CN109559140A - It expands the other method, apparatus of user interest degree group, calculate equipment and storage medium - Google Patents
It expands the other method, apparatus of user interest degree group, calculate equipment and storage medium Download PDFInfo
- Publication number
- CN109559140A CN109559140A CN201710880869.3A CN201710880869A CN109559140A CN 109559140 A CN109559140 A CN 109559140A CN 201710880869 A CN201710880869 A CN 201710880869A CN 109559140 A CN109559140 A CN 109559140A
- Authority
- CN
- China
- Prior art keywords
- user
- interest
- degree group
- degree
- natural quality
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of other method, apparatus of expansion user interest degree group, calculate equipment and storage medium.User corresponding natural quality is classified to according to the natural quality of user to be grouped, and the other ranking of interest-degree group of interior big data user, the interest-degree group for recommending in the top and user to be not yet marked to user or the other information of interest-degree group for belonging to recommendation are grouped based on natural quality.As a result, while making up the deficiency precisely recommended, it is ensured that the information of recommendation can better meet the potential demand of user, dabble face so as to expand the information of user, and promote the satisfaction of user.
Description
Technical field
The present invention relates to information technology fields, more particularly to one kind for expanding user interest degree group method for distinguishing, dress
It sets, calculate equipment and storage medium.
Background technique
Precisely recommend mainly by obtaining the browsing behavior data of user, and analyzed based on special algorithm, so as to
Recommend the relevant information for meeting its browsing behavior to user.For example, can be user by the browsing behavior data of acquisition user
User's portrait is constructed, is drawn a portrait according to the user of user and is precisely recommended to user, for example information class application " today's tops " is just
It is the Typical Representative precisely recommended.
The navigation interest of user can relatively accurately be met based on the mode information recommended to the user precisely recommended, but
It is that the type of the information that can also obtain to user to a certain extent is limited.For example, commonly using today's tops
User is it can be found that in use for some time, the information that can be contacted often is confined to limited several classification.
Therefore, in order to make up the deficiency of the accurate way of recommendation, it also will use " recommending to explore " when to user's recommendation information
Mode to user's recommendation information.Explore the new letter recommended to refer to and recommended based on certain rule to user except precisely recommending
Breath dabbles face, viscosity of the increase user to it to widen information of the user in specific business or application.
Currently, recommending to explore mainly by the way of following to user's recommendation information.
1. it randomly selects and organizes other information except the interest-degree group that individual consumer is possessed, a part as recommendation.
2. establishing the mapping between different interest-degree groups or incidence relation based on specific strategy, choose and individual
Relevant group of other information except the interest-degree group that user is possessed, a part as recommendation.
3. choosing the different other high-quality resources of interest-degree group, user is indistinguishably recommended.
Although being pushed away as it can be seen that existing recommend exploring mode that can recommend the new information except precisely recommending to user
Recommend process is not close in conjunction with user, do not fully take into account the potential demand of user so that information recommended to the user by with
A possibility that family receives is lower, can not accurately expand the potential of user and dabble face, or even can bring disagreeableness browsing to user
Experience.
Summary of the invention
The main purpose of the present invention is to provide one kind to set for expanding the other method, apparatus of user interest degree group, calculating
Standby and storage medium, by the interest-degree group for recommending the other users identical or essentially identical with the natural quality of user to user
The other or corresponding information of interest-degree group, can be while making up the deficiency precisely recommended, it is ensured that the information energy of recommendation
The potential demand of user is enough better meet, dabbles face so as to expand the information of user, and promote the satisfaction of user.
According to an aspect of the invention, there is provided a kind of expansion user interest degree group method for distinguishing, comprising: according to user
Natural quality user be classified to corresponding natural quality be grouped, natural quality includes at least two belonging to for the natural of grouping
Property dimension, and user be classified to natural quality dimension with identical value natural quality be grouped；Based on natural quality point
The other ranking of interest-degree group of big data user, the interest-degree group for recommending in the top and user to be not yet marked to user in group
The other information of interest-degree group that is other or belonging to recommendation.
This understanding of same or similar hobby would generally be had by being grouped identical user based on natural quality, in group
Interest-degree group in the top can reflect to a certain extent with the general of group user in the interest-degree group of big data user
All over interest.Therefore, the interest-degree group or the other information recommendation of interest-degree group by the way that in the top and user to be not yet marked
To user, the probability that recommended interest-degree group or information are easily accepted by a user can be further improved, so as to expand
While the information of user dabbles face, the satisfaction of user is further promoted.
Preferably, natural quality, which is grouped the other ranking of interest-degree group of interior big data user, is: belonging to natural quality grouping
All users the other ranking of interest-degree group；Or belong to the big data use of natural quality grouping under the classification of other attribute dimensions
The other ranking of interest-degree group at family.
Thus, it is possible to which flexible choice is used to divide the range and dimension with group user according to application demand.Such as specific
Application or business dimension, if under browser novel read this dimension of user under, the user in group is screened, with sieve
Select the user for meeting actual demand.And it is opened up to the interest-degree group in the user's progress novel reading range for meeting demand is other
Exhibition.
Preferably, other attribute dimensions and in the natural quality dimension of grouping be at least partly based on setting can be mutual
The dimension changed.Thus, it is possible to further promote the flexibility for dividing range and dimension with group user, preferably to fit
Needs for various practical businesses.
Preferably, the other ranking of interest-degree group that natural quality is grouped interior big data user is asked according at least to one of following
It takes: the sum of the interest-degree group of each user in being grouped；The sum of the interest-degree group weighting of each user in being grouped.As a result, may be used
Neatly to determine the other ranking machine system of interest-degree group in group based on conditions such as source-information type, calculated loads.
Preferably, the interest-degree group of recommending in the top and user to be not yet marked to user or the interest for belonging to recommendation
The other information of degree group may include: according to the multiple other weighted sums of interest-degree group for recommending in the top and user to be not yet marked
Value, recommend the other information of interest-degree group for belonging to recommendation of corresponding percentage.Thus, it is possible to accurate when recommending multipacket message
The particular number to be recommended to each group.
