Embodiment
Below in conjunction with accompanying drawing, the technical scheme that the embodiment of the present invention provides is described in detail.
Figure 1 shows that a kind of method obtaining user interest label that the embodiment of the present invention provides, specifically comprise following operation:
The user behavior data of the first smart machine in step 100, acquisition object time section.
In the embodiment of the present invention, the first smart machine refers to the smart machine needing to obtain user interest label.Preferably, the first smart machine is the less smart machine of intelligent television, intelligent earphone etc. user's input information.
In the embodiment of the present invention, the user behavior data of the first smart machine, refers to that user uses the data of the first smart machine.Can but be not limited only to comprise following at least one: the data of the viewing data of behavior, the data listening to behavior, Shopping Behaviors, the data of splitting glass opaque, the data of collection behavior, the data downloading behavior etc.
In the embodiment of the present invention, preferably, object time section refers to certain time period in the middle of one day.Certainly, object time section also can be one day, such as using Monday as object time section, etc.
In the embodiment of the present invention, the behavior time of origin of user behavior data record is in object time section, then this user behavior data is exactly the user behavior data of the first smart machine in object time section.
Step 110, according to the user behavior data of above-mentioned first smart machine obtained, generate the user interest label of above-mentioned object time section correspondence, this user interest label at least indicates the interested first content theme of user.
In the embodiment of the present invention, content topic refers to the theme belonging to the content that the first smart machine provides; First content theme be first smart machine of behavior time of origin in object time section user behavior data for content belonging to theme.
In the embodiment of the present invention, user can not be concrete someone.Because the first smart machine may multiple user share, therefore, user refers to all users using this first smart machine.Certainly, user also can refer to concrete someone, and this just needs the different user distinguishing use first smart machine.
Step 120, obtain the social network user data of the second smart machine of above-mentioned first smart machine place LAN (Local Area Network), these social network user data comprise above-mentioned user at the log-on message of social networks and/or the above-mentioned user user behavior data at social networks.
In the embodiment of the present invention, user social networks log-on message can but be not limited only to comprise following at least one item: the occupation of the age of user, the sex of user, user, the region of user, user tag (such as good-for-nothing, animation fan, CS intelligent) etc.
In the embodiment of the present invention, user social networks user behavior data can but be not limited only to comprise following at least one item: user makes a speech the data of behavior on social networks, such as, on the social networks such as microblogging, micro-letter, bean cotyledon network, carry out comment on data for content; User marks the data etc. of behavior on social networks.
Step 130, according to obtain the above-mentioned user interest label of above-mentioned social network user Data Update.
In above-mentioned flow process, do not limit step 120 and perform after step 110, if before step 130 completing steps 120.
Inventor is realizing finding in process of the present invention, and user behavior data can reflect the interested content topic of user.Such as, user watches certain this behavior of film by smart machine, may reflect user interested in this content topic of film.When the interested content topic of user cannot be determined according to the fragmented information of user's input, adopt the method that the embodiment of the present invention provides, by obtaining the user behavior data of the first smart machine in object time section, just can according to the user behavior data obtained, generate the user interest label of this object time section correspondence, indicating user is at the interested first content theme of this object time section.Because the user behavior data of the first smart machine is preserved in the mode of user behaviors log usually, therefore, obtain the user behavior data that user behaviors log can obtain the first smart machine, therefore, the method that the embodiment of the present invention provides realizes simple.Further, in order to improve the accuracy of the user interest label of acquisition further, the social network user data of the second smart machine belonging to same LAN (Local Area Network) with the first smart machine can also be utilized to upgrade user interest label.Because social networks has friendly man-machine interaction mode usually, be convenient to evaluation, scoring etc. the data of user's input to content; Again owing to being in same LAN, can think that the user of use second smart machine is also the user of use first smart machine, therefore, utilize the social network user Data Update user interest label of the second smart machine, the user interest label interested content topic of indicating user more exactly can be made.
After the method adopting the embodiment of the present invention to provide obtains user interest label, this user interest label can be applied in multiple business.Such as, the user interest label obtained is utilized to carry out intelligent recommendation etc. on the first smart machine.
In the embodiment of the present invention, the implementation of step 100 can be described above, namely from the user behaviors log of the first smart machine, obtains user behavior data.
