CN106033581A - Analysis method of behavior data and apparatus thereof - Google Patents
Analysis method of behavior data and apparatus thereof Download PDFInfo
- Publication number
- CN106033581A CN106033581A CN201510117145.4A CN201510117145A CN106033581A CN 106033581 A CN106033581 A CN 106033581A CN 201510117145 A CN201510117145 A CN 201510117145A CN 106033581 A CN106033581 A CN 106033581A
- Authority
- CN
- China
- Prior art keywords
- data
- dimension
- dimensional information
- behavioral data
- historical behavior
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an analysis method of behavior data and an apparatus thereof. The method comprises the following steps of collecting behavior data which needs to be predicted, wherein the behavior data includes dimension information on at least one dimension; acquiring historical behavior data and historical dimension information corresponding to the historical behavior data on at least one dimension; and according to a prediction model established in advance, comparing the behavior data and the historical behavior data which need to be predicted on the corresponding dimension and acquiring an analysis result of the behavior data. In the prior art, a behavior implementation result analysis process is time-consuming and power-consuming; an implementation opportunity can not be grasped; and an implementation result is difficult to predict before the implementation. By using the method and the apparatus of the invention, the above problems are solved.
Description
Technical field
The present invention relates to data analysis field, in particular to the analysis method and device of a kind of behavioral data.
Background technology
For people's behavioral data, such as: marketing behavior, recommendation information, market public relation etc., the issue of behavior
Person is most concerned with the result after its behavior is issued, and this result generally includes economic consequence, attention rate and degree of recognition.
The object information of this feedback not only facilitates behavior publisher and understands the impact that the most issued data cause, and also makes
Obtain when publisher issues data afterwards and can pass through planning in advance, to reach expected results as far as possible.With common
As a example by marketing behavior, publisher selects published method, carries out each marketing behavior relevant special often through artificial behavior
Determine the modeling of attribute, collect historical data, by reaching a conclusion for a long time, often delay the optimal of marketing behavior
Action opportunity, takes time and effort.Or carrying out public opinion monitoring during marketing behavior, the very first time is to marketing behavior
Effect is fed back, but is only capable of after behavior is initiated just being monitored, it is impossible to the result of analytical behavior, and is expert at
During for existing problems, it is difficult to the problem of prediction behavior itself, cause the undesirable even counter productive of behavior outcome.
Take time and effort for behavior result of implementation is analyzed process by prior art, action opportunity, behavior cannot be held in fact
It is difficult to predict the problem of result of implementation before executing, the most not yet proposes effective solution.
Summary of the invention
Present invention is primarily targeted at the analysis method and device that a kind of behavioral data is provided, to solve in prior art
To behavior result of implementation analyze process take time and effort, cannot hold action opportunity, behavior implement before be difficult to prediction implement knot
The problem of fruit.
To achieve these goals, an aspect according to embodiments of the present invention, it is provided that the analysis of a kind of behavioral data
Method.The analysis method of the present invention includes: gathering the behavioral data that needs are predicted, wherein, behavioral data includes:
Dimensional information at least one dimension；Obtain historical behavior data and historical behavior data at least one dimension
Corresponding history dimensional information；Behavioral data needs being predicted according to the forecast model pre-build and history row
Compare in corresponding dimension for data, obtain the analysis result of behavioral data.
To achieve these goals, another aspect according to embodiments of the present invention, it is provided that the analysis of a kind of behavioral data
Device.The inventive system comprises: acquisition module, for gathering the behavioral data needing to be predicted, wherein, OK
Include for data: the dimensional information at least one dimension；First acquisition module, be used for obtaining historical behavior data with
And the history dimensional information that historical behavior data are corresponding at least one dimension；Analyze module, for according to building in advance
The behavioral data that needs are predicted by vertical forecast model is compared in corresponding dimension with historical behavior data,
Obtain the analysis result of behavioral data.
According to inventive embodiments, according to substantial amounts of history case, behavior number needs being predicted by forecast model
Compare in corresponding dimension according to historical behavior data, draw the analysis result needing to be predicted behavioral data.
Solve behavior result of implementation is analyzed by technology process take time and effort, action opportunity cannot be held, behavior difficult before implementing
To predict the problem of result of implementation, reach quick and precisely to predict the effect of behavior outcome before behavior is implemented.
Accompanying drawing explanation
The accompanying drawing of the part constituting the application is used for providing a further understanding of the present invention, and the present invention's is schematic real
Execute example and illustrate for explaining the present invention, being not intended that inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is the flow chart of the analysis method of behavioral data according to embodiments of the present invention；And
Fig. 2 is the schematic diagram of the analytical equipment of behavioral data according to embodiments of the present invention.
Detailed description of the invention
It should be noted that in the case of not conflicting, the embodiment in the application and the feature in embodiment can phases
Combination mutually.Describe the present invention below with reference to the accompanying drawings and in conjunction with the embodiments in detail.