It preferably, may include at least one following: gender for the natural quality dimension of grouping；Age bracket；Region；
Occupation；Know-how；And curious degree.As a result, by the reasonable selection and setting to user's essential attribute, can more subject to
Reasonable user grouping is really neatly realized, to provide more accurate recommendation.
Preferably, this method can also include: to obtain user to the interest-degree group of recommendation or belong to the interest-degree of recommendation
The click behavior of the other information of group；When the behavior of click is higher than threshold condition, user's sheet is converted by the interest-degree group of recommendation
The interest-degree group that body includes.Thus, it is possible to be realized by conversion to the other real extension of the interest-degree group of user.
Preferably, this method can also include: the value of the curious degree dimension based on conversion adjustment user.
As a result, after the value of the curious degree dimension to user is adjusted, it can classify again to user, and
Recommend interest-degree group or the other information of interest-degree group again accordingly, so as to form benign cycle, so that recommending to user
Information can preferably meet the needs of users.Also, when the value of the curious degree dimension in adjustment user is too small, may be used also
To stop recommending the information except precisely recommending to user.
Preferably, in the natural quality dimension of grouping at least one of can at predetermined intervals periodically more
Newly.Thus, it is possible to the natural quality dimension for having time variability in the natural quality dimension for grouping according to scheduled
Time interval regularly updates, so that updated natural quality dimension is able to reflect the variation of the natural quality of user.
According to another aspect of the present invention, a kind of other device of expansion user interest degree group is additionally provided, comprising: classification
User is classified to corresponding natural quality for the natural quality according to user and is grouped by unit, and natural quality includes at least two
A natural quality dimension for grouping, and user is to be classified to the natural quality dimension natural quality with identical value point
Group；Recommendation unit recommends ranking to user for the other ranking of interest-degree group of big data user in being grouped based on natural quality
Interest-degree group that forward and user is not yet marked or the other information of interest-degree group for belonging to recommendation.
Preferably, natural quality, which is grouped the other ranking of interest-degree group of interior big data user, is: belonging to natural quality grouping
All users the other ranking of interest-degree group；Or belong to the big data use of natural quality grouping under the classification of other attribute dimensions
The other ranking of interest-degree group at family.
Preferably, other attribute dimensions and in the natural quality dimension of grouping be at least partly based on setting can be mutual
The dimension changed.
Preferably, the other ranking of interest-degree group that natural quality is grouped interior big data user is asked according at least to one of following
It takes: the sum of the interest-degree group of each user in being grouped；The sum of the interest-degree group weighting of each user in being grouped.
Preferably, recommendation unit is according to the multiple other weightings of interest-degree group for recommending in the top and user to be not yet marked
The value of sum recommends the other information of interest-degree group for belonging to recommendation of corresponding percentage.
It preferably, include at least one following: gender for the natural quality dimension of grouping；Age bracket；Region；Duty
Industry；Curious degree.
Preferably, which can also include: acquiring unit, for obtaining user to the interest-degree group of recommendation or belonging to
The click behavior of the other information of interest-degree group of recommendation；Conversion unit, for will recommend when the behavior of click is higher than threshold condition
Interest-degree group be converted into the interest-degree group that user itself includes.
Preferably, which can also include: adjustment unit, for the curious degree dimension based on conversion adjustment user
Value.
Preferably, it is regularly updated at predetermined intervals at least one in the natural quality dimension of grouping.
According to a further aspect of the invention, a kind of calculating equipment is additionally provided, comprising: processor；And memory,
On be stored with executable code, when executable code is executed by processor, processor is made to execute the method addressed above.
According to a further aspect of the invention, a kind of non-transitory machinable medium is additionally provided, is stored thereon
There is executable code, when executable code is executed by the processor of electronic equipment, processor is made to execute the method addressed above.
The other method, apparatus of expansion user interest degree group of the invention calculates equipment and storage, by user recommend with
The interest-degree group or the corresponding information of interest-degree group of the identical or essentially identical other users of the natural quality of user, can be with
While making up the deficiency precisely recommended, guarantee that the information recommended can more precisely meet the potential demand of user, from
And information can be expanded for user itself and dabble face and promote user's stickiness.
Detailed description of the invention
Disclosure illustrative embodiments are described in more detail in conjunction with the accompanying drawings, the disclosure above-mentioned and its
Its purpose, feature and advantage will be apparent, wherein in disclosure illustrative embodiments, identical reference label
Typically represent same parts.
Fig. 1 is to show the schematic flow according to an embodiment of the invention for expanding user interest degree group method for distinguishing
Figure.
Fig. 2 be show it is according to an embodiment of the invention expand the other device of user interest degree group structure it is schematic
Block diagram.
Fig. 3 is to show the schematic block diagram of the structure according to an embodiment of the invention for calculating equipment.
Specific embodiment
The preferred embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although showing the disclosure in attached drawing
Preferred embodiment, however, it is to be appreciated that may be realized in various forms the disclosure without the embodiment party that should be illustrated here
Formula is limited.On the contrary, these embodiments are provided so that this disclosure will be more thorough and complete, and can be by the disclosure
Range is completely communicated to those skilled in the art.
Before describing the present invention, concept of the present invention and invention mechanism are described briefly first.
User portrait: be the virtual representations of real user, be built upon a series of truthful datas (Marketing data,
Usability data) on user model.Product or service side can be directed to targeted user population, by using effective
Mark system and implement respective algorithms, sketch the contours of the synthesis prototype of real user in data plane.The user's portrait constructed
It can apply in itself or other business scenarios.
Mark system: in particular to the scheme classified for the information under particular range.Classification and label are marks
Common means in system.For example the classification of novel " hiding day " is " ancient customs ", label is " warm blood " and " pursuing a goal with determination ".These are to this
The classification annotation information of novel belongs to mark system.It is also a kind of machine for distinguishing between information individual in information
System.Meanwhile mark system is the system basis recommended.
Interest-degree group system: corresponding with mark system.For information, classification or grouping are dependent on mark
System.And for a user, classification or grouping are then interest-degree group systems.In general, system and interest are marked
Spending group system is mutual corresponding in realization, that is to say, that directly the mark system of information can be applied as user's
Interest-degree group system.In the present invention, interest-degree group system can be equal to mark system.