In the embodiment of the present invention, the implementation of step 110 has multiple.Such as, can define objective behavior, when the user behavior data obtained in step 100 have recorded goal behavior, just think user to the behavior for content interested.Accordingly, when the user behavior data finding acquisition in step 100 have recorded the goal behavior of regulation, generate the user interest label of object time section correspondence, at least indicating user in object time section to goal behavior for content belonging to first content theme interested.Again such as, can define objective behavior, and the duration of define objective behavior, when the user behavior data obtained in step 100 have recorded goal behavior, and the duration that have recorded this goal behavior reaches the duration of regulation, just think user to the behavior for content interested.Accordingly, when the user behavior data finding acquisition in step 100 have recorded the goal behavior of regulation, and the duration of goal behavior reaches the duration of regulation, generate the user interest label of object time section correspondence, at least indicating user in object time section to goal behavior for content belonging to first content theme interested.Certainly, also can have other implementation, the present invention no longer enumerates.
In the embodiment of the present invention, the implementation of step 120 has multiple.Such as, the social network user data of IP address second smart machine identical with the IP address of above-mentioned first smart machine are obtained.That is, whether identical by IP address, determine the second smart machine belonging to same LAN (Local Area Network) with the first smart machine.Again such as, by the mode of LAN (Local Area Network) interscan, obtain the social network user data of the second smart machine of above-mentioned first smart machine place LAN (Local Area Network).That is, by capturing data in the mode of LAN (Local Area Network) interscan.
In the embodiment of the present invention, the implementation of step 130 has multiple.
A kind of implementation is: according to the above-mentioned user obtained the log-on message of social networks and/or this user at the user behavior data of social networks, determine following at least one the information of this user: age, occupation, sex, region; Search above-mentioned first smart machine usage time interval corresponding to the information determined and interested second content topic of this user; If this first smart machine usage time interval has overlapping with above-mentioned object time section, upgrade above-mentioned user interest label, user interest label after renewal also indicates interested second content topic of this user, and the second content topic and first content theme are side by side or be the sub-content topic of first content theme.Concrete, user comprises age, occupation, sex, region etc. at the log-on message of social networks, then can determine age of user, occupation, sex, region according to log-on message.Also can be that user have recorded the age, occupation, sex, region etc. of user at the user behavior data of social networks, determines the age of user, occupation, sex, region accordingly; Such as, user delivers content " we are after 90s " on microblogging, then can determine the roughly age of user accordingly.Can also be, by analyzing the user behavior data of user at social networks, by extracting feature, determine the age of user, occupation, sex, region etc.Further, the pre-configured corresponding relation of above-mentioned information and the first smart machine usage time interval and interested second content topic of user; Such as, 18 ~ 30 years old age bracket, the first corresponding smart machine usage time interval is point ~ 24 point in evening 8; Interested second content topic of user corresponding to finance practitioner is financial and economic news etc.So, by searching the first smart machine usage time interval and interested second content topic of this user that just can determine that above-mentioned information is corresponding.Further, if the first smart machine usage time interval and above-mentioned object time section have overlapping (can specify must be completely overlapping, as long as or regulation overlap just can), upgrade above-mentioned user interest label.Wherein, if the second content topic and first content theme are side by side, the information that can refer to and add interested second content topic of this user of instruction in user interest label is upgraded; Such as, it is interested in " film " this content topic that the user interest label that step 110 generates indicates user, and in this step, the user interest label after renewal adds indicating user to the interested information of " news " this content topic; If the second content topic is the sub-content topic of first content theme, renewal can refer to the information of adding interested second content topic of this user of instruction in user interest label, also can refer to that the information by interested second content topic of this user of instruction replaces the information of the interested first content theme of this user of instruction; Such as, it is interested in " film " this content topic that the user interest label that step 110 generates indicates user, in this step, the user interest label after renewal adds or replaces with indicating user to the interested information of " action class film " this sub-content topic.Further, if the second content topic is the sub-content topic of first content theme, in this step, uninterested 5th content topic of this user corresponding to the above-mentioned information determined can also be searched, after upgrading user interest label, uninterested 5th content topic of the further indicating user of user interest label.
Another kind of implementation is: according to the user behavior data of the above-mentioned user obtained at social networks, analyze interested 3rd content topic of this user and/or uninterested 4th content topic of this user, the 3rd content topic and the 4th content topic are the sub-content topic of first content theme; Upgrade above-mentioned user interest label, the user interest label after renewal also indicates interested 3rd content topic of above-mentioned user and uninterested 4th content topic of above-mentioned user.Its specific implementation with reference to above-mentioned implementation, can repeat no more herein.