In order to make those skilled in the art be more fully understood that the present invention program, below in conjunction with in the embodiment of the present invention
Accompanying drawing, is clearly and completely described the technical scheme in the embodiment of the present invention, it is clear that described embodiment
It is only the embodiment of a present invention part rather than whole embodiments.Based on the embodiment in the present invention, ability
The every other embodiment that territory those of ordinary skill is obtained under not making creative work premise, all should belong to
The scope of protection of the invention.
It should be noted that term " first " in description and claims of this specification and above-mentioned accompanying drawing, "
Two " it is etc. for distinguishing similar object, without being used for describing specific order or precedence.Should be appreciated that this
The data that sample uses can be exchanged in the appropriate case, in order to embodiments of the invention described herein.Additionally, term
" include " and " having " and their any deformation, it is intended that cover non-exclusive comprising, such as, comprise
The process of series of steps or unit, method, system, product or equipment are not necessarily limited to those steps clearly listed
Rapid or unit, but can include that the most clearly list or intrinsic for these processes, method, product or equipment
Other step or unit.
Embodiment 1
Embodiments provide a kind of analysis method of behavioral data.
Fig. 1 is the flow chart of the analysis method of behavioral data according to embodiments of the present invention.As it is shown in figure 1, the behavior
The analysis method of data comprises the following steps that
Step S20, gathers the behavioral data needing to be predicted, and wherein, behavioral data includes: at least one dimension
On dimensional information；
Concrete, the dimension of the behavioral data in above-mentioned steps S20 and dimensional information can be one, it is also possible to be
Multiple, such as: dimension can be the supporting rate of the product object preset, neutral rate and opposition rate, corresponding dimension letter
Cease e.g. supporting rate 40%, neutral rate 50%, opposition rate 10%；Or dimension is behavior duration span,
Corresponding dimensional information can be one month.
Step S40, obtains the history dimension that historical behavior data and historical behavior data are corresponding at least one dimension
Degree information；
Concrete, historical behavior data are collected the most in advance, including the history dimensional information of each dimension and correspondence.
At least one dimension herein refers to the dimension identical with the behavioral data being predicted in step S20.Such as: behavior
The dimension of data is the supporting rate of default product object, neutral rate and opposition rate, needs to obtain corresponding history dimension
Information, e.g. supporting rate 30%, neutral rate 50%, opposition rate 20%.The dimension of behavioral data is time span,
Need to obtain corresponding history dimensional information, can be two weeks.
Step S60, behavioral data needs being predicted according to the forecast model pre-build and historical behavior data
Corresponding dimension is compared, obtains the analysis result of behavioral data.
Concrete, it was predicted that model pre-builds, it was predicted that model dimension can include the product object scale of construction, time span
Degree, the supporting rate of product object, neutral rate and opposition rate, important historical behavior, product object attention rate, product pair
As sales value etc..The analysis mode of forecast model be by needs prediction behavioral data with historical behavior data corresponding
Dimension contrast, the dimension of above-mentioned correspondence refer to need prediction behavioral data with historical behavior data must be
Same dimension, the behavioral data dimension needing prediction is the default product object scale of construction, the pin of default product object
Selling value, corresponding historical behavior data dimension is the historical product object scale of construction, the sales value of historical product object equally.
The dimension of this forecast model and alignments can store in the server.By the data prestored and prediction
Model, by modeling analysis, can be predicted the outcome fast and accurately, be solved behavioral data in prior art
Analysis result takes time and effort, the problem that behavior is difficult to before implementing predict result of implementation.
Needing exist for explanation, the analysis result of behavioral data can be the sales volume value of default product object, it is possible to
It it is the supporting rate etc. of default product object.Such as, the analysis result of behavioral data can be default product object pin
Amount increase by 20% or the product object supporting rate preset increase by 10%.
Optionally, in above-mentioned steps S60, behavioral data needs being predicted according to forecast model and historical behavior
Data are compared in corresponding dimension, and the step of the analysis result obtaining behavioral data may include that
Step S601, behavioral data needs being predicted according to the forecast model pre-build and multiple historical behaviors
Data are compared in corresponding dimension respectively, obtain corresponding with the behavioral data that described needs are predicted multiple
Preprocessing result；
Concrete, get N group historical behavior data, it would be desirable to the behavioral data being predicted is gone through with each group respectively
History behavioral data is compared in corresponding dimension, obtains the Preprocessing result of N number of behavior data.
Step S602, carries out denoising to obtaining multiple Preprocessing result；
Concrete, there is noise data in the Preprocessing result of the N number of behavioral data obtained, noise data is by shadow
Ring behavioral data precision of analysis, in order to obtain the analysis result of behavioral data more accurately, need pre-place
Reason analysis result carries out denoising.Such as: the N number of Preprocessing result obtained meets normal distribution, and normal state is divided
Cloth rear and front end may be considered noise data, affects the precision of analysis of behavioral data, therefore can remove
The data of normal distribution rear and front end.