Interest-degree group: classifying to user or be grouped based on interest-degree group system, for each user,
It can have one or more interest-degree groups.Wherein, the interest-degree group of user can be under same business or application
Interest-degree group, the interest-degree group being also possible under different business or application.For example, the interest of the user A in novel business
Degree group can be swordsman, suspense, science fiction, and swordsman, suspense, science fiction these three interest-degree groups can also have it is certain
Weight such as can be 20%, 30%, 50% respectively.
As described in the background section, it explores and recommends as the supplement precisely recommended, can expand user dabbles face, with
The limitation precisely recommended is made up to a certain extent.But the existing exploration way of recommendation is recommending precisely to recommend it to user
When outer new information, it is generally difficult to the potential demand in view of user, so that exploring what the mode recommended was recommended based on existing
A possibility that information is easily accepted by a user is lower, so that the information that will not only expand user dabbles face, can also cause not to user
Good viewing experience.
For this problem, inventor has found after extensive studies it, in a certain or certain several natural quality dimension
Same or similar multiple users, the interest between them are usually to convert each other.For example, A and B are high school students, and A and
B is boy student.A likes seeing that class novel of pursuing a goal with determination, B are liked seeing swordsman's class TV play usually usually.It is expected that if recommending A to B
The class novel of pursuing a goal with determination for liking seeing, B can also attempt viewing and pursue a goal with determination class novel, or even class novel of pursuing a goal with determination is seen on liking, equally, if
B is recommended to like the swordsman's class TV play seen to A, A is also possible to like to see swordsman's class TV play.For another example A is computer
The programmer for having entered into work of major, B are the students for being just admitted to computer major, and A likes browser design usually
Website and professional website, it is contemplated that if received before recommending to B not in contact with the programming website crossed or professional website, B
And the probability for liking such upper recommendation information is larger.
In view of this, the present invention proposes, can according to the natural quality of user, by natural quality some or certain it is several
Same or similar user is divided into same natural quality grouping in natural quality dimension.It, can be with for a certain user in group
Recommend the interest-degree group or the corresponding recommendation information of interest-degree group that it does not have but other users have in group to it.
New interest-degree group recommended to the user or the new corresponding information of interest-degree group can better meet user's as a result,
While making up precisely recommendation bring deficiency, user can be improved to dabble face in the information for expanding user in potential demand
To the acceptance probability of the information of recommendation, and then promote the satisfaction of user.Also, it can also realize the interest-degree to user itself
Thus the other expansion of group promotes the amount of reading of user, and then promotes usage rate of the user.
Further, when dividing natural quality grouping according to natural quality dimension, it usually needs to large-scale consumer number
According to being grouped, so that the user in the grouping of each natural quality is big data user.In this way, recommending in user a certain into group
When new interest-degree group or the corresponding information of new interest-degree group, can according in organizing other big datas user it is all emerging
Interest-degree group in the top is recommended user by the interesting other ranking of degree group.The interest-degree group of big data user in organizing as a result,
Interest-degree group in the top can reflect the use with the corresponding natural quality dimension of the grouping to a certain extent in not
The common interest at family can be considered as the standard configuration interest of the user with the natural quality dimension.Therefore, big data is used in organizing
Interest-degree group in the top or the corresponding information recommendation of interest-degree group in family can be further improved and pushed away to user
The interest-degree group or the corresponding information of interest-degree group recommended, the probability being easily accepted by a user, so as in the letter for expanding user
While breath dabbles face, the satisfaction of user is further promoted.
In the present invention, natural quality can be the attribute for referring to reflection user's oneself state information.It can be first
The attribute for the reflection user's body states such as whether gender, age, height, weight, body sound, next can also be the think of of itself
Dimension mode attribute, such as to the curious degree of unknown things, to the reception preference and logical level of architecture and fragmentation information
Deng.
Above-mentioned mode of thinking attribute can not directly acquire, but can usually be retouched by the behavior and other attributes of user
It states.User is directly related to a possibility that user receives recommendation information to the curious degree of unknown things.In one embodiment,
Face degree can be dabbled by information to be described from curious degree of the side to user, and mainly anti-using the dimension
Reflect user receive new recommendation information resistance wish it is strong with it is weak.In another embodiment, user's history behavior can be passed through
New recommendation information is received to assess user to the reception degree of the recommendation information of other classifications except concern classification in data
Resist wish it is strong with it is weak.It similarly, can also be from the occupation of user, region, education level, classification of receiving an education, assets, wedding
The labels such as relation by marriage situation are started with come the attribute for the mode of thinking level for reflecting user.For example, can be according to Regional Distribution to belonging to west
The user in portion province recommends northwest folk rhyme；It can also be according to the modernization level of locating region (for example, go up north wide deep, a line, two
Line city etc.) recommend different commodity；Current hot marketing books or online course are recommended to sales force；It avoids to height
Net value crowd (for example, can be assessed from educational background, profession, condition of assets etc.) or high support system recommend " title party " class letter
Breath etc..
Natural quality dimension can divide different groupings according to specific natural quality, and the value of natural quality dimension can
To be set according to actual conditions.Such as this natural quality of age, juvenile (under-18s), youth can be divided
Multiple natural quality dimensions such as (40 years old 18 years old or more or less), middle aged (60 years old 40 years old or more or less), old (60 years old or more).Often
The specific value of a nature dimension can specifically be set according to concrete application scene or demand.
Since the natural quality of user may vary over, such as the attributes such as age, region, height, weight
It can or be possible to vary over.It is therefore preferred that user is classified to pair according to the natural quality of user
After the natural quality grouping answered, user can also be grouped again at predetermined intervals.For example, according to user
Gender, age, region, information is when dabbling the natural qualities such as face degree and being grouped, the gender of user and region are usually phase
To fixed, the age can change over time, and information dabbles face degree and then may be used after a relatively long period
It can will appear and wave variation, therefore be referred to the user of natural quality dimension having the same according to this four natural qualities
After the grouping of same natural quality, scheduled time interval can be set, it is specific be spaced length can according to business self-characteristic into
Row adjustment, such as can be 30 days, after meeting time interval, re-starts grouping to user, so that grouping can be anti-in time
Reflect the variation of the certain natural qualities of user.