Can be cooperatively interacted between the embodiment that it should be pointed out that above steps enforcement.
Below using intelligent television as the first smart machine, and in conjunction with embody rule scene, the technical scheme that the embodiment of the present invention provides is described in detail.
Under a scene, the method that the embodiment of the present invention provides server in the wide area network realizes.And have following configuration:
The configuration of object time section: in this scene, pay close attention to Monday ~ day part of Friday, concrete, determine the following object time section in the middle of a day: 06:00-06:30,06:30-08:00,08:00-10:00,18:00-20:00,20:00-22:00.In the embodiment of the present invention, object time section both can be artificial configuration, also automatically can be configured by certain algorithm.
User uses the goal behavior of intelligent television to configure: in this scene, goal behavior comprises viewing behavior, such as, watch film, viewing news etc., Shopping Behaviors, game behavior.
The configuration of content topic: be provided with multiple content topic (corresponding above-mentioned first content theme) in advance, and the content that each content topic is corresponding.In this scene, content topic comprises film, news, game, shopping program, variety show etc.
The configuration of sub-content topic: be provided with the sub-content topic (corresponding above-mentioned second content topic, the 3rd content topic, the 4th content topic and the 5th content topic) under each content topic in advance, and the content that every sub-content topic is corresponding.In this scene, for this content topic of film, the sub-content topic under this content topic can comprise: action class film, science fiction class film, Culture Revolution class film, war of resistance class film etc.
The configuration of target social networks: pre-configured target social networks, the then user data of follow-up acquisition target social networks.In this scene, the social user of target comprises microblogging, micro-letter, bean cotyledon net, TIME dotCom etc.
The configuration of target social network information: pre-configured target social network information, follow-up from social network user extracting data target social network information.In this scene, configuration target social information comprises: the information relating to content scores, relates to the information of resource content evaluation, log-on message of user etc.
Obtain the period assignment of user interest label: usually user on Monday ~ Friday uses the rule of intelligent television substantially fixing, therefore, in this scene, configuring this cycle is a week.It should be pointed out that in actual use, also can configure other cycles, the present invention is not construed as limiting this.
Based on above-mentioned configuration, server each week obtains user interest label for each object time section above-mentioned respectively.The implementation in one of them cycle as shown in Figure 2, specifically comprises following operation:
Step 200, server obtain the user behavior data of the intelligent television in each object time section above-mentioned.
In the present embodiment, server obtain last Monday ~ every day Friday in the middle of, the user behavior data of the intelligent television in each object time section above-mentioned.Such as, obtain last Monday ~ every day Friday this period of 06:00 ~ 06:30, the user behavior data of intelligent television.
In the present embodiment, user behavior data comprises behavior time of origin, user behavior, behavior for content.Wherein, to play the program that television channel provides, behavior for content can but be not limited only to comprise: the station symbol (such as CCTV2-channel for finance and economics) of television channel, (what such as play is finance and economic news to the classification of the program play, the classification of this program can be finance and economics, news etc.), if what play is film, content can also comprise the director, performer, films types etc. of film.
In the present embodiment, owing to being configured with goal behavior, therefore, server only obtains user behavior data corresponding to goal behavior.Namely the user behavior data that viewing behavior, Shopping Behaviors, game behavior are corresponding is only obtained.Concrete, can according to the play content identification goal behavior of the intelligent television recorded in user behavior data.Such as, a certain bar user behavior data have recorded the program having play CCTV-2 (channel for finance and economics) at 06:10 ~ 06:30, then identify that the user behavior that this user behavior data is corresponding is viewing behavior.
Step 210, server generate the user interest label of each object time section correspondence above-mentioned according to the above-mentioned user behavior data obtained, and this user interest label indicates the interested content topic of user.
Wherein, in the present embodiment, the particular content of user interest label is as follows: " liking film ", " liking listening song ", " liking game ", " liking buying clothes " etc.