Step S603, calculating that the multiple Preprocessing results after denoising are averaged, obtain behavior number
According to analysis result.
Concrete, after removing noise data, remaining all Preprocessing results are the analysis results to behavioral data
There are the data of actual reference significance, calculate the analysis result of behavioral data according to remaining all Preprocessing results.
In above-mentioned steps S601, the behavioral data being predicted by needs according to the forecast model pre-build is gone through with multiple
History behavioral data is compared in corresponding dimension respectively, obtains corresponding many with the behavioral data that needs are predicted
Individual Preprocessing is as a result, it is possible to include:
Step S6011, obtains historical behavior result Y that any one historical behavior data is corresponding；
Concrete, historical behavior result can be economically, and the sales volume such as default product object changes, it is possible to be
Preset the supporting rate of product object, the attention rate etc. of default product object, such as historical behavior result is the increase in
The sales volume of 40%.
Concrete, historical behavior result Y can calculate in the following way, such as when historical behavior result Y is for selling
During increased percentage,
Such as when historical behavior result Y is share price amount of increase,
Above-mentioned N can preset as required.
Step S6012, is calculated the Preprocessing result of the behavioral data that needs are predicted by equation below
X:
Wherein, i is the number of dimension, is natural number, xiFor the dimensional information in i-th dimension, aiTie up for i-th
The weighted value of the dimensional information on degree, yiFor the history dimensional information in i-th dimension, biFor in i-th dimension
The weighted value of history dimensional information.
Herein it should be noted that above-mentioned calculated Preprocessing result X can be with multiple historical behavior numbers
Based on first historical behavior data according to, calculated Preprocessing result, and with multiple historical behaviors
Based on other historical behavior data in data, calculate the calculating that the Preprocessing result of behavior data is used
Formula is identical with the above-mentioned formula for being calculated X.
Concrete, the analysis result of behavioral data is based on historical behavior result and calculates, such as historical behavior result
Being the ratio of sales volume rising, the analysis result of the behavioral data that correspondence draws also is the ratio that sales volume rises.Wherein, ai and
biFor characterizing the default value set according to the behavioral data preset with historical behavior data dependence.I is natural number,
The behavioral data representing default has multiple dimensional information that multiple dimension is corresponding with historical behavior data.
Optionally, when company A needs to predict certain market behavior, the dimension of behavior and the dimension of correspondence will be implemented
Data include: time span: 1 month；The product object scale of construction preset: 500,000,000 market values；The product object support preset
Rate: 40%；The product object neutrality rate preset: 50%；The product object opposition rate preset: 10%；The product preset
Object supporting rate variation tendency: increase by 20%；The product object sales value preset: 1,500,000,000；Behavioral agent historical act
Importance: 50；Behavioral agent historical act attention rate: 380,000；Behavioral agent historical act coverage: 710,000；
Behavioral agent historical act supporting rate variation tendency: increase by 50%.The dimension of B houses market behavior and the dimension of correspondence
Data include: time span: 0.5 month；The product object scale of construction preset: 1,500,000,000 market values；The product object preset
Supporting rate: 30%；The product object neutrality rate preset: 50%；The product object opposition rate preset: 20%；Preset
Product object supporting rate variation tendency: increase by 10%；The product object sales value preset: 4,500,000,000；Behavioral agent history
Movable importance: 32；Behavioral agent historical act attention rate: 670,000；Behavioral agent historical act coverage: 107
Ten thousand；Behavioral agent historical act supporting rate variation tendency: increase by 35%；Historical behavior result: sales value increases by 40%.
Weighted value ai=1, weighted value bi=1 can be calculated by above-mentioned dimension data:
X=-0.74
In above-mentioned steps S60, behavioral data needs being predicted according to the forecast model pre-build and history row
Comparing in corresponding dimension for data, before obtaining the analysis result of behavioral data, the method may include that
Receive the amendment information of outside input, revise forecast model.
Concrete, it was predicted that model can be modified according to arranging of external information.Such as, for default product
Object is different, according to the significance level of dimension, for the weighted value that different dimension set is different, to obtain more accurately
The analysis result of behavioral data.
Optionally, in above-mentioned steps S40, tie up at least one obtaining historical behavior data and historical behavior data
Before history dimensional information corresponding on degree, the method may include that
Multiple history dimensional information corresponding for historical behavior data are stored in multiple data base by step S301, wherein,
It is associated by same behavior parameter between each data base.Behavioral parameters in this step refers to institute in dimensional information
The concrete behavior data comprised, such as, can be to hold certain theme charitable activity in dimensional information, it is also possible to be to hold
Certain theme marathon grand prix.
