CN110033323A - Data analysing method, device, electronic equipment and readable storage medium storing program for executing - Google Patents
Data analysing method, device, electronic equipment and readable storage medium storing program for executing Download PDFInfo
- Publication number
- CN110033323A CN110033323A CN201910290693.5A CN201910290693A CN110033323A CN 110033323 A CN110033323 A CN 110033323A CN 201910290693 A CN201910290693 A CN 201910290693A CN 110033323 A CN110033323 A CN 110033323A
- Authority
- CN
- China
- Prior art keywords
- grouping
- model
- test result
- objects
- cryptographic hash
- 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
Abstract
The embodiment of the present disclosure discloses a kind of data analysing method, device, electronic equipment and readable storage medium storing program for executing, the data analysing method includes calculating the corresponding cryptographic Hash of object according to the respective identification information of multiple objects, and the object includes user or user equipment；The multiple object is grouped according to the cryptographic Hash；Each object is grouped and applies different models；Obtain the test result obtained to each object grouping using corresponding model.The technical solution is by carrying out small flux experiment on line for new model, it avoids and new model is directly applied into generated greater risk on line, simultaneously, multiple objects are carried out automation grouping and are grouped each object according to cryptographic Hash to apply different models, the method of salary distribution is more random and objective, reduce the influence that subjective factor evaluates modelling effect, and improves the testing efficiency of model.
Description
Technical field
This disclosure relates to computer application technology, and in particular to a kind of data analysing method, device, electronic equipment and
Readable storage medium storing program for executing.
Background technique
In the epoch of internet fast development, it is corresponding to realize that the network operator of Internet enterprises can develop various models
Function, such as: the network operator for taking out industry can develop various models for taking out intelligent scheduling, sequence or recommendation of industry etc.
Scene.During proposing the present invention, inventors have found that when the network operator of Internet enterprises develops a new model,
It is often directly applied on line without too small flow test, by virtue of experience judges the application effect of the new model.
The above method has the following deficiencies: firstly, new model directly applies on line, and risk is larger, especially when this is new
When model is the model of a failure, lost part client can be directly resulted in；Secondly, by virtue of experience judging answering for new model
It is more subjective with effect, it is easy to appear mistake and inefficiency.Therefore, new model is subjected to small flow test on line, and to this
New model carries out effect analysis and is a problem to be solved.
Summary of the invention
In order to solve the problems in the relevant technologies, the embodiment of the present disclosure provides a kind of data analysing method, device, electronics and sets
Standby and readable storage medium storing program for executing.
In a first aspect, providing a kind of data analysing method in the embodiment of the present disclosure.
Specifically, the data analysing method, comprising:
Calculate the corresponding cryptographic Hash of object according to the respective identification information of multiple objects, the object include user or
User equipment；
The multiple object is grouped according to the cryptographic Hash；
Each object is grouped and applies different models；
Obtain the test result obtained to each object grouping using corresponding model.
With reference to first aspect, the disclosure is in the first implementation of first aspect, and the acquisition is to described each right
As being grouped the test result obtained using corresponding model, comprising:
Bury a little for the object in each object grouping；
After to the object application corresponding model in each object grouping, obtain in each object grouping
Object buries a log；
The determining test result obtained to each object grouping using corresponding model of a log is buried according to described.
With reference to first aspect, the disclosure is in second of implementation of first aspect, and the method also includes to described
Test result is visualized.
Second of implementation with reference to first aspect, the disclosure is in the third implementation of first aspect, to institute
It states test result to be visualized, including the test result is carried out visually by any one or more following mode
Change and show: table, curve graph, column diagram.
With reference to first aspect, the disclosure is described to be incited somebody to action according to the cryptographic Hash in the 4th kind of implementation of first aspect
The multiple object grouping, comprising:
The multiple object is divided into N parts according to the cryptographic Hash, N >=2；
It is divided into the multiple object grouping for described N parts.