In the present embodiment, if certain user behavior duration of user behavior data record reach the schedule time, then think user to the behavior for content interested.Accordingly, in this step, server, according to the record of user behavior data, extracts user's interested content in each object time section above-mentioned respectively; According to the content topic of configuration, determine the content topic belonging to these contents, thus determine user's interested content topic in each object time section above-mentioned, generate the user interest label of each object time section correspondence.Such as, by analyzing, following user interest label is determined:
06:00-06:30 likes news category
06:30-08:00 likes song class
08:00-10:00 likes variety show
18:00-20:00 likes animated type program
20:00-22:00 likes film, TV play class program
In the present embodiment, by the IP address of intelligent television, object time section, and the corresponding relation between the user interest label three got saves (because TV is towards multiple people, be generally towards one family, that is, TV is the electronic equipment that a lot of people shares, so a fixing user name generally can not be used on TV to log in, just this element of user name is not joined in corresponding relation here).
Step 220, server obtain the social network user data with the smart machine of above-mentioned intelligent television identical ip addresses, and these social network user data comprise above-mentioned user at the log-on message of social networks and the above-mentioned user user behavior data at social networks.
Concrete, by report with the smart machine of above-mentioned intelligent television identical ip addresses, the social network user data of target social networks.
If reach data with target social networks to use agreement, then, in this step, server directly obtains the back-end data of target social networks.If do not reach data with target social networks to use agreement, then by the public data of browsing objective social networks, obtain desired data.
Step 230, server are according to the user interest label of each object time section correspondence of above-mentioned social network user Data Update obtained.
In this step, server, from the social network user extracting data target social network information obtained, determines the interested sub-content topic of user and uninterested sub-content topic according to the target social network information extracted.
Wherein, for the information such as user's sex, occupation, educational background, region, age got, can directly add in user interest label as the content of user interest label.Also can analyze the interested sub-content topic of user and uninterested sub-content topic further according to these information, the information of this little content topic of instruction is added in user interest label.
Such as, have recorded " viewed from today " the fast and the furious 7 "; joyride scene is good enjoyable " from social network user data of meagre upper acquisition, " do not understand the thought of people in " the ordinary world "; see to go down ", determine that user is interested in this sub-content topic of action class film, and this sub-content topic of Culture Revolution class film is lost interest in.Then upgrade user interest label, indicating user likes action class film, does not like Culture Revolution class film.Because the interest tags that the above-mentioned 20:00 of determining ~ 22:00 is corresponding likes film, TV play class program, after this step upgrades, the interest tags that 20:00 ~ 22:00 is corresponding likes action class film, TV play class program, do not like Culture Revolution class film, TV play class program.Follow-up based on user interest label recommendations TV play time, will eliminate with the reaction Cultural Revolution of " the ordinary world " class after the TV play of the subject matter such as class struggle.
It should be noted that, for some object time section, if the sub-content topic under the interested content topic of this object time section cannot be analyzed by the social network user data obtained, then for these object time sections, only based on the user behavior data determination user interest label of intelligent television.Such as, the daytime of the week may be all that old man or child are in, and captures and waits social network user behavioral data less than the micro-letter of real-time microblogging, can only according to the user interest label determined in above-mentioned steps 210 to user's recommended program.
Utilize set about mechanical, electrical brain, intelligent television of home network environment to share the feature of a network egress, by under same time period, same IP condition, social network user label and intelligent television user tag combine, and obtain the user interest label of relative abundance.
Under another scene, the equipment of the method that the embodiment of the present invention provides in LAN (Local Area Network) realizes, and has above-mentioned configuration.
Wherein, the equipment in above-mentioned LAN (Local Area Network) can be intelligent television, also can be smart mobile phone, panel computer etc. smart machine, also can be LAN gateway etc.
Its implementation can with reference to the description of above-described embodiment, but the means obtaining social network user data can be different.Concrete, can by obtaining social network user data in the mode of LAN (Local Area Network) interscan.
Based on the inventive concept same with method, the embodiment of the present invention also provides a kind of device obtaining user interest label, as shown in Figure 3, comprising:
First data acquisition module 301, for obtaining the user behavior data of the first smart machine in object time section;
Interest tags generation module 302, for the user behavior data according to described first smart machine obtained, generate the user interest label of described object time section correspondence, described user interest label at least indicates the interested first content theme of user;
Second data acquisition module 303, for obtaining the social network user data of the second smart machine of described first smart machine place LAN (Local Area Network), described social network user data comprise described user at the log-on message of social networks and/or the described user user behavior data at social networks;
Interest tags update module 304, for user interest label according to the described social network user Data Update obtained.