Concrete, persistently the dimensional information of the different behavior of storage, accumulates substantial amounts of historical data, and dimensional information is pressed
Classifying according to dimension, the dimensional information belonging to identical dimensional is stored in same data base, multiple dimensions and many numbers
According to storehouse one_to_one corresponding.By said method, substantial amounts of historical behavior information is stored in data base, can be easily
Query history behavioral data, improves the efficiency obtaining behavioral data analysis result.
Needing exist for explanation, above-mentioned dimension can be adjusted, and increases dimension, can make the dimension of analysis more
Comprehensively, the conclusion type that model can draw also will increase.Meanwhile, dimension can also be modified, be deleted.
After multiple history dimensional information corresponding for historical behavior data are stored in multiple data base by step S301,
The method that the present embodiment provides can also include:
Step S3021, gathers new history dimensional information；
Concrete, it being continuously increased new history dimensional information, the analysis result that can make behavioral data is more accurate.
Step S3022, stores new history dimensional information in new data base；
Concrete, it is stored in same data base with the dimensional information of dimension, sets up new number for new dimensional information
According to storehouse.
Step S3023, new data base is by same behavior parameter and any one or more database associations.
Concrete, new data base is associated with original multiple data bases, can be to be associated with one of them,
Can also be to be associated with plurality of data base.
In conjunction with above-described embodiment one, below, the scheme just provided the embodiment of the present application one with following instantiation is carried out
Describe in detail:
Step 1, when company A needs to predict certain market behavior, will implement the dimension of behavior and the dimension of correspondence
Data include: time span: 1 month；The product object scale of construction preset: 500,000,000 market values；The product object support preset
Rate: 40%；The product object neutrality rate preset: 50%；The product object opposition rate preset: 10%；The product preset
Object supporting rate variation tendency: increase by 20%；The product object sales value preset: 1,500,000,000；Behavioral agent historical act
Importance: 50；Behavioral agent historical act attention rate: 380,000；Behavioral agent historical act coverage: 710,000；
Behavioral agent historical act supporting rate variation tendency: increase by 50%.
Step 2, the dimension of B houses market behavior and the dimension data of correspondence include: time span: 0.5 month；In advance
If the product object scale of construction: 1,500,000,000 market values；The product object supporting rate preset: 30%；The product object preset is neutral
Rate: 50%；The product object opposition rate preset: 20%；The product object supporting rate variation tendency preset: increase by 10%；
The product object sales value preset: 4,500,000,000；Behavioral agent historical act importance: 32；Behavioral agent historical act is closed
Note degree: 670,000；Behavioral agent historical act coverage: 1,070,000；The change of behavioral agent historical act supporting rate becomes
Gesture: increase by 35%；Historical behavior result: sales value increases by 40%.
Step 3, weighted value ai=1, weighted value biAccording to weight calculation formula when=1:
Thus can draw result X that the activity that will perform drawn by B case may be gathered in the cropsb=-0.74.
Step 4, when the historical behavior data of acquisition for organizing more, such as C, D, E, count respectively according to above-mentioned steps 3
Calculate effect X that the activity that will perform may be gathered in the cropsc=-0.42, Xd=-0.7, Xe=-0.65.
Step 5, presets the activity dependence of B, C, D, E, such as: the activity dependence of B is kb=0.9,
The activity dependence of C is kcThe activity dependence of=0.8, D is kdThe activity dependence of=0.6, E is ke=0.3.
According to activity dependence threshold value k set in advance, such as k=0.6, reject the dependency activity less than 0.6
E, the meansigma methods calculating remaining B, C, D movable is:
Step 6, calculates according to formula and will perform movable final result:
Understand X=-0.624.
By according to foregoing, need this activity of prediction that sales volume will be made to reduce by 62.4%.
To sum up, the present invention according to substantial amounts of history case, behavioral data needs being predicted by forecast model with
Historical behavior data are compared in corresponding dimension, draw the analysis result needing to be predicted behavioral data.Solve
In technology of having determined to behavior result of implementation analyze process take time and effort, cannot hold action opportunity, behavior implement before be difficult to
The problem of prediction result of implementation, has reached quick and precisely to predict the effect of behavior outcome before behavior is implemented.
Embodiment 2
Embodiments provide the analytical equipment of a kind of behavioral data.It should be noted that the embodiment of the present invention
The analytical equipment of behavioral data may be used for performing the analysis method of the behavioral data that the embodiment of the present invention is provided, this
The analysis method of the behavioral data of bright embodiment can also be filled by the analysis of the behavioral data that the embodiment of the present invention is provided
Put and perform.
Fig. 2 is the flow chart of the analytical equipment of behavioral data according to embodiments of the present invention.As in figure 2 it is shown, the behavior
The analytical equipment of data may include that
Acquisition module 40, for gathering the behavioral data needing to be predicted, wherein, behavioral data includes: at least one
Dimensional information in individual dimension.
