Invention content
In view of this, technical problems to be solved in this application are there is provided a kind of method of alternate analysis user behavior,
By for user setting dimension, and then alternate analysis user behavior data, the Behavior law of user is therefrom obtained, and then infer each
Class user psychology and conditions of demand.
In order to solve the above-mentioned technical problem, the application has following technical solution:
A kind of method of alternate analysis user behavior, including:
Customer data base is established, several dimensions are set in the customer data base, are corresponded in each dimension one or more
Tag name;
The user data of multiple users, the use of each user are acquired and preserved in the subscriber data set of customer data base
User data includes the matching tag name of the User ID of the user and the user in one or more dimension;
It is acquired in the tag set of customer data base and preserves corresponding label data in each dimension, each label data
Including tag name and several user's business cards for matching the label;
Default alternate analysis calculative strategy calls subscriber data set and/or mark according to the alternate analysis calculative strategy
Label set, and extract user data and/or label data in subscriber data set and/or tag set;
Analytic operation is carried out to the user data and/or label data that are extracted using the alternate analysis calculative strategy,
Obtain user behavior data.
It is described to include using alternate analysis calculative strategy progress analytic operation:
Project sort operation, screening operation and/or set intersection union.
The method further includes:
User list is set, the User ID of all users is preserved in the user list;
List of labels is set, all tag names are preserved in the list of labels.
The dimension includes:
Age of user, user's gender, user's occupation, user tendency, user activity and/or user operation records.
When the dimension is specially age of user, then the corresponding tag name of the dimension includes:
0 ~ 15 year old, 15 ~ 25 years old, 25 ~ 35 years old, 35 ~ 50 years old and 50 years old or more.
When the dimension is specially user's gender, then the corresponding tag name of the dimension includes:
Male and female.
User's business card includes:
User ID, user's weighted index and User support index.
Compared with prior art, method and system described herein has reached following effect:
(1)Establish customer data base and a large amount of gathered datas so that user behavior analysis is provided with more massive data
Sample so that analysis result is more accurately enriched;
(2)By the way that diversified alternate analysis calculative strategy is arranged, open user behavior analysis, Neng Gougeng are realized
It is extensive to obtain the potential Behavior law of user;
(2)By obtained user behavior data, it can be better understood by the psychological activity and use demand of user, carried
High user experience.
Certainly, implementing any product of the application must be not necessarily required to reach all the above technique effect simultaneously.
Specific implementation mode
Some vocabulary has such as been used to censure specific components in specification and claim.Those skilled in the art answer
It is understood that hardware manufacturer may call the same component with different nouns.This specification and claims are not with name
The difference of title is used as the mode for distinguishing component, but is used as the criterion of differentiation with the difference of component functionally.Such as logical
The "comprising" of piece specification and claim mentioned in is an open language, therefore should be construed to " include but do not limit
In "." substantially " refer in receivable error range, those skilled in the art can be described within a certain error range solution
Technical problem basically reaches the technique effect.In addition, " coupling " word includes any direct and indirect electric property coupling herein
Means.Therefore, if it is described herein that a first device is coupled to a second device, then representing the first device can directly electrical coupling
It is connected to the second device, or the second device indirectly electrically coupled through other devices or coupling means.Specification
Subsequent descriptions be implement the application better embodiment, so it is described description be for the purpose of the rule for illustrating the application,
It is not limited to scope of the present application.The protection domain of the application is when subject to appended claims institute defender.
The alternate analysis based on user behavior is a kind of analysis mode that opening is extremely strong.In analysis number of users
According to before, or even it is not necessarily required to one expected result of setting.In fact, in conjunction with data analysis user behavior, precisely in order to
From statistical angle try to find out it is certain can not intuitively discover, user potential Behavior law comes further whereby
Speculate the psychological needs of user.Finally again by the psychological needs of user, it is transmitted in the middle of the design of product.To product
There is specific specific aim for the psychological needs of user in design and improvement.It is having for a kind of raising product user experience
Effect means.
Exactly according to this process in the application, the comprehensive analysis of more in-depth has been carried out to user data.Namely it is
The a variety of dimensions of user setting, and user behavior data is acquired on a large scale;And then on each dimension, to large-scale user
Data carry out the analytic operation of all kinds of openings.Referring specifically to following embodiment.
It is shown in Figure 1, it is the specific embodiment of herein described method.Method described in the present embodiment specifically include with
Lower step:
Step 101 establishes customer data base, and several dimensions are arranged in the customer data base, one is corresponded in each dimension
A or multiple tag names.
Be different from the mode of the mode of traditional sampling questionnaire survey in the present embodiment, will be first collect and count with
The related Various types of data of user behavior, establishes special customer data base.And several users in the database, are arranged to retouch
State the dimension of user.For example, the information such as the age of user, gender, occupation, tendentiousness, liveness and operation note, belong to retouch
State the dimension of user.