Inventor is realizing finding in process of the present invention, and user behavior data can reflect the interested content topic of user.Such as, user watches certain this behavior of film by smart machine, may reflect user interested in this content topic of film.When the interested content topic of user cannot be determined according to the fragmented information of user's input, adopt the device that the embodiment of the present invention provides, by obtaining the user behavior data of first smart machine of behavior time of origin in object time section, just can generate the user interest label of this object time section correspondence according to the user behavior data obtained, indicating user is at the interested first content theme of this object time section.Because the user behavior data of the first smart machine is preserved in the mode of user behaviors log usually, therefore, obtain the user behavior data that user behaviors log can obtain the first smart machine, therefore, the device that the embodiment of the present invention provides realizes simple.Further, in order to improve the accuracy of the user interest label of acquisition further, the social network user data of the second smart machine belonging to same LAN (Local Area Network) with the first smart machine can also be utilized to upgrade user interest label.Because social networks has friendly man-machine interaction mode usually, be convenient to evaluation, scoring etc. the data of user's input to content; Again owing to being in same LAN, can think that the user of use second smart machine is also the user of use first smart machine, therefore, utilize the social network user Data Update user interest label of the second smart machine, the user interest label interested content topic of indicating user more exactly can be made.
Optionally, described second data acquisition module is used for:
Obtain the social network user data of IP address second smart machine identical with the IP address of described first smart machine.
Optionally, described second data acquisition module is used for:
By the mode of LAN (Local Area Network) interscan, obtain the social network user data of the second smart machine of described first smart machine place LAN (Local Area Network).
Optionally, described interest tags update module is used for:
According to the described user obtained the log-on message of social networks and/or described user at the user behavior data of social networks, determine following at least one the information of described user: age, occupation, sex, region;
Search described first smart machine usage time interval corresponding to the information determined and interested second content topic of described user;
If described first smart machine usage time interval has overlapping with described object time section, upgrade described user interest label, user interest label after renewal also indicates interested second content topic of described user, and described second content topic and described first content theme are side by side or be the sub-content topic of described first content theme.
Optionally, described interest tags update module is used for:
According to the user behavior data of the described user obtained at social networks, analyze interested 3rd content topic of described user and/or uninterested 4th content topic of described user, described 3rd content topic and the 4th content topic are the sub-content topic of described first content theme;
Upgrade described user interest label, the user interest label after renewal also indicates interested 3rd content topic of described user and uninterested 4th content topic of described user.
Those skilled in the art should understand, embodiments of the invention can be provided as method, system or computer program.Therefore, the present invention can adopt the form of complete hardware embodiment, completely software implementation or the embodiment in conjunction with software and hardware aspect.And the present invention can adopt in one or more form wherein including the upper computer program implemented of computer-usable storage medium (including but not limited to magnetic disk memory, CD-ROM, optical memory etc.) of computer usable program code.
The present invention describes with reference to according to the process flow diagram of the method for the embodiment of the present invention, equipment (system) and computer program and/or block scheme.Should understand can by the combination of the flow process in each flow process in computer program instructions realization flow figure and/or block scheme and/or square frame and process flow diagram and/or block scheme and/or square frame.These computer program instructions can being provided to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device to produce a machine, making the instruction performed by the processor of computing machine or other programmable data processing device produce device for realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.
These computer program instructions also can be stored in can in the computer-readable memory that works in a specific way of vectoring computer or other programmable data processing device, the instruction making to be stored in this computer-readable memory produces the manufacture comprising command device, and this command device realizes the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.
These computer program instructions also can be loaded in computing machine or other programmable data processing device, make on computing machine or other programmable devices, to perform sequence of operations step to produce computer implemented process, thus the instruction performed on computing machine or other programmable devices is provided for the step realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.
Although describe the preferred embodiments of the present invention, those skilled in the art once obtain the basic creative concept of cicada, then can make other change and amendment to these embodiments.So claims are intended to be interpreted as comprising preferred embodiment and falling into all changes and the amendment of the scope of the invention.
Obviously, those skilled in the art can carry out various change and modification to the present invention and not depart from the spirit and scope of the present invention.Like this, if these amendments of the present invention and modification belong within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention is also intended to comprise these change and modification.