Concrete, concrete, the dimension of the behavioral data in above-mentioned acquisition module 40 and dimensional information can be one,
Can also be multiple, such as: dimension can be the supporting rate of the product object preset, neutral rate and opposition rate, corresponding
Dimensional information can be supporting rate 40%, neutral rate 50%, opposition rate 10%；Or dimension be behavior lasting time
Between span, corresponding dimensional information can be one month.
First acquisition module 42, is used for obtaining historical behavior data and historical behavior data is right at least one dimension
The history dimensional information answered.
Concrete, historical behavior data are collected the most in advance, including the history dimensional information of each dimension and correspondence.
At least one dimension herein refers to the dimension identical with the behavioral data being predicted in acquisition module 40.Such as: OK
For the supporting rate that dimension is default product object of data, neutral rate and opposition rate, need to obtain corresponding history dimension
Degree information, can be supporting rate 30%, neutral rate 50%, opposition rate 20%.The dimension of behavioral data is time span,
Need to obtain corresponding history dimensional information, can be two weeks.
Analyze module 44, for the behavioral data being predicted by needs according to the forecast model pre-build and history row
Compare in corresponding dimension for data, obtain the analysis result of behavioral data.
Concrete, it was predicted that model pre-builds, it was predicted that model dimension can include the product object scale of construction, time span
Degree, the supporting rate of product object, neutral rate and opposition rate, important historical behavior, product object attention rate, product pair
As sales value etc..The analysis mode of forecast model be by needs prediction behavioral data with historical behavior data corresponding
Dimension contrast, the dimension of above-mentioned correspondence refer to need prediction behavioral data with historical behavior data must be
Same dimension, the behavioral data dimension needing prediction is the default product object scale of construction, the pin of default product object
Selling value, corresponding historical behavior data dimension is the historical product object scale of construction, the sales value of historical product object equally.
The dimension of this forecast model and alignments can store in the server.By the data prestored and prediction
Model, by modeling analysis, can be predicted the outcome fast and accurately, be solved behavioral data in prior art
Analysis result takes time and effort, the problem that behavior is difficult to before implementing predict result of implementation.
Needing exist for explanation, the analysis result of behavioral data can be default product object sales volume value, it is possible to
It it is the supporting rate etc. of default product object.Such as, the analysis result of behavioral data can be default product object pin
Amount increase by 20% or the product object supporting rate preset increase by 10%.
Optionally, this device analysis module 44 may include that
Pretreatment module 441, the behavioral data being used for being predicted by needs according to the forecast model pre-build is with many
Individual historical behavior data are compared in corresponding dimension respectively, obtain corresponding with needing the behavioral data being predicted
Multiple Preprocessing results；
Concrete, get N group historical behavior data, it would be desirable to the behavioral data being predicted is gone through with each group respectively
History behavioral data is compared in corresponding dimension, obtains the Preprocessing result of N number of behavior data.
Denoising module 442, for carrying out denoising to obtaining multiple Preprocessing result；
Concrete, there is noise data in the Preprocessing result of the N number of behavioral data obtained, noise data is by shadow
Ring behavioral data precision of analysis, in order to obtain the analysis result of behavioral data more accurately, need pre-place
Reason analysis result carries out denoising.Such as: the N number of Preprocessing result obtained meets normal distribution, and normal state is divided
Cloth rear and front end may be considered noise data, affects the precision of analysis of behavioral data, therefore can remove
The data of normal distribution rear and front end.
First computing module 443, for calculating that the multiple Preprocessing results after denoising are averaged,
Obtain the analysis result of behavioral data.
Concrete, after removing noise data, remaining all Preprocessing results are the analysis results to behavioral data
There are the data of actual reference significance, calculate the analysis result of behavioral data according to remaining all Preprocessing results.
Optionally, this device pretreatment module 441 may include that
Second acquisition module 4411, for obtaining historical behavior result Y that any one historical behavior data is corresponding；
Concrete, historical behavior result can be economically, and the sales volume such as default product object changes, it is possible to be
Preset the supporting rate of product object, the attention rate etc. of default product object, such as historical behavior result is the increase in
The sales volume of 40%.
Concrete, historical behavior result Y can calculate in the following way, such as when historical behavior result Y is for selling
During increased percentage,
Such as when historical behavior result Y is share price amount of increase,
Above-mentioned N can preset as required.
Second computing module 4412, for being calculated the pre-place of the behavioral data that needs are predicted by equation below
Reason analysis result X:
Wherein, i is the number of dimension, is natural number, xiFor the dimensional information in i-th dimension, aiTie up for i-th
The weighted value of the dimensional information on degree, yiFor the history dimensional information in i-th dimension, biFor going through in i-th dimension
The weighted value of history dimensional information.