The 4th kind of implementation with reference to first aspect, the disclosure are described in the 5th kind of implementation of first aspect
The multiple object is divided into N parts according to the cryptographic Hash, including according to the cryptographic Hash divided by the resulting remainder of N, it will be described
Multiple objects are divided into N parts.
With reference to first aspect, the disclosure is described to answer each object grouping in the 6th kind of implementation of first aspect
New model is applied with different models, including to the grouping of the first object, the second object is grouped and applies old model, first object
The number of objects of grouping is less than the number of objects of second object grouping；
In the case where the test result of the new model is better than the test result of the old model, third object is grouped
Using the new model, the number of objects of the third object grouping is more than the number of objects of first object grouping.
The 6th kind of implementation with reference to first aspect, the disclosure are described in the 7th kind of implementation of first aspect
The grouping of first object includes at least part same object with third object grouping；Or first object grouping with it is described
The grouping of third object does not include same object.
The 6th kind of implementation with reference to first aspect, the disclosure in the 8th kind of implementation of first aspect, according to
The test result of the new model determines the third object grouping compared to the promotion degree of the test result of the old model
Number of objects；
The promotion degree is bigger, and the number of objects of the third object grouping is more.
With reference to first aspect, in the 9th kind of implementation of first aspect, the multiple object is based on following the disclosure
At least one of attribute and belong to the same category: city, the age, gender, occupation, work address, inhabitation address, browsing habit, disappear
Take habit.
Second aspect provides a kind of data analysis set-up in the embodiment of the present disclosure.
Specifically, the data analysis set-up, comprising:
Computing module is configured as calculating the corresponding cryptographic Hash of object according to the respective identification information of multiple objects,
The object includes user or user equipment；
Grouping module is configured as being grouped the multiple object according to the cryptographic Hash；
Application module is configured as being grouped each object and applies different models；
Module is obtained, is configured as obtaining the test result for obtaining each object grouping using corresponding model.
In conjunction with second aspect, the disclosure is in the first implementation of second aspect, and the acquisition is to described each right
As being grouped the test result obtained using corresponding model, comprising:
Bury a little for the object in each object grouping；
After to the object application corresponding model in each object grouping, obtain in each object grouping
Object buries a log；
The determining test result obtained to each object grouping using corresponding model of a log is buried according to described.
In conjunction with second aspect, the disclosure further includes display module, is configured in second of implementation of second aspect
Are as follows:
The test result is visualized.
In conjunction with second of implementation of second aspect, the disclosure is in the third implementation of second aspect, to institute
It states test result to be visualized, including the test result is carried out visually by any one or more following mode
Change and show: table, curve graph, column diagram.
In conjunction with second aspect, the disclosure is described to be incited somebody to action according to the cryptographic Hash in the 4th kind of implementation of second aspect
The multiple object grouping, comprising:
The multiple object is divided into N parts according to the cryptographic Hash, N >=2；
It is divided into the multiple object grouping for described N parts.
In conjunction with the 4th kind of implementation of second aspect, the disclosure is described in the 5th kind of implementation of second aspect
The multiple object is divided into N parts according to the cryptographic Hash, including according to the cryptographic Hash divided by the resulting remainder of N, it will be described
Multiple objects are divided into N parts.
In conjunction with second aspect, the disclosure is described to answer each object grouping in the 6th kind of implementation of second aspect
New model is applied with different models, including to the grouping of the first object, the second object is grouped and applies old model, first object
The number of objects of grouping is less than the number of objects of second object grouping；
In the case where the test result of the new model is better than the test result of the old model, third object is grouped
Using the new model, the number of objects of the third object grouping is more than the number of objects of first object grouping.
In conjunction with the 6th kind of implementation of second aspect, the disclosure is described in the 7th kind of implementation of second aspect
The grouping of first object includes at least part same object with third object grouping；Or first object grouping with it is described
The grouping of third object does not include same object.