So far the basic principle of concept of the present invention and invention is briefly explained, below with reference to specific implementation
Just specific implementation process of the invention is described in further details example.
Fig. 1 is to show the schematic flow according to an embodiment of the invention for expanding user interest degree group method for distinguishing
Figure.
User is classified to by corresponding natural quality according to the natural quality of user and is grouped in step S110 referring to Fig. 1.
Definition about natural quality may refer to above description, and details are not described herein again.It, can be with when executing step S110
User corresponding natural quality is classified to according to one or more natural qualities of user to be grouped.Wherein, each natural quality
It may include at least two natural quality dimensions for grouping.
In one embodiment of the invention, can be belonged to naturally according to the gender of user, age, region, curious degree etc.
Property is grouped.As an example, gender can be divided into male, female, unknown three dimensions.Age can be divided into four according to age bracket
Dimension, respectively juvenile (under-18s), young (40 years old 18 years old or more or less), middle aged (60 years old 40 years old or more or less), old age
(60 years old or more).Region can be divided according to country/city up-to-dateness height, such as region is State-level
When, can be divided into three developed country, developing country and less developed country dimensions can be divided into a line when region is domestic
City, tier 2 cities, three line cities, four line cities, five, five line city dimension.In addition it also can according to need according to specifically
Domain distribution is divided (for example, middle part, western part, southeastern coast, overseas etc.).
Know-how is mainly used for inferring that user is ready to receive the type of new information.Know-how is higher, then is more possible to
Receive more professional and system information recommendation, correspondingly can more dislike the recommendation of such as " title party " category information.Knowledge water
It is flat such as can be assessed according to the education level of user, profession, reading habit before.
Curious degree is mainly used for characterizing the wish degree that user receives new information, and curious degree is high, then user receives new
The wish of information is stronger, and curious degree is low, then user receive new information wish it is weaker.In the present invention, information can be used
Dabble the curious degree that face degree represents user, narrow, general, extensive three dimensions can be divided at this time.Wherein, information relates to
The face degree of hunting can be defined according to historical behavior data of the user in concrete application or business.For example, can be to user
Historical behavior data analyzed, when user browses the information compared with polymorphic type, it is believed that the information of user relates to
The face degree of hunting be it is extensive, when information of the user to less type browses, it is believed that the information of user dabbles face degree
Be it is narrow, in being between the two general.Determine that the information of user dabbles face alternatively, it is also possible to draw a portrait according to the user of user
Degree, such as can by user representation data of the user in a certain service/application analyzed to obtain user the business/
Information in dabbles face degree.
When by user grouping, user can be classified to natural quality dimension and be grouped with the natural quality of identical value.
It wherein, can be by the user point under the natural quality with identical natural quality dimension when being classified according to a natural quality
Class a to natural quality is grouped.When being classified according to multiple natural qualities, preferably will can have under multiple natural qualities
There is the user of identical natural quality dimension to be classified to a natural quality grouping.It can so guarantee that a user is only divided into
One natural quality is grouped, the potential relevance in raising group between user.
As described above, since the natural quality of user may change over time, correspondingly, user
Affiliated natural quality dimension is also possible to change.It therefore, can be to the time in the natural quality dimension for grouping
The stronger natural quality dimension of variability regularly updates at predetermined intervals.For example, it may be age bracket, region etc. are tieed up
Degree.Update described herein refers to, periodically carries out again at predetermined intervals to the user under the natural quality dimension
Classification to obtain natural quality dimension new belonging to user, and then obtains new natural quality grouping.Wherein, in order to guarantee
The accuracy regularly updated, it is also necessary to periodically obtain the new natural quality information of user, it is all can periodically to obtain user herein
Natural quality information, can also only obtain and change over time more natural quality information such as age, region.
In step S120, it is grouped the other ranking of interest-degree group of interior big data user based on natural quality, recommends to user
Interest-degree group that in the top and user is not yet marked or the other information of interest-degree group for belonging to recommendation.
It is divided into natural quality dimension having the same between the user in the grouping of same natural quality, according to retouching above
It states, same or similar multiple users on a certain or certain several natural quality, the interest between them is usually to turn each other
Change.Therefore, for a certain user in group, interest-degree belonging to other users in the group being not yet marked can be recommended to it
Group, or directly recommend the corresponding information of interest-degree group belonging to other users in the group being not yet marked to user.
It should be noted that the recommendation for focusing on interest-degree group or the corresponding information of interest-degree group of the invention.
Other obtain of interest-degree group about user is not emphasis of the invention, can be obtained using existing mode, such as may be used
To be analyzed to obtain the interest-degree group of user by user to user portrait, can also by user in specific industry
Behavior record information in business/application is analyzed, and the user interest degree group of user is obtained.
In order to improve the interest-degree group or the probability that is easily accepted by a user of the corresponding information of interest-degree group of recommendation, Ke Yigen
According to the interest-degree group ranking of big data user in organizing, recommend new interest-degree group or new interest-degree to user according to ranking
The corresponding information of group.
Specifically, can be according to the sum of the interest-degree group of each user in being grouped, or it is grouped the interest of interior each user
The sum of group weighting or the combination of both modes are spent, the other ranking of interest-degree group is obtained.That is, can be by tool
There is the number of the other user of the interest-degree group to be summed to obtain each other score value of interest-degree group in group, it can also be by tool
There is the other weighting of interest-degree group in the other user of the interest-degree group to be summed to obtain each other score value of interest-degree group in group,
In addition the combination of above two mode can also be taken to ask.Then interest-degree group is ranked up according to score value size
Obtain the other ranking of interest-degree group for being grouped interior big data user.
Thus, it is possible to according to the multiple other weighted sums of interest-degree group for recommending in the top and user to be not yet marked
Value recommends the other information of interest-degree group for belonging to recommendation of corresponding percentage.It can be pressed according to the other weighted value of interest-degree group
The information that ratio chooses corresponding interest-degree mark is recommended.For example, the 3 interest-degree group L1/L2/L3 recommended to user UU
Percentage weight value be respectively 20%/30%/50%, it is assumed that need to user recommend 30 information, then can be proportionally
6 are chosen from L1 classification, L2 classification chooses 9, and L3 classification chooses 15.May be used also in addition, the information of user dabbles face degree
Be used to determine every time to user's recommendation information type and item number number and/or information recommendation frequency.For example, for information
Dabble the lower user grouping of face degree, even if being recommended, it should also which but information more identical than other dimension values dabbles face journey
Spending the information type that higher user grouping is recommended every time will lack, and item number is less and/or recommended frequency is lower.