And several tag names is corresponded under each dimension, represents attribute of the user in the dimension.Such as this
In embodiment, age, gender and occupation are the dimensions for describing user's self character.On this dimension of age, corresponding it can mark
Signature has<0 ~ 15 year old>、<15 ~ 25 years old>、<25 ~ 35 years old>、<35 ~ 50 years old>With<50 years old or more>, each user in the dimension,
It is inevitable that there are one the tag names to match with own situation.Equally, the corresponding tag name of gender dimension includes<Male>With<Women
>;The corresponding tag name of professional dimension includes<Engineer>、<Teacher>、<Doctor>、<Law practitioner>Or<Civil servant>Deng
Deng.The dimensions such as tendentiousness, liveness and operation note describe the concrete condition that user uses network service.Tendentiousness dimension
Can weigh user interest tendency, including tag name for example have<TV play>、<Film>Or<Variety show>Deng.Liveness
Dimension can weigh the frequent degree that user participates in network service.Operation note dimension specifically has recorded in the browsing of user
Hold.
Since the opening of analysis method described in the present embodiment is extremely strong, the user information sample of preservation is bigger, in covering
Appearance is more comprehensive, then the result analyzed will more be enriched.So in actual customer data base, including dimension and dimension
It is that type is extremely more to spend corresponding tag name, no longer enumerated here more.
Step 102, the user data for being acquired in the subscriber data set of customer data base and preserving multiple users, each
The user data of user includes the matching tag name of the User ID of the user and the user in one or more dimension.
The subscriber data set is using individual consumer as base unit, including multiple or perhaps a large amount of user data.
Each user data describes all kinds of association attributes of a user.Specifically, user data has included the user of the user
The matching tag name of ID and the user in one or more dimension.
For example, for user A, user data is represented by following:
<User A_0001;Male;25 ~ 35 years old;Engineer;Variety show;Quantity 3 is browsed weekly;Browsing " Guo De recently
Guiding principle ">.Wherein " user A_0001 " is the ID of the user, remaining item is tag name of the user in certain specific dimension.
According to above-mentioned pattern, each user is provided with a similar user data.So the customer data base can be with
The user data of a large amount of even whole users of acquisition, is saved in the subscriber data set.
Step 103 acquires in the tag set of customer data base and preserves corresponding label data in each dimension, each
Label data includes tag name and several user's business cards for matching the label.
In the customer data base, there is also a tag sets.With label data for substantially singly in the tag set
Position is preserved.Necessarily there are a certain number of user groups to be provided with the same tag name in customer data base, then the label
Name and its user group constitute a label data.It will include a tag name in a namely label data, and use
User's business card of user in the group of family.May include that User ID, user's weighted index and User support refer in user's business card
Three kinds of information of number.Weighted index means influence power of the user in the user group;Index support mean the user with
The height of the tag name matching degree.
For example, label<Film>Label data it is specific as follows:
<Film>;<User J_0010;xj;yj>;<User K_0011;xk;yk>;<User L_0012;xl;yl >。
Wherein xj、xk、xlFor weighted index, yj、yk、ylFor index support, equal coefficient word.
In fact, user group's quantity of a label is perhaps thousands of, the label is only simply shown herein
The form of data.
According to above-mentioned pattern, each tag name is provided with a similar label data.So the customer data base can
To acquire the label data of a large amount of even whole tag names, it is saved in the label data set.
By step 101 ~ step 103 in the present embodiment, the foundation of customer data base is realized, and in user data set
Close and label data set in save data resource abundant enough, with for carrying out the analysis of user behavior.
In addition preferred, for ease of being safeguarded to subscriber data set and tag set, can also be set in the present embodiment
User list is set, the User ID of all users is preserved in the user list;And list of labels is set, in the list of labels
Preserve all tag names.All User ID and tag name are directly managed by above-mentioned two list.
Step 104, default alternate analysis calculative strategy, subscriber data set is called according to the alternate analysis calculative strategy
And/or tag set, and extract user data and/or label data in subscriber data set and/or tag set.
It is risen to step 104, that is, starts with the data resource of above-mentioned preservation, that is, user data and label data carry out
The analysis of user behavior.
Aforementioned known, the user behavior analysis described in the present embodiment is a kind of analysis of opening, in order to from
Statistical angle try to find out it is certain can not intuitively discover, user potential Behavior law.Even it is not required setting
One expected analysis result.So corresponding analytic process, that is, the content of the alternate analysis calculative strategy should be
It is very diversified, arbitrarily it can be arranged and adjust.
Based on the above, in the present embodiment, the alternate analysis calculative strategy is not limited specifically.It is every relevant
Calculation and analysis methods are counted, it can be in conjunction under the overall technical architecture of the present embodiment.In general, alternate analysis calculates
The analytic operation that strategy includes mainly has projection sort operation, screening operation and/or set intersection union etc. several important
Operation mode is independent or is used in conjunction with one another.
Step 105 divides the user data and/or label data that are extracted using the alternate analysis calculative strategy
Operation is analysed, user behavior data is obtained.