In conjunction with the 6th kind of implementation of second aspect, the disclosure in the 8th kind of implementation of second aspect, according to
The test result of the new model determines the third object grouping compared to the promotion degree of the test result of the old model
Number of objects；
The promotion degree is bigger, and the number of objects of the third object grouping is more.
In conjunction with second aspect, in the 9th kind of implementation of second aspect, the multiple object is based on following the disclosure
At least one of attribute and belong to the same category: city, the age, gender, occupation, work address, inhabitation address, browsing habit, disappear
Take habit.
The third aspect, the embodiment of the present disclosure provide a kind of electronic equipment, including memory and processor, wherein described
Memory is for storing one or more computer instruction, wherein one or more computer instruction is by the processor
It executes to realize following methods step:
Calculate the corresponding cryptographic Hash of object according to the respective identification information of multiple objects, the object include user or
User equipment；
The multiple object is grouped according to the cryptographic Hash；
Each object is grouped and applies different models；
Obtain the test result obtained to each object grouping using corresponding model.
In conjunction with the third aspect, the disclosure is in the first implementation of the third aspect, and the acquisition is to described each right
As being grouped the test result obtained using corresponding model, comprising:
Bury a little for the object in each object grouping；
After to the object application corresponding model in each object grouping, obtain in each object grouping
Object buries a log；
The determining test result obtained to each object grouping using corresponding model of a log is buried according to described.
In conjunction with the third aspect, the disclosure is in second of implementation of the third aspect, one or more computer
Instruction is also executed by the processor to realize following methods step:
The test result is visualized.
In conjunction with second of implementation of the third aspect, the disclosure is in the third implementation of the third aspect, to institute
It states test result to be visualized, including the test result is carried out visually by any one or more following mode
Change and show: table, curve graph, column diagram.
In conjunction with the third aspect, the disclosure is described to be incited somebody to action according to the cryptographic Hash in the 4th kind of implementation of the third aspect
The multiple object grouping, comprising:
The multiple object is divided into N parts according to the cryptographic Hash, N >=2；
It is divided into the multiple object grouping for described N parts.
In conjunction with the 4th kind of implementation of the third aspect, the disclosure is described in the 5th kind of implementation of the third aspect
The multiple object is divided into N parts according to the cryptographic Hash, including according to the cryptographic Hash divided by the resulting remainder of N, it will be described
Multiple objects are divided into N parts.
In conjunction with the third aspect, the disclosure is described to answer each object grouping in the 6th kind of implementation of the third aspect
New model is applied with different models, including to the grouping of the first object, the second object is grouped and applies old model, first object
The number of objects of grouping is less than the number of objects of second object grouping；
In the case where the test result of the new model is better than the test result of the old model, third object is grouped
Using the new model, the number of objects of the third object grouping is more than the number of objects of first object grouping.
In conjunction with the 6th kind of implementation of the third aspect, the disclosure is described in the 7th kind of implementation of the third aspect
The grouping of first object includes at least part same object with third object grouping；Or first object grouping with it is described
The grouping of third object does not include same object.
In conjunction with the 6th kind of implementation of the third aspect, the disclosure in the 8th kind of implementation of the third aspect, according to
The test result of the new model determines the third object grouping compared to the promotion degree of the test result of the old model
Number of objects；
The promotion degree is bigger, and the number of objects of the third object grouping is more.
In conjunction with the third aspect, in the 9th kind of implementation of the third aspect, the multiple object is based on following the disclosure
At least one of attribute and belong to the same category: city, the age, gender, occupation, work address, inhabitation address, browsing habit, disappear
Take habit.
Fourth aspect provides a kind of readable storage medium storing program for executing in the embodiment of the present disclosure, is stored thereon with computer instruction, should
Realize the first implementation such as first aspect, first aspect to the 9th kind of realization side when computer instruction is executed by processor
The described in any item methods of formula.