In the present invention, the other ranking of interest-degree group that natural quality is grouped interior big data user, which can be, belongs to natural quality
The other ranking of interest-degree group of all users of grouping is also possible to belong to natural quality grouping in the case where other attribute dimensions are classified
Big data user the other ranking of interest-degree group.
For example, for the nature for using gender, age, region, curious this four natural qualities of degree to be grouped
Attribute grouping can be browser (such as search dog, QQ browser) user, the novel in browser client reads user, browser
Belong to the interest-degree group of the big data user of natural quality grouping under other attribute dimensions classification such as old Beijing user in user
Other ranking.Wherein, region " old Beijing user " described herein is considered as other attribute dimensions, not for the nature category of grouping
Property " region " in the included natural quality dimension for grouping.Thus, it is possible to which (such as product/business needs according to actual needs
Ask), targetedly using other attribute dimensions (the specific dimension under such as product/business) to the big number in natural quality grouping
It is screened according to user, to obtain the big data user for meeting other attribute dimensions classifications, and the big data obtained based on screening
The interest-degree group of user recommends new interest-degree group or the corresponding information of interest-degree group to user.
In one embodiment of the invention, other attribute dimensions and for at least portion in the natural quality dimension of grouping
Dividing can be based on the interchangeable dimension of setting.Such as described in upper example, other attribute dimensions, which can be, to be different from for grouping
Natural quality dimension, other attribute dimensions can be interchangeable with one or more in the natural quality dimension for grouping,
It can be grouped other attribute dimensions as natural quality dimension, and using natural quality dimension as other attribute dimensions
It is screened.
So far it combines Fig. 1 to be just grouped according to the natural quality of user, and recommends interest-degree group or interest to user
The process of the corresponding information of degree group elaborates.Furthermore it is also possible to according to the opposite interest-degree group that it is recommended of user
Or the click behavior of the corresponding information of interest-degree group, determine whether recommended information can characterize the true hobby of user,
In the case where it can characterize the true hobby of user, it is translated into the interest-degree group of user itself, thus realization pair
The other expansion of interest-degree group of user.
In specific implementation, can after the interest-degree group or the corresponding information of interest-degree group to user's recommendation,
Click behavior of the user to the interest-degree group of recommendation or the other information of interest-degree group for belonging to recommendation is obtained, and in the behavior of click
When higher than threshold condition, it converts the interest-degree group of recommendation to the interest-degree group of user itself.
As an alternative embodiment of the present invention, it can also be directed to the interest-degree group recommended according to user or belong to and push away
The click behavior for the other information of interest-degree group recommended adjusts the value of the curious degree dimension of user.Specifically, in user
When click probability is too low, the value of the curious degree dimension of user can be turned down, when user's click probability is higher, can be incited somebody to action
The value of the curious degree dimension of user is turned up.And when the value of the curious degree dimension in user is too low, it can stop making
Information except precisely being recommended with from the mode that recommendation of the invention is explored to user's recommendation.It is alternatively possible to be pushed away according to user
Whether the interest-degree group recommended is converted into the interest-degree group of user itself, adjusts the value of the curious degree dimension of user.
So far, expansion user interest degree group method for distinguishing of the invention is described in detail in conjunction with Fig. 1.Expansion of the invention is used
Family interest-degree group method for distinguishing is also implemented as a kind of other device of expansion user interest degree group.Fig. 2 is shown according to this
Invent the schematic block diagram of the expansion other structure of user interest degree group of an embodiment.Wherein, each functional module in device
It can be realized by the combination of the hardware of the realization principle of the invention, software or hardware and software.Those skilled in the art can manage
Solution, Fig. 2 described function module can combine or be divided into submodule, to realize the original of foregoing invention
Reason.Therefore, description herein can support to functions described herein module it is any it is possible combination or division or
It is further to limit.Letter is done in the operation that the functional module and each functional module that only can have below with regard to device can execute
Illustrate, above description may refer to for the detail section being directed to, which is not described herein again.
Referring to fig. 2, expanding the other device 200 of user interest degree group includes taxon 210 and recommendation unit 220.
Taxon 210 is used to that user to be classified to corresponding natural quality according to the natural quality of user and is grouped, natural
Attribute includes at least two natural quality dimension for grouping, and user is to be classified to natural quality dimension with identical
The natural quality of value is grouped.Wherein, include at least one of following for the natural quality dimension of grouping: gender, age bracket,
Domain, occupation, curious degree.
Recommendation unit 220 is used to be grouped the other ranking of interest-degree group of interior big data user based on natural quality, to user
The interest-degree group of recommending in the top and user to be not yet marked or the other information of interest-degree group for belonging to recommendation.
In one embodiment of the invention, natural quality be grouped in the other ranking of interest-degree group of big data user can be with
It is the other ranking of interest-degree group for belonging to all users of the natural quality grouping, is also possible to classify in other attribute dimensions
Under belong to natural quality grouping big data user the other ranking of interest-degree group.Also, it other attribute dimensions and is used for
At least partly can be based on setting interchangeable dimension in the natural quality dimension of grouping.
In one embodiment of the invention, natural quality is grouped the other ranking of interest-degree group of interior big data user at least
It is sought according to one of following: the sum of the interest-degree group of each user in the grouping；The interest of each user in the grouping
Spend the sum of group weighting.Recommendation unit 220 can be multiple according to recommending the in the top and described user to be not yet marked as a result,
The value of the other weighted sum of interest-degree group recommends the other information of interest-degree group for belonging to recommendation of corresponding percentage.
As shown in Fig. 2, device 200 optionally can also include acquiring unit 230 shown in dotted line frame and conversion list in figure
Member 240.