Concrete, the analysis result of behavioral data is based on historical behavior result and calculates, such as historical behavior result
Being the ratio of sales volume rising, the analysis result of the behavioral data that correspondence draws also is the ratio that sales volume rises.Wherein, ai and
biFor characterizing the default value set according to the behavioral data preset with historical behavior data dependence.I is natural number,
The behavioral data representing default has multiple dimensional information that multiple dimension is corresponding with historical behavior data.
Herein it should be noted that above-mentioned calculated Preprocessing result X can be with multiple historical behavior numbers
Based on first historical behavior data according to, calculated Preprocessing result, and with multiple historical behaviors
Based on other historical behavior data in data, calculate the calculating that the Preprocessing result of behavior data is used
Formula is identical with the above-mentioned formula for being calculated X.
Optionally, when company A needs to predict certain market behavior, the dimension of behavior and the dimension of correspondence will be implemented
Data include: time span: 1 month；The product object scale of construction preset: 500,000,000 market values；The product object support preset
Rate: 40%；The product object neutrality rate preset: 50%；The product object opposition rate preset: 10%；The product preset
Object supporting rate variation tendency: increase by 20%；The product object sales value preset: 1,500,000,000；Behavioral agent historical act
Importance: 50；Behavioral agent historical act attention rate: 380,000；Behavioral agent historical act coverage: 710,000；
Behavioral agent historical act supporting rate variation tendency: increase by 50%.The dimension of B houses market behavior and the dimension of correspondence
Data include: time span: 0.5 month；The product object scale of construction preset: 1,500,000,000 market values；The product object preset
Supporting rate: 30%；The product object neutrality rate preset: 50%；The product object opposition rate preset: 20%；Preset
Product object supporting rate variation tendency: increase by 10%；The product object sales value preset: 4,500,000,000；Behavioral agent history
Movable importance: 32；Behavioral agent historical act attention rate: 670,000；Behavioral agent historical act coverage: 107
Ten thousand；Behavioral agent historical act supporting rate variation tendency: increase by 35%；Historical behavior result: sales value increases by 40%.
Can be calculated by above-mentioned dimension data:
X=-0.74
Optionally, this device can also include: modified module, for receiving the amendment information of outside input, revises
Forecast model.
Concrete, it was predicted that model can be modified according to arranging of external information.Such as, for default product
Object is different, according to the significance level of dimension, for the weighted value that different dimension set is different, to obtain more accurately
The analysis result of behavioral data.
Optionally, this device can also include: data memory module 431, for by corresponding for historical behavior data many
Individual history dimensional information stores in multiple data base, wherein, is carried out by same behavior parameter between each data base
Association.
Concrete, persistently the dimensional information of the different behavior of storage, accumulates substantial amounts of historical data, and dimensional information is pressed
Classifying according to dimension, the dimensional information belonging to identical dimensional is stored in same data base, and multiple dimension correspondences are multiple
Data base.By said method, substantial amounts of historical behavior information is stored in data base, can inquire about easily and go through
History behavioral data, improves the efficiency obtaining behavioral data analysis result.Behavioral parameters refers to concrete behavior data, example
As can be that dimensional information holds certain theme charitable activity, it is also possible to be to hold certain theme marathon grand prix.
Needing exist for explanation, above-mentioned dimension can be adjusted, and increases dimension, can make the dimension of analysis more
Comprehensively, the conclusion type that model can draw also will increase.Meanwhile, dimension can also be modified, be deleted.
Optionally, this device can also include:
Second acquisition module 4321, for gathering new history dimensional information；
Concrete, it being continuously increased new history dimensional information, the analysis result that can make behavioral data is more accurate.
Second data memory module 4322, for storing new history dimensional information in new data base；
Concrete, it is stored in same data base with the dimensional information of dimension, sets up new number for new dimensional information
According to storehouse.
Relating module 4323, is closed with any one or more data bases by same behavior parameter for new data base
Connection.
Concrete, new data base is associated with original multiple data bases, can be to be associated with one of them,
Can also be to be associated with plurality of data base.
It should be noted that for aforesaid each method embodiment, in order to be briefly described, therefore it is all expressed as one it be
The combination of actions of row, but those skilled in the art should know, the present invention not limiting by described sequence of movement
System, because according to the present invention, some step can use other orders or carry out simultaneously.Secondly, art technology
Personnel also should know, embodiment described in this description belongs to preferred embodiment, involved action and module
Not necessarily necessary to the present invention.
In the above-described embodiments, the description to each embodiment all emphasizes particularly on different fields, and does not has the portion described in detail in certain embodiment
Point, may refer to the associated description of other embodiments.
In several embodiments provided herein, it should be understood that disclosed device, can be by other side
Formula realizes.Such as, device embodiment described above is only schematically, the division of the most described unit, only
Being only a kind of logic function to divide, actual can have other dividing mode when realizing, and the most multiple unit or assembly can
To combine or to be desirably integrated into another system, or some features can be ignored, or does not performs.Another point, is shown
The coupling each other shown or discuss or direct-coupling or communication connection can be by some interfaces, device or unit
INDIRECT COUPLING or communication connection, can be being electrical or other form.