The technical solution that the embodiment of the present disclosure provides can include the following benefits:
It is avoided according to the technical solution that the embodiment of the present disclosure provides by the way that new model is carried out small flux experiment on line
New model is directly applied into generated greater risk on line, meanwhile, multiple objects are carried out by automation point according to cryptographic Hash
Group is simultaneously grouped using different models each object, and the method for salary distribution is more random and objective, reduces subjective factor and imitates to model
The influence of fruit evaluation, and improve the testing efficiency of model.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
The disclosure can be limited.
Detailed description of the invention
In conjunction with attached drawing, by the detailed description of following non-limiting embodiment, the other feature of the disclosure, purpose and excellent
Point will be apparent.In the accompanying drawings:
Fig. 1 shows the flow chart of data analysing method according to an embodiment of the present disclosure；
Fig. 2 shows the test knots according to an embodiment of the present disclosure for obtaining and obtaining to each object grouping using corresponding model
The flow chart of fruit；
Fig. 3 A shows the table schematic diagram that test result according to an embodiment of the present disclosure visualizes；
Fig. 3 B shows the curve graph schematic diagram that test result according to an embodiment of the present disclosure visualizes；
Fig. 4 is shown in accordance with an embodiment of the present disclosure to the flow chart of multiple objects grouping；
Fig. 5 shows the structural block diagram of data analysis set-up according to an embodiment of the present disclosure；
Fig. 6 shows the structural block diagram of electronic equipment according to an embodiment of the present disclosure；
Fig. 7, which is shown, to be suitable for being used to realizing that the structure of the computer system of the data analysing method according to the embodiment of the present disclosure is shown
It is intended to.
Specific embodiment
Hereinafter, the exemplary embodiment of the disclosure will be described in detail with reference to the attached drawings, so that those skilled in the art can hold
It changes places and realizes them.In addition, for the sake of clarity, the part unrelated with description exemplary embodiment is omitted in the accompanying drawings.
In the disclosure, it should be appreciated that the term of " comprising " or " having " etc. is intended to refer to disclosed in this specification
Feature, number, step, behavior, the presence of component, part or combinations thereof, and be not intended to exclude other one or more features,
A possibility that number, step, behavior, component, part or combinations thereof exist or are added.
It also should be noted that in the absence of conflict, the feature in embodiment and embodiment in the disclosure
It can be combined with each other.The disclosure is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 shows the flow chart of data analysing method according to an embodiment of the present disclosure.As shown in Figure 1, the data point
Analysis method includes the following steps S101-S104:
In step s101, the corresponding cryptographic Hash of object is calculated according to the respective identification information of multiple objects, it is described
Object includes user or user equipment；
In step s 102, the multiple object is grouped according to the cryptographic Hash；
In step s 103, each object is grouped and applies different models；
In step S104, the test result obtained to each object grouping using corresponding model is obtained.
In accordance with an embodiment of the present disclosure, object includes user or user equipment, and the respective mark of object may include user
Unique identification or unique user equipment identifier.The respective mark of object may include in number, letter, Chinese character or additional character
Any one or any combination thereof is not especially limited this in the disclosure.Can uniquely it be known by the identification information of object
The not object.The corresponding cryptographic Hash of object is calculated according to the respective identification information of multiple objects, such as can be using safety
Hash algorithm or use information-digest algorithm, as long as the Hash that may be implemented to convert the respective mark of object to cryptographic Hash is calculated
Method, the disclosure are not especially limited the hash algorithm for calculating cryptographic Hash.
After model to be tested has been determined, these different models to be tested can be applied to multiple objects to test
The effect of model.In accordance with an embodiment of the present disclosure, multiple objects can be grouped according to the number of model to be tested.For example, false
If the quantity of model to be tested is 5, then the multiple object can be divided into 5 groups, to test this 5 models respectively.
According to test needs, the split ratio of each model can be set, i.e., using the ratio of the number of objects of each model.
For example, lesser split ratio can first be arranged for new model, its split ratio is adjusted according to test result.For test result
Preferable new model can increase effect of its split ratio with testing needle to more objects.