User data/label data in customer data base is calculated and divided using the alternate analysis calculative strategy
Analysis, just it can be found that certain potential rules on user behavior.It can be using this rule as user behavior data in the present embodiment.
Intuitive not noticeable user behavior feature is often shown in the user behavior data, can help to infer
Go out all types of user when being serviced using network, more the psychological activity and conditions of demand of deep layer;And in product design and optimization
During, the psychological activity and conditions of demand of user are efficiently used, achievees the purpose that improve user experience.
As seen through the above technical solutions, advantageous effect existing for the present embodiment is:It establishes customer data base and largely adopts
Collect data so that user behavior analysis is provided with more massive data sample so that analysis result is more accurately enriched;Pass through
Diversified alternate analysis calculative strategy is set, realizes open user behavior analysis, can widely obtain user
Potential Behavior law improves user experience in order to subsequently be better understood by the psychological activity and use demand of user.
It is shown in Figure 2, it is another specific embodiment of herein described method.It will be shown in Fig. 1 in the present embodiment
On the basis of embodiment, a kind of alternate analysis calculative strategy is more specifically described, so that the application overall technical architecture
It is more convenient for understanding.The data acquisition phase of overall technical architecture is consistent with 101 ~ step 103 of abovementioned steps in the present embodiment, herein
It is not repeated narration.Only description carries out the detailed process of user behavior analysis using alternate analysis calculative strategy.
User data in step 201, extraction subscriber data set finds out band in tendentiousness dimension using screening operation
Have<Father go where>The user of label obtains user group α.
In the present embodiment, the first step of user behavior analysis is to filter out a specific user group.Specifically, sieve
It selects to the interested certain customers of variety show " father where go ".In customer data base it can be found that<Father go where
>Belong to a label in tendentiousness dimension, so being found with the complete of the label from subscriber data set in the present embodiment
Portion's user data, as user group α to get to " father where go " interested user group.
Step 202, using filtering algorithm, find out in user group α carry label<Beijing>User, obtain user group
Body β.
After obtaining user group α, further user group α is further screened using region dimension, is obtained
To Beijing user in user group α.In customer data base,<Beijing>For a label in the dimension of region, thus with
It finds and carries in family group α<Beijing>The user of label, that obtain is exactly Beijing user group for paying close attention to " father where go ", this
Certain customers group, that is, user group β.
Step 203, using filtering algorithm, find out in user group β carry label<iphone>User, obtain user
Group γ.
With it is aforementioned similarly, further filtered out in user group β and receive the user group of network service using iphone mobile phones
Body, as user group γ.
That is the use in user group γ meets following characteristics per family:Using the concern of iphone mobile phones, " which father goes
Beijing user of youngster ".
Step 204 calculates the percentage that user group γ accounts for user group β, and using percent data as user behavior
Data.
Based on analysis result obtained above, the percentage that user group γ accounts for user group β can be further calculated.
Namely learn that the user of much ratios in user group β is to utilize iphone mobile phones concern " father where go ".
Assuming that the data result obtained in the present embodiment is the percentage 75% that user group γ accounts for user group β.This just says
Bright, in Beijing, majority likes the user of " father where go " iphone mobile phones, or explanation is used to use the use of iphone mobile phones
There is higher interest at family to " father where go ".The user that is learned in above i.e. the present embodiment is potential with for rule.
It is obtained by analysis to draw a conclusion in the present embodiment:It " is used in the user 75% of Beijing concern " father where go "
Iphone mobile phones ".The conclusion can be used as this and analyze obtained user behavior data.
As seen through the above technical solutions, advantageous effect existing for the present embodiment is:Specifically describe a kind of intersection point
Analyse calculative strategy so that the application overall technical architecture is more complete, open more abundant.
By the above various embodiments it is found that advantageous effect existing for the application is:
(1)Establish customer data base and a large amount of gathered datas so that user behavior analysis is provided with more massive data
Sample so that analysis result is more accurately enriched;
(2)By the way that diversified alternate analysis calculative strategy is arranged, open user behavior analysis, Neng Gougeng are realized
It is extensive to obtain the potential Behavior law of user;
(2)By obtained user behavior data, it can be better understood by the psychological activity and use demand of user, carried
High user experience.
It should be understood by those skilled in the art that, embodiments herein can be provided as method, apparatus or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, the application can be used in one or more wherein include computer usable program code computer
Usable storage medium(Including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)The computer program of upper implementation produces
The form of product.
Several preferred embodiments of the application have shown and described in above description, but as previously described, it should be understood that the application
Be not limited to form disclosed herein, be not to be taken as excluding other embodiments, and can be used for various other combinations,
Modification and environment, and the above teachings or related fields of technology or knowledge can be passed through in the scope of the invention is set forth herein
It is modified.And changes and modifications made by those skilled in the art do not depart from spirit and scope, then it all should be in this Shen
It please be in the protection domain of appended claims.