The described unit illustrated as separating component can be or may not be physically separate, shows as unit
The parts shown can be or may not be physical location, i.e. may be located at a place, or can also be distributed to
On multiple NEs.Some or all of unit therein can be selected according to the actual needs to realize the present embodiment
The purpose of scheme.
It addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is also possible to
It is that unit is individually physically present, it is also possible to two or more unit are integrated in a unit.Above-mentioned integrated
Unit both can realize to use the form of hardware, it would however also be possible to employ the form of SFU software functional unit realizes.
If described integrated unit realizes and as independent production marketing or use using the form of SFU software functional unit
Time, can be stored in a computer read/write memory medium.Based on such understanding, technical scheme
Completely or partially can producing with software of the part that the most in other words prior art contributed or this technical scheme
The form of product embodies, and this computer software product is stored in a storage medium, including some instructions in order to make
Obtain a computer equipment (can be personal computer, mobile terminal, server or the network equipment etc.) and perform this
All or part of step of method described in each embodiment bright.And aforesaid storage medium includes: USB flash disk, read-only storage
Device (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory),
The various media that can store program code such as portable hard drive, magnetic disc or CD.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.All within the spirit and principles in the present invention, made
Any modification, equivalent substitution and improvement etc., should be included within the scope of the present invention.
Claims (12)
1. the analysis method of a behavioral data, it is characterised in that including:
Gathering the behavioral data needing to be predicted, wherein, described behavioral data includes: at least one dimension
Dimensional information；
Obtain the history that historical behavior data and described historical behavior data are corresponding at least one dimension described
Dimensional information；
The behavioral data described needs being predicted according to the forecast model pre-build and described historical behavior number
Compare according in corresponding dimension, obtain the analysis result of described behavioral data.
Method the most according to claim 1, it is characterised in that according to described forecast model by described need to carry out pre-
The behavioral data surveyed is compared in corresponding dimension with described historical behavior data, obtains described behavioral data
Analysis result include:
The behavioral data described needs being predicted according to the forecast model pre-build and multiple described history row
Compare respectively in corresponding dimension for data, obtain corresponding with the behavioral data that described needs are predicted
Multiple Preprocessing results；
Denoising is carried out to obtaining the plurality of Preprocessing result；
The plurality of Preprocessing result after denoising is averaged calculating, obtain described behavior
The analysis result of data.
Method the most according to claim 2, it is characterised in that according to the forecast model pre-build by described needs
The behavioral data being predicted is compared in corresponding dimension respectively with multiple described historical behavior data,
Step to the multiple Preprocessing results corresponding with the behavioral data that described needs are predicted includes:
Obtain historical behavior result Y that any one historical behavior data is corresponding；
Preprocessing result X of the behavioral data that described needs are predicted it is calculated by equation below:
Wherein, i is the number of described dimension, is natural number, xiFor the dimensional information in i-th dimension, aiFor
The weighted value of the dimensional information in described i-th dimension, yiFor the history dimensional information in described i-th dimension,
biWeighted value for the history dimensional information in described i-th dimension.
Method the most according to claim 1, it is characterised in that according to described forecast model by described need to carry out pre-
The behavioral data surveyed is compared in corresponding dimension with described historical behavior data, obtains described behavioral data
Analysis result before, described method also includes:
Receive the amendment information of outside input, revise described forecast model.
Method the most according to claim 1, it is characterised in that obtain described historical behavior data and described in go through
Before the history dimensional information that history behavioral data is corresponding at least one dimension described, described method also includes:
Multiple history dimensional information corresponding for described historical behavior data are stored in multiple data base, wherein,
It is associated by same behavior parameter between each data base.
Method the most according to claim 5, it is characterised in that by multiple history corresponding for described historical behavior data
After dimensional information stores in multiple data base, described method also includes:
Gather new history dimensional information；
Described new history dimensional information is stored in new data base；
Described new data base is by described same behavior parameter and any one or more database associations.
7. the analytical equipment of a behavioral data, it is characterised in that including:
Acquisition module, for gathering the behavioral data needing to be predicted, wherein, described behavioral data includes:
Dimensional information at least one dimension；
First acquisition module, is used for obtaining historical behavior data and described historical behavior data described at least one
History dimensional information corresponding in individual dimension；
Analyze module, for the behavioral data that described needs is predicted according to the forecast model that pre-builds with
Described historical behavior data are compared in corresponding dimension, obtain the analysis result of described behavioral data.