According to the split ratio of each model, the quantity of the object in each object grouping can be correspondingly set.For example, false
If model to be tested is A1 model and A2 model, and A1 model is identical with the split ratio of A2 model, then can be according to cryptographic Hash
The multiple object is divided into a1 group and a2 group, to multiple object application A1 models in a1 group, to multiple in a2 group
Object application A2 model, can also be to multiple object application A1 models in a2 group, to multiple object application A2 moulds in a1 group
Type.Or, it is assumed that model to be tested is model M 1 and model M 2, and the split ratio of model M 1 and model M 2 is 9:1, then can be with
The multiple object is divided into m1 group and m2 group according to split ratio according to cryptographic Hash, wherein object accounting is that object is total in m1 group
Object accounting is the 10% of object sum in several 90%, m2 groups, can be to multiple object application model M1 in m1 group, to m2
Multiple object application model M2 in group.
After different models are applied to corresponding object grouping, the test result of available difference model.According to this public affairs
The embodiment opened, for different models, the exposure of object is (for example, commodity or businessman are browsed or opened up on a user device by users
Show), the operation information of the behaviors such as clicking or place an order can be preserved in the form of burying a journal file, it is different to pass through analysis
Model buries a journal file, so that it may obtain the test result of corresponding model.The embodiment of the present disclosure is to model test results
Evaluation index is not specifically limited, for example, take out industry, evaluation index may include exposure number, click number, it is lower one
Number, odd number, clicking rate, click conversion ratio, transformation in planta rate, rate of return on investment, average visitor at conversion ratio, exposure conversion ratio in shop
It is one or more in monovalent and total flowing water, or can the weight different to different evaluation setup measures as desired, such as
Lower list number and the weight of odd number are higher than exposure number and click the weight of number.
Fig. 2 shows the test knots according to an embodiment of the present disclosure for obtaining and obtaining to each object grouping using corresponding model
The flow chart of fruit.Each object grouping is included the following steps using the test result that corresponding model obtains as shown in Fig. 2, obtaining
S201-S203:
In step s 201, bury a little for the object in each object grouping；
In step S202, after to the object application corresponding model in each object grouping, obtain described each
Object in the grouping of a object buries a log；
In step S203, bury what log determination obtained each object grouping using corresponding model according to described
Test result.
In accordance with an embodiment of the present disclosure, bury a little for each object in the grouping of different objects, the disclosure pair
The implementation buried a little is not especially limited, for example, the operation information that object can be a little collected by Javascript method is buried,
Operation information such as can be the exposure of object, click or place an order at the behaviors, when object, which executes, to be operated, the corresponding behaviour of server record
Make information, these operation informations preserve in the form of burying a log.Recommend the page for example, object is opened and click some
Recommended businessman, but do not place an order, then the exposure of burying in log record object and click behaviour of the server in the object
Make.
Object after object application corresponding model in being grouped to each object, in available each object grouping
Bury a log, bury log by counting and determine the test result obtained to each object grouping using corresponding model.For example,
Exposure number by burying the available model E 1 of a log is 12509, click number is 9007, lower single number is 4631, single
Number is 4957, so that calculating conversion ratio in clicking rate be 72% (click number/exposure number) and shop is 51.42% (to place an order
Number/click number) etc. evaluation indexes；The exposure number for obtaining model E 2 is 29026, click number is 20691, lower single number
For 10896, odd number 11639, thus calculate clicking rate be 71.28% and shop in conversion ratio be the evaluation indexes such as 52.66%,
Model E 1 and the test effect of E2 can be known by comparing above-mentioned evaluation index.
In accordance with an embodiment of the present disclosure, the data analysing method further includes carrying out visualization exhibition to the test result
Show.
In order to more intuitively show the test result of different models, such as show pair between the test result of different models
It, can be more straight by visualizing than as a result, in embodiment of the disclosure, can be visualized to test result
Contrast effect between the variation tendency and different models of the understanding model of sight.