Acquiring unit 230 is used to obtain user to the interest-degree group of recommendation or belongs to the other information of interest-degree group of recommendation
Click behavior.Conversion unit 240 is used to convert use for the interest-degree group of recommendation when the behavior of click is higher than threshold condition
The interest-degree group that family itself includes.
As shown in Fig. 2, device 200 can also include optionally adjustment unit 250 shown in dotted line frame in figure, for being based on
Conversion adjusts the value of the curious degree dimension of the user.
Fig. 3 is to show the schematic block diagram of the structure according to an embodiment of the invention for calculating equipment 300.Its
In, calculating equipment 300 can be and may be embodied as various types of computer installations, such as desktop computer, portable computer, flat
Plate computer, smart phone, personal digital assistant (PDA) or other kinds of computer installation, but be not limited to any specific
Form, the navigation equipment that can be such as mounted in vehicle.
As shown in figure 3, calculating equipment 300 of the invention may include processor 310 and memory 320.Processor 310
It can be the processor of a multicore, also may include multiple processors.In some embodiments, processor 310 may include
One general primary processor and one or more special coprocessors, such as at graphics processor (GPU), digital signal
Manage device (DSP) etc..In some embodiments, the circuit that customization can be used in processor 510 is realized, such as special-purpose is integrated
Circuit (application specific integrated circuit, ASIC) or field programmable gate array
(field programmable gate arrays, FPGA).
Memory 320 may include various types of storage units, such as Installed System Memory, read-only memory (ROM), and forever
Long storage device.Wherein, ROM can store the static data of other modules needs of processor 310 or computer or refer to
It enables.Permanent storage can be read-write storage device.Permanent storage can be after computer circuit breaking not
The non-volatile memory device of the instruction and data of storage can be lost.In some embodiments, permanent storage device uses
Mass storage device (such as magnetically or optically disk, flash memory) is used as permanent storage.In other embodiment, permanently deposit
Storage device can be removable storage equipment (such as floppy disk, CD-ROM drive).Installed System Memory can be read-write storage equipment or
The read-write storage equipment of volatibility, such as dynamic random access memory.Installed System Memory can store some or all processors
The instruction and data needed at runtime.In addition, memory 320 may include the combination of any computer readable storage medium,
Including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read only memory), disk and/or
CD can also use.In some embodiments, memory 120 may include that removable storage that is readable and/or writing is set
It is standby, for example, laser disc (CD), read-only digital versatile disc (such as DVD-ROM, DVD-dual layer-ROM), read-only Blu-ray Disc,
Super disc density, flash card (such as SD card, min SD card, Micro-SD card etc.), magnetic floppy disc etc..It is computer-readable to deposit
It stores up medium and does not include carrier wave and the momentary electron signal by wirelessly or non-wirelessly transmitting.
In embodiments of the present invention, executable code is stored on memory 320, processor 310 can be executed and is stored in
Executable code on memory 320.When executable code is executed by processor 310, processor 310 can be made to execute this hair
The bright other scheme of expansion user interest degree group.Wherein, place can also be stored in addition to storing executable code on memory 320
It manages device 310 and is executing some or all data needed for the other solution processes of expansion user interest degree group of the invention.
Concrete application example
Assuming that some particular user Un, the pre-generated user's representation data for it of applied business.User Un's is each
Interest-degree group and its weighted value are as follows: L_1, P_1；L_2, P_2；…；L_n, P_n, n are the natural number more than or equal to 1.
Interest-degree group is class indication of the applied business to the interest-degree of user Un, and interest-degree weight is corresponding interest
Spend group ratio shared in all interest-degree groups of user Un.For example, for some user in novel business
A, interest-degree group and weight are as follows: ancient customs, and 20%；Suspense, 30%；Science fiction, 50%.As can be seen that user A is in novel industry
There are three interest-degree groups in business, account for different ratio respectively, then novel business, can be according to interest when recommending to user A
The specific novel information that ratio shared by group recommends corresponding proportion quantity is spent, it, can be according to such as when recommending 10 novels
Percent, recommend 2 ancient customs novels, 3 suspense novels, 5 science fictions to user A.
For utilize group technology of the invention, be referred to some natural quality grouping G K user, based on be grouped G to
The process of each user's recommendation information is as follows.Herein, it such as can be related to acceptable age, gender, region, know-how and information
This five dimensions of face degree are hunted to carry out natural quality grouping.Grouping G for example can correspond to women, youth, a line city, height
Know crowd and information is dabbled face and is widely specifically grouped.
Step 1, according to K user in group, generate the other superset of interest-degree group for covering all users, L_ can be denoted as
super_set.The other weighted value of interest-degree group in the superset is initialized to 0.
Each of which other weighted value of interest-degree group is added to superset L_super_ by each user of step 2, ordered retrieval
The other weight of same interest-degree group in set.
Step 3, according to the other weighted value of interest-degree group in superset L_super_set, arranged from high to low, generate
One ordered list List_L_super_set.
The each user of step 4, ordered retrieval is its selection for recommendation in ordered list List_L_super_set
Interest-degree group can be limited to each user herein and select 3 interest-degree groups for recommendation.Wherein, interest-degree group
Specific value can be adjusted according to business demand.Select the other detailed process of interest-degree group as follows.
1) since the highest interest-degree group of the weight of ordered list List_L_super_set, successively order traversal,
Select the interest-degree group in the interest-degree group set that 3 are not present in the user.
2) the interest-degree group that this 3 are chosen is recorded in the representation data of user, and other to this 3 interest-degree groups
Weighted value carries out percentage normalized.As an example it is assumed that the group interest-degree of some user UU is identified as L1/L2/L3,
The group weighted value that the group weight that the group weighted value of L1 is 10000%, L2 is 15000%, L3 is 25000%, then L1 percentage is returned
One change after weighted value be 20%, L2 percentage normalization after weighted value be 30%, L3 percentage normalize after weighted value
It is 50%.
Step 5, the representation data storage of user.The interest-degree group or interest-degree chosen into user's recommendation step 4
The corresponding specific business information of group.
In addition, interest-degree group recommended to the user can also be converted into the interest-degree group of user itself.For example, can be with
Recommend the statistics of conversion ratio, it is when high conversion rate of the user in some interest-degree group is after the threshold value of setting, this is emerging
Interesting degree group is converted into the interest-degree group of user itself, and the interest-degree group is added to the interest-degree group set of user.
