device |
| TW107144112A TWI703862B (en) | 2018-01-17 | 2018-12-07 | Content recommendation method and device |
| PCT/CN2019/070830 WO2019141109A1 (en) | 2018-01-17 | 2019-01-08 | Method and device for content recommendation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810043114.2A CN108363733A (en) | 2018-01-17 | 2018-01-17 | Content recommendation method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108363733A true CN108363733A (en) | 2018-08-03 |
Family
ID=63006224
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810043114.2A Pending CN108363733A (en) | 2018-01-17 | 2018-01-17 | Content recommendation method and device |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN108363733A (en) |
| TW (1) | TWI703862B (en) |
| WO (1) | WO2019141109A1 (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409959A (en) * | 2018-10-31 | 2019-03-01 | 广州品唯软件有限公司 | A kind of user information analysis method, device, equipment and medium |
| WO2019141109A1 (en) * | 2018-01-17 | 2019-07-25 | 阿里巴巴集团控股有限公司 | Method and device for content recommendation |
| CN110377195A (en) * | 2019-07-15 | 2019-10-25 | 腾讯科技（深圳）有限公司 | The method and apparatus for showing interactive function |
| CN110909250A (en) * | 2018-09-14 | 2020-03-24 | 阿里巴巴集团控股有限公司 | Information processing method and device, storage medium and processor |
| WO2020187070A1 (en) * | 2019-03-19 | 2020-09-24 | 腾讯科技（深圳）有限公司 | Region division method and device, storage medium and electronic device |
| CN111815361A (en) * | 2020-07-10 | 2020-10-23 | 北京思特奇信息技术股份有限公司 | Region boundary calculation method and device, electronic equipment and storage medium |
| CN112395486A (en) * | 2019-08-12 | 2021-02-23 | 中国移动通信集团重庆有限公司 | Broadband service recommendation method, system, server and storage medium |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102412057B1 (en) * | 2021-06-07 | 2022-06-23 | 쿠팡 주식회사 | Operating method for electronic apparatus for providing store information and electronic apparatus supporting thereof |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106776776A (en) * | 2016-11-11 | 2017-05-31 | 广东小天才科技有限公司 | A kind of recommendation method and device of sports center information |
| CN107391605A (en) * | 2017-06-30 | 2017-11-24 | 北京奇虎科技有限公司 | Information-pushing method, device and mobile terminal based on geographical position |
| CN107545052A (en) * | 2017-08-23 | 2018-01-05 | 广东欧珀移动通信有限公司 | Information recommendation method, device, mobile terminal and storage medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8229458B2 (en) * | 2007-04-08 | 2012-07-24 | Enhanced Geographic Llc | Systems and methods to determine the name of a location visited by a user of a wireless device |
| US20090094189A1 (en) * | 2007-10-08 | 2009-04-09 | At&T Bls Intellectual Property, Inc. | Methods, systems, and computer program products for managing tags added by users engaged in social tagging of content |
| CN103023977B (en) * | 2012-11-19 | 2015-07-01 | 华南理工大学 | Recommendation system and method based on positional information |
| US9699491B1 (en) * | 2014-10-10 | 2017-07-04 | ThinkAnalytics | Content recommendation engine |
| CN106294489B (en) * | 2015-06-08 | 2022-09-30 | 北京三星通信技术研究有限公司 | Content recommendation method, device and system |
| CN108363733A (en) * | 2018-01-17 | 2018-08-03 | 阿里巴巴集团控股有限公司 | Content recommendation method and device |
-
2018
- 2018-01-17 CN CN201810043114.2A patent/CN108363733A/en active Pending
- 2018-12-07 TW TW107144112A patent/TWI703862B/en active
-
2019
- 2019-01-08 WO PCT/CN2019/070830 patent/WO2019141109A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106776776A (en) * | 2016-11-11 | 2017-05-31 | 广东小天才科技有限公司 | A kind of recommendation method and device of sports center information |
| CN107391605A (en) * | 2017-06-30 | 2017-11-24 | 北京奇虎科技有限公司 | Information-pushing method, device and mobile terminal based on geographical position |
| CN107545052A (en) * | 2017-08-23 | 2018-01-05 | 广东欧珀移动通信有限公司 | Information recommendation method, device, mobile terminal and storage medium |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019141109A1 (en) * | 2018-01-17 | 2019-07-25 | 阿里巴巴集团控股有限公司 | Method and device for content recommendation |
| CN110909250B (en) * | 2018-09-14 | 2023-05-02 | 阿里巴巴集团控股有限公司 | Information processing method and device, storage medium and processor |