In accordance with an embodiment of the present disclosure, the test result is visualized, including by it is following any one
Or various ways visualize the test result: table, curve graph, column diagram.
Different visual presentation modes has the effect of different, in the embodiments of the present disclosure, can choose one of which
Mode visualizes test result, also can choose various ways and visualizes to test result, such as:
Table will be seen that the specific value of evaluation index, curve graph will be seen that evaluation index with the variation tendency of time.
The example visualized according to the test result of the embodiment of the present disclosure is illustrated below with reference to Fig. 3 A and Fig. 3 B
Property process.Assuming that have B1 model and B2 model, and it is 70% that the split ratio of B1 model, which is the split ratio of 30%, B2 model,.
Fig. 3 A shows the table schematic diagram that test result according to an embodiment of the present disclosure visualizes.Such as Fig. 3 A institute
Show, illustrate the test result of B1 model and B2 model in tabular form, wherein exposure number, click number, lower single number and
Odd number can be obtained by burying a log mode, other evaluation indexes can be obtained by calculating, for example, clicking rate=click people
Conversion ratio=lower single number/click number, exposure conversion ratio=odd number/exposure number, click conversion in number/exposure number, shop
Rate=one/click number, transformation in planta rate=lower single number/exposure number, by table can intuitively understand B1 model and
The test result of B2 model.Simultaneously by calculating correlation data, i.e. (B1 model-B2 model)/B2 model, it will be appreciated that B1 model
With the test comparison result of B2 model.
Fig. 3 B shows the curve graph schematic diagram that test result according to an embodiment of the present disclosure visualizes.Such as Fig. 3 B institute
Showing, abscissa is time shaft, and specifically from 2019-01-01 to 2019-01-05, ordinate is the evaluation index of test result,
Such as odd number, city are Beijing, commercial circle selects whole commercial circles, the test of B1 model and B2 model can be intuitively found out from Fig. 3 B
As a result and between comparing result.
In accordance with an embodiment of the present disclosure, column diagram can be used also to visualize to test result, such as with
The column of different colours represents different models, and the height of column indicates the numerical value of test result.
Fig. 4 is shown in accordance with an embodiment of the present disclosure to the flow chart of multiple objects grouping.As shown in figure 4, described according to institute
It states cryptographic Hash and the grouping of the multiple object is included the following steps into S401-S402:
In step S401, the multiple object is divided into N parts according to the cryptographic Hash, N >=2；
In step S402, it is divided into the multiple object for described N parts and is grouped.
It in accordance with an embodiment of the present disclosure, can be first multiple according to cryptographic Hash when being grouped multiple objects according to cryptographic Hash
Object is divided into N parts, then is divided into multiple objects for N parts and is grouped.In accordance with an embodiment of the present disclosure, N parts of the number of objects can be with
It is equal, it can also be unequal.First it is divided into N parts according to the multiple objects of cryptographic Hash, then is divided into multiple objects for N parts and is grouped, convenient for dynamic
State adjusts the number of objects of each object grouping.For example, when to increase the number of objects of certain object grouping, it can be to the object point
Group increases one or more parts, and when to reduce the number of objects of certain object grouping, it is a or more can be grouped reduction from the object
Part.It is grouped by this method, it is not necessary to be grouped again in the number of objects of each regulating object grouping, reduce operation
And processing load.
The embodiment of the present disclosure to the specific value of N without limitation.For example, it is assumed that N is 100, then it can be first according to cryptographic Hash
The multiple object is divided into 100 parts at random, then is divided into multiple objects for 100 parts and is grouped.For example, it is assumed that point of model E 1
Stream is 80%, and model E 2 splits into 20%, then the multiple object first can be divided into 100 at random according to cryptographic Hash
Part, to 80 parts of application model E1 therein, to 20 parts of application model E2 therein.For example, it is assumed that N is 10, then it can first basis
Multiple objects are divided into 10 parts at random by cryptographic Hash, then are divided into multiple objects for 10 parts and are grouped.For example, it is assumed that model F1 and mould
The split ratio of type F2 is 7:3, then multiple objects first can be divided into 10 parts at random according to cryptographic Hash, then to 7 parts therein
Application model F1, to 3 parts of application model F2 therein.