Meanwhile the statistical data can assess adjustment for " information the dabbles face degree " natural quality of user and provide according to (user is to emerging
The interesting other conversion ratio of degree group is higher, and it is wider to illustrate that its information dabbles face, can more receive new things).
Expansion user interest degree group method for distinguishing according to the present invention, dress above is described in detail by reference to attached drawing
It sets, calculate equipment.
In addition, being also implemented as a kind of computer program or computer program product, the meter according to the method for the present invention
Calculation machine program or computer program product include the calculating for executing the above steps limited in the above method of the invention
Machine program code instruction.
Alternatively, the present invention can also be embodied as a kind of (or the computer-readable storage of non-transitory machinable medium
Medium or machine readable storage medium), it is stored thereon with executable code (or computer program or computer instruction code),
When the executable code (or computer program or computer instruction code) by electronic equipment (or calculate equipment, server
Deng) processor execute when, so that the processor is executed each step according to the above method of the present invention.
Those skilled in the art will also understand is that, various illustrative logical blocks, mould in conjunction with described in disclosure herein
Block, circuit and algorithm steps may be implemented as the combination of electronic hardware, computer software or both.
The flow chart and block diagram in the drawings show the possibility of the system and method for multiple embodiments according to the present invention realities
Existing architecture, function and operation.In this regard, each box in flowchart or block diagram can represent module, a journey
A part of sequence section or code, a part of the module, section or code include one or more for realizing defined
The executable instruction of logic function.It should also be noted that in some implementations as replacements, the function of being marked in box can also
To be occurred with being different from the sequence marked in attached drawing.For example, two continuous boxes can actually be basically executed in parallel,
They can also be executed in the opposite order sometimes, and this depends on the function involved.It is also noted that block diagram and/or stream
The combination of each box in journey figure and the box in block diagram and or flow chart, can the functions or operations as defined in executing
Dedicated hardware based system realize, or can realize using a combination of dedicated hardware and computer instructions.
Various embodiments of the present invention are described above, above description is exemplary, and non-exclusive, and
It is not limited to disclosed each embodiment.Without departing from the scope and spirit of illustrated each embodiment, for this skill
Many modifications and changes are obvious for the those of ordinary skill in art field.The selection of term used herein, purport
In the principle, practical application or improvement to the technology in market for best explaining each embodiment, or make the art
Other those of ordinary skill can understand each embodiment disclosed herein.
Claims (20)
1. a kind of expansion user interest degree group method for distinguishing, comprising:
The user is classified to corresponding natural quality according to the natural quality of user to be grouped, the natural quality includes at least
Two natural quality dimensions for grouping, and the user is classified to the natural quality dimension with identical value
Natural quality grouping；
The other ranking of interest-degree group of interior big data user is grouped based on the natural quality, Xiang Suoshu user recommends in the top
And the interest-degree group that is not yet marked of the user or the other information of interest-degree group for belonging to recommendation.
2. the method for claim 1, wherein the natural quality is grouped the other row of interest-degree group of interior big data user
Name is:
Belong to the other ranking of interest-degree group of all users of the natural quality grouping；Or
Belong to the other ranking of interest-degree group of the big data user of the natural quality grouping under the classification of other attribute dimensions.
3. method according to claim 2, wherein other described attribute dimensions and the natural quality dimension for grouping
In be at least partly based on interchangeable dimension is arranged.
4. the method for claim 1, wherein the natural quality is grouped the other row of interest-degree group of interior big data user
Name is sought according at least to one of following:
The sum of the interest-degree group of each user in the grouping；
The sum of the interest-degree group weighting of each user in the grouping.
5. method as claimed in claim 4, wherein Xiang Suoshu user recommends the in the top and described user to be not yet marked
Interest-degree group or the other information of interest-degree group for belonging to recommendation include:
According to the value for the multiple other weighted sums of interest-degree group for recommending the in the top and described user to be not yet marked, recommend to correspond to
The other information of interest-degree group for belonging to recommendation of percentage.
6. the method for claim 1, wherein the natural quality dimension for grouping includes at least one following:
Gender；
Age bracket；
Region；
Occupation；
Know-how；And
Curious degree.
7. method as claimed in claim 6, further includes:
Obtain click behavior of the user to the interest-degree group of recommendation or the other information of interest-degree group for belonging to recommendation；
When the click behavior is higher than threshold condition, converting described user itself for the interest-degree group of recommendation includes
Interest-degree group.
8. the method for claim 7, further includes:
The value of the curious degree dimension of the user is adjusted based on the conversion.
9. the method for claim 1, wherein at least one in the natural quality dimension of grouping according to scheduled
Time interval regularly updates.
10. a kind of other device of expansion user interest degree group, comprising:
Taxon is grouped for the user to be classified to corresponding natural quality according to the natural quality of user, it is described from
Right attribute includes at least two natural quality dimension for grouping, and the user is to be classified to the natural quality dimension
Spending, there is the natural quality of identical value to be grouped；
Recommendation unit, for the other ranking of interest-degree group of big data user in being grouped based on the natural quality, to the use
Interest-degree group that family recommends the in the top and described user to be not yet marked or the other letter of interest-degree group for belonging to recommendation
Breath.
11. device as claimed in claim 10, wherein the interest-degree group that the natural quality is grouped interior big data user is other
Ranking is:
Belong to the other ranking of interest-degree group of all users of the natural quality grouping；Or
Belong to the other ranking of interest-degree group of the big data user of the natural quality grouping under the classification of other attribute dimensions.
12. device as claimed in claim 11, wherein other described attribute dimensions and the natural quality dimension for grouping
In degree is at least partly based on the interchangeable dimension of setting.
13. device as claimed in claim 10, wherein the interest-degree group that the natural quality is grouped interior big data user is other
Ranking is sought according at least to one of following:
The sum of the interest-degree group of each user in the grouping；
The sum of the interest-degree group weighting of each user in the grouping.
14. device as claimed in claim 13, wherein the recommendation unit is according to recommending the in the top and user not yet
The value for the multiple other weighted sums of interest-degree group being marked recommends the interest-degree group for belonging to recommendation of corresponding percentage other
Information.