| CN110909250A (en) * | 2018-09-14 | 2020-03-24 | 阿里巴巴集团控股有限公司 | Information processing method and device, storage medium and processor |
| CN109409959A (en) * | 2018-10-31 | 2019-03-01 | 广州品唯软件有限公司 | A kind of user information analysis method, device, equipment and medium |
| WO2020187070A1 (en) * | 2019-03-19 | 2020-09-24 | 腾讯科技（深圳）有限公司 | Region division method and device, storage medium and electronic device |
| CN111723959A (en) * | 2019-03-19 | 2020-09-29 | 腾讯科技（深圳）有限公司 | Region dividing method, region dividing device, storage medium and electronic device |
| CN111723959B (en) * | 2019-03-19 | 2023-12-12 | 腾讯科技（深圳）有限公司 | Region dividing method and device, storage medium and electronic device |
| US11966424B2 (en) | 2019-03-19 | 2024-04-23 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for dividing region, storage medium, and electronic device |
| CN110377195A (en) * | 2019-07-15 | 2019-10-25 | 腾讯科技（深圳）有限公司 | The method and apparatus for showing interactive function |
| CN110377195B (en) * | 2019-07-15 | 2022-09-30 | 腾讯科技（深圳）有限公司 | Method and device for displaying interaction function |
| CN112395486A (en) * | 2019-08-12 | 2021-02-23 | 中国移动通信集团重庆有限公司 | Broadband service recommendation method, system, server and storage medium |
| CN112395486B (en) * | 2019-08-12 | 2023-11-03 | 中国移动通信集团重庆有限公司 | Broadband service recommendation method, system, server and storage medium |
| CN111815361A (en) * | 2020-07-10 | 2020-10-23 | 北京思特奇信息技术股份有限公司 | Region boundary calculation method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019141109A1 (en) | 2019-07-25 |
| TWI703862B (en) | 2020-09-01 |
| TW201933879A (en) | 2019-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108363733A (en) | Content recommendation method and device | |
| CN106462565B (en) | Text is updated in document | |
| CN102142003B (en) | Method and device for providing point of interest information | |
| CN106033466A (en) | Database query method and device | |
| CN104731917A (en) | Recommendation method and device | |
| CN107292365A (en) | Binding method, device, equipment and the computer-readable recording medium of Commercial goods labelses | |
| CN103473230A (en) | Service range determining method, logistics service provider recommending method and corresponding device | |
| CN105677780A (en) | Scalable user intent mining method and system thereof | |
| CN107027100A (en) | The method and system of accessible location is marked based on associated person information | |
| JP4950508B2 (en) | Facility information management system, facility information management device, facility information management method, and facility information management program | |
| US9330135B2 (en) | Method, apparatus and computer readable recording medium for a search using extension keywords | |
| CN111680144A (en) | Method and system for multi-turn dialogue voice interaction, storage medium and electronic equipment | |
| CN106204190B (en) | Information processing method, terminal and server | |
| CA2709116A1 (en) | Method and system for enabling location entry | |
| CN104079640A (en) | User service delivery method and system | |
| US20160241497A1 (en) | Invoking an application to perform a service based on message content | |
| CN106164941A (en) | The additional information relevant to fuzzy item in the message is provided | |
| CN103631946A (en) | Content pushing system based on geographic positions | |
| CN109726295A (en) | Brand knowledge map display methods, device, figure server and storage medium | |
| CN106202440B (en) | Data processing method, device and equipment | |
| CN112819544A (en) | Advertisement putting method, device, equipment and storage medium based on big data | |
| US9336237B2 (en) | Content distribution | |
| CN110990714A (en) | User behavior intention prediction method and device | |
| US9449110B2 (en) | Geotiles for finding relevant results from a geographically distributed set | |
| KR100806698B1 (en) | Method for storing content database reflecting use-log of user |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1257388 Country of ref document: HK |
|
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180803 |
|
| RJ01 | Rejection of invention patent application after publication |