In accordance with an embodiment of the present disclosure, described that the multiple object is divided into N parts according to the cryptographic Hash, including according to institute
Cryptographic Hash is stated divided by the resulting remainder of N, the multiple object is divided into N parts.For example, the identical multiple objects of remainder can be made
For portion, for example, it is assumed that cryptographic Hash is respectively as follows: 0,1,2 divided by the resulting remainder of N ... ..., remainder is then 0 by N-2, N-1
Multiple objects are as first part, and multiple objects that remainder is 1 are as second part ... ..., and multiple objects that remainder is N-1 are as the
N parts, so that multiple objects are divided into N parts.
By taking N takes 100 as an example, cryptographic Hash is respectively as follows: 0,1,2 divided by 100 resulting remainders ... ..., and 98,99, then by remainder
For 0 multiple objects as first part, multiple objects that remainder is 1 are as second part ... ..., and multiple objects that remainder is 99 are made
It is the 100th part, multiple objects is divided into 100 parts to realize.
In accordance with an embodiment of the present disclosure, the number of objects in the grouping of at least two objects is different；Or at least two object
Number of objects in grouping is identical.
In accordance with an embodiment of the present disclosure, model to be tested can all be new model, also may include one or more new moulds
Type and one or more old model (for example, once or model currently in use).When model to be measured is all new model, if to it
Split ratio does not have particular/special requirement, and the split ratio that can set all new models is equal, and is carried out multiple objects according to cryptographic Hash
Average packet, at this point, the number of objects in the grouping of at least two objects is identical.For example, it is assumed that model to be tested is 2 new moulds
Multiple objects can be divided into two groups, respectively x1 group and x2 according to cryptographic Hash by type, respectively X1 model and X2 model
Group, i.e. x1 group are identical with the number of objects in x2 group.
Alternatively, when model to be measured includes one or more new models and one or more old models, it can be first by new mould
The flow-rate ratio of type is set below the flow-rate ratio of old model, at this point, the number of objects in the grouping of at least two objects is different.Example
Such as, it is assumed that model to be tested is 2 models, respectively old model P1 and new model P2, and old model P1 and new model P2
Split ratio is 8:2, multiple objects can be divided into two groups according to split ratio according to cryptographic Hash, wherein number of objects is in p1 group
Number of objects is the 20% of the total quantity of the multiple object, i.e. p1 group in 80%, the p2 group of the total quantity of the multiple object
It is different with the number of objects in p2 group.
In accordance with an embodiment of the present disclosure, described be grouped to each object applies different models, including is grouped to the first object
Using new model, the second object is grouped and applies old model, the number of objects of the first object grouping is less than described second pair
As the number of objects of grouping.In the case where the test result of the new model is better than the test result of the old model, to the
The new model is applied in the grouping of three objects, and the number of objects of the third object grouping is more than the object of first object grouping
Quantity.
When there is new model online, due to can not also predict the application effect of new model, it generally can first be directed to number of objects
Less the first object grouping is measured using new model, the second packet more to number of objects applies old model, to reduce new mould
Type tests the influence to actual operation.
When the test result of new model is better than the test result of old model, illustrate that the application effect of new model can be better than
Old model can gradually expand the test object quantity of new model, and number of objects is more than to the third object of the first object grouping
Grouping is applied to new model, wherein the number of objects for improving object grouping can be realized by improving split ratio.It below will be with
It is explained for take-away industry, it is assumed that the shunting difference that the grouping of the first object, the grouping of the second object, third object are grouped
It is 10%, 60% and 30%, but average visitor's unit price of new model is 32 yuan, average visitor's unit price of old model is 28 yuan, i.e., newly
The test result of model is better than the test result of old model, at this point it is possible to which the grouping of third object is applied to new model.