15. device as claimed in claim 10, wherein the natural quality dimension for grouping includes at least one following:
Gender；
Age bracket；
Region；
Occupation；
Know-how；And
Curious degree.
16. device as claimed in claim 15, further includes:
Acquiring unit, for obtaining the user to the interest-degree group of recommendation or belonging to the other letter of interest-degree group of recommendation
The click behavior of breath；
Conversion unit, for converting institute for the interest-degree group of recommendation when the click behavior is higher than threshold condition
State the interest-degree group that user itself includes.
17. device as claimed in claim 16, further includes:
Adjustment unit, the value of the curious degree dimension for adjusting the user based on the conversion.
18. device as claimed in claim 10, wherein at least one in the natural quality dimension of grouping according to predetermined
Time interval regularly update.
19. a kind of calculating equipment, comprising:
Processor；And
Memory is stored thereon with executable code, when the executable code is executed by the processor, makes the processing
Device executes the method as described in any one of claim 1-9.
20. a kind of non-transitory machinable medium, is stored thereon with executable code, when the executable code is electric
When the processor of sub- equipment executes, the processor is made to execute method as claimed in any one of claims 1-9 wherein.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710880869.3A CN109559140A (en) | 2017-09-26 | 2017-09-26 | It expands the other method, apparatus of user interest degree group, calculate equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710880869.3A CN109559140A (en) | 2017-09-26 | 2017-09-26 | It expands the other method, apparatus of user interest degree group, calculate equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109559140A true CN109559140A (en) | 2019-04-02 |
Family
ID=65862318
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710880869.3A Pending CN109559140A (en) | 2017-09-26 | 2017-09-26 | It expands the other method, apparatus of user interest degree group, calculate equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109559140A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130290289A1 (en) * | 2012-04-30 | 2013-10-31 | Brightedge Technologies, Inc. | Integration of third party information |
| CN104156366A (en) * | 2013-05-13 | 2014-11-19 | 中国移动通信集团浙江有限公司 | Method and network server for recommending network applications to mobile terminal |
| CN104731954A (en) * | 2015-04-01 | 2015-06-24 | 天翼爱音乐文化科技有限公司 | Music recommendation method and system based on group perspective |
| CN105005582A (en) * | 2015-06-17 | 2015-10-28 | 深圳市腾讯计算机系统有限公司 | Recommendation method and device for multimedia information |
| CN105574216A (en) * | 2016-03-07 | 2016-05-11 | 达而观信息科技（上海）有限公司 | Personalized recommendation method and system based on probability model and user behavior analysis |
-
2017
- 2017-09-26 CN CN201710880869.3A patent/CN109559140A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130290289A1 (en) * | 2012-04-30 | 2013-10-31 | Brightedge Technologies, Inc. | Integration of third party information |
| CN104156366A (en) * | 2013-05-13 | 2014-11-19 | 中国移动通信集团浙江有限公司 | Method and network server for recommending network applications to mobile terminal |
| CN104731954A (en) * | 2015-04-01 | 2015-06-24 | 天翼爱音乐文化科技有限公司 | Music recommendation method and system based on group perspective |
| CN105005582A (en) * | 2015-06-17 | 2015-10-28 | 深圳市腾讯计算机系统有限公司 | Recommendation method and device for multimedia information |
| CN105574216A (en) * | 2016-03-07 | 2016-05-11 | 达而观信息科技（上海）有限公司 | Personalized recommendation method and system based on probability model and user behavior analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103823908B (en) | Content recommendation method and server based on user preference | |
| Chen et al. | Learning to rank features for recommendation over multiple categories | |
| Moro et al. | Factors influencing hotels’ online prices | |
| Lazari et al. | Designs of discrete choice set experiments for estimating both attribute and availability cross effects | |
| Dowell et al. | Understanding value creation and word-of-mouth behaviour at cultural events | |
| Grisolía et al. | A latent class model of theatre demand | |
| Keller et al. | The effects of brand name suggestiveness on advertising recall | |
| CN108416625A (en) | The recommendation method and apparatus of marketing product | |
| CN109902708A (en) | A kind of recommended models training method and relevant apparatus | |
| CN109597973A (en) | A kind of recommendation, generation method and the device of official documents and correspondence information | |
| CN108829808A (en) | A kind of page personalized ordering method, apparatus and electronic equipment | |
| Caldito et al. | Tourist behaviour and trends | |
| Adirinekso et al. | Measurement of performance, effort, social influence, facilitation, habit and hedonic motives toward pay later application intention: Indonesia evidence | |
| CN109242593A (en) | A kind of clothes recommended method, device and the storage medium of case-based reasioning | |
| Scheider et al. | Distinguishing extensive and intensive properties for meaningful geocomputation and mapping | |
| CN114862516A (en) | Document recommendation method, storage medium, and program product | |
| Russell et al. | Motivations behind posting travel pictures for status: Developing and testing the conspicuous consumption posting scale (CCPS) | |
| von Rimscha | It’s not the economy, stupid! External effects on the supply and demand of cinema entertainment | |
| CN116188120B (en) | Method, device and system for recommending audio books and storage medium | |
| Shaker Ardekani | Tracing musical tastes in Tehran: How urbanism selects its sound | |
| Thorogood et al. | A framework for computer-assisted sound design systems supported by modelling affective and perceptual properties of soundscape | |
| CN109559140A (en) | It expands the other method, apparatus of user interest degree group, calculate equipment and storage medium | |
| Oey et al. | Integrating factor analysis and multi dimensional scaling for brand positioning of e-commerce platforms | |
| CN110035111A (en) | A kind of method for pushing and device dispensing entity | |
| McIlwraith | Algorithms of the intelligent web |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200709 Address after: 310052 room 508, floor 5, building 4, No. 699, Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Applicant after: Alibaba (China) Co.,Ltd. Address before: 510627 Guangdong city of Guangzhou province Whampoa Tianhe District Road No. 163 Xiping Yun Lu Yun Ping B radio square 14 storey tower Applicant before: Guangzhou Dongjing Computer Technology Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190402 |