Device the most according to claim 7, it is characterised in that described analysis module includes:
Pretreatment module, for the behavioral data being predicted by described needs according to the forecast model pre-build
Compare respectively in corresponding dimension with multiple described historical behavior data, obtain and need to be predicted
Multiple Preprocessing results that behavioral data is corresponding；
Denoising module, for carrying out denoising to obtaining the plurality of Preprocessing result；
First computing module, for averaging to the plurality of Preprocessing result after denoising
Calculate, obtain the analysis result of described behavioral data.
Device the most according to claim 8, it is characterised in that described pretreatment module includes:
Second acquisition module, for obtaining described historical behavior result Y that any one historical behavior data is corresponding；
Second computing module, for being calculated the pre-place of the behavioral data that needs are predicted by equation below
Reason analysis result X:
Wherein, i is the number of described dimension, is for natural number, xiFor the dimensional information in i-th dimension, ai
For the weighted value of the dimensional information in described i-th dimension, yiFor the history dimensional information in described i-th dimension,
biWeighted value for the history dimensional information in described i-th dimension.
Device the most according to claim 7, it is characterised in that described device also includes:
Modified module, for receiving the amendment information of outside input, revises described forecast model.
11. devices according to claim 7, it is characterised in that described device also includes:
Data memory module, for storing many by multiple history dimensional information corresponding for described historical behavior data
In individual data base, wherein, it is associated by same behavior parameter between each data base.
12. devices according to claim 11, it is characterised in that described device also includes:
Second acquisition module, for gathering new history dimensional information；
Second data memory module, for storing described new history dimensional information in new data base；
Relating module, for described new data base by described same behavior parameter and any one or more numbers
Associate according to storehouse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510117145.4A CN106033581A (en) | 2015-03-17 | 2015-03-17 | Analysis method of behavior data and apparatus thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510117145.4A CN106033581A (en) | 2015-03-17 | 2015-03-17 | Analysis method of behavior data and apparatus thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106033581A true CN106033581A (en) | 2016-10-19 |
Family
ID=57151196
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510117145.4A Pending CN106033581A (en) | 2015-03-17 | 2015-03-17 | Analysis method of behavior data and apparatus thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106033581A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409875A (en) * | 2018-09-25 | 2019-03-01 | 阿里巴巴集团控股有限公司 | A kind of bill method of calibration, device and electronic equipment |
-
2015
- 2015-03-17 CN CN201510117145.4A patent/CN106033581A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409875A (en) * | 2018-09-25 | 2019-03-01 | 阿里巴巴集团控股有限公司 | A kind of bill method of calibration, device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106485562B (en) | Commodity information recommendation method and system based on user historical behaviors | |
| CN108205766A (en) | Information-pushing method, apparatus and system | |
| CN109325691A (en) | Abnormal behaviour analysis method, electronic equipment and computer program product | |
| CN109241669A (en) | A kind of method for automatic modeling, device and its storage medium | |
| CN109583468A (en) | Training sample acquisition methods, sample predictions method and corresponding intrument | |
| CN102541920A (en) | Method and device for improving accuracy degree by collaborative filtering jointly based on user and item | |
| Limbong et al. | Implementation of Multi Factor Evaluation Process (MFEP) in Assessment of Employee Performance Achievement | |
| CN110147389A (en) | Account number treating method and apparatus, storage medium and electronic device | |
| CN108846695A (en) | The prediction technique and device of terminal replacement cycle | |
| CN104346698A (en) | Catering member big data analysis and checking system based on cloud computing and data mining | |
| CN111738331A (en) | User classification method and device, computer-readable storage medium and electronic device | |
| CN106407305A (en) | Data mining system and method | |
| CN111369344A (en) | Method and device for dynamically generating early warning rule | |
| CN108304975A (en) | A kind of data prediction system and method | |
| CN106528774A (en) | Method and apparatus for predicting distribution network project management trend | |
| CN113569162A (en) | Data processing method, device, equipment and storage medium | |
| CN106033581A (en) | Analysis method of behavior data and apparatus thereof | |
| Shumway | Allocation of scarce resources to agricultural research: Review of methodology | |
| KR101632537B1 (en) | Technical ripple effect analysis method | |
| Valença et al. | Construction of a software measurement tool using systematic literature review | |
| Tudor et al. | Data mining algorithms and techniques research in CRM systems | |
| CN114661750A (en) | Feature derivation method and device, nonvolatile storage medium and electronic equipment | |
| Li et al. | iMiner: mining inventory data for intelligent management | |
| Rodpysh | Model to predict the behavior of customers churn at the industry | |
| Töhönen et al. | Towards systemic evaluation of the business value of IT |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 100083 No. 401, 4th Floor, Haitai Building, 229 North Fourth Ring Road, Haidian District, Beijing Applicant after: Beijing Guoshuang Technology Co.,Ltd. Address before: 100086 Cuigong Hotel, 76 Zhichun Road, Shuangyushu District, Haidian District, Beijing Applicant before: Beijing Guoshuang Technology Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161019 |