If the new model test result for being applied to the grouping of third object is still better than the test result of old model, can be into one
Step improves the number of objects for applying new model, until to whole object application new models.
In accordance with an embodiment of the present disclosure, the first object grouping is identical including at least part as third object grouping
Object；Or the first object grouping and third object grouping do not include same object.
It can be in the grouping of the first object and setting unit same object in the grouping of third object, for example, can be at first pair
As grouping all or part of object on the basis of, increase new object, with obtain third object grouping.In this way, can be to phase
Retest is carried out with object, new model effect is verified, to exclude the influence of accidentalia.
Alternatively, the object of third grouping can be totally different from the object in the grouping of the first object.In this way, not to identical right
As carrying out retest, testing efficiency can be improved and promote test result for the universality of object.Moreover, by increasing quilt
The number of objects of test also can reduce the influence of accidentalia.
In accordance with an embodiment of the present disclosure, the test result according to the test result of the new model compared to the old model
Promotion degree, determine the number of objects of third object grouping, the promotion degree is bigger, the third object grouping
Number of objects is more.
When testing new model, it can be guiding with test result, be stepped up the object applied to new model
Quantity, it can the promotion degree according to the test result of new model compared to the test result of old model, to determine third pair
As the number of objects of grouping, specifically, when the test result of new model is got over compared to the promotion degree of the test result of old model
Greatly, greater number of object just is distributed for the grouping of third object.Such as: when new model test result compared to old model survey
Test result improves 10%, the number of objects that third object is grouped can be increased by 10% compared to the grouping of the first object, when new
The test result of model improves 20% compared to the test result of old model, can be by number of objects phase that third object is grouped
Than increasing by 20% in the grouping of the first object.
In accordance with an embodiment of the present disclosure, the multiple object is based at least one of following attribute and belongs to the same category: city
City, age, gender, occupation, work address, inhabitation address, browsing habit, consumption habit.
Model measurement is carried out for same class object, test result can be made more accurate and there is specific aim.For example,
Certain model may be especially effective for certain a kind of user, and unobvious for another kind of user's effect.Completely subtract for example, increasing
Dynamics may be compared to user's more attractive to other ages for the user at certain ages；Alternatively, providing certain present
May user to some city compared to user's more attractive to other cities.If not carrying out class area to object
Point, then the modelling effect tested may be because that object properties are different and are diluted.When for same class user progress model
When test, available model is directed to the effect of different object, accomplishes to apply plan because of city Shi Ce or because of object type, raising is determined
The specific aim and accuracy of plan.
Fig. 5 shows the structural block diagram of data analysis set-up 500 according to an embodiment of the present disclosure.Wherein, which can be with
Pass through being implemented in combination with as some or all of of electronic equipment for software, hardware or both.
As shown in figure 5, the data analysis set-up includes computing module 510, grouping module 520, application module 530 and obtains
Modulus block 540.
The computing module 510 is configured as being breathed out accordingly according to multiple objects respective identification information calculating object
Uncommon value, the object includes user or user equipment；
The grouping module 520 is configured as being grouped the multiple object according to the cryptographic Hash；
The application module 530, which is configured as being grouped each object, applies different models；
The module 540 that obtains is configured as obtaining the test knot for obtaining each object grouping using corresponding model
Fruit.
In accordance with an embodiment of the present disclosure, described to obtain the test knot obtained to each object grouping using corresponding model
Fruit, comprising:
Bury a little for the object in each object grouping；
After to the object application corresponding model in each object grouping, obtain in each object grouping
Object buries a log；
The determining test result obtained to each object grouping using corresponding model of a log is buried according to described.
In accordance with an embodiment of the present disclosure, the data analysis set-up includes display module 550, is configured as: to the survey
Test result is visualized.
In accordance with an embodiment of the present disclosure, the test result is visualized, including by it is following any one
Or various ways visualize the test result: table, curve graph, column diagram.
It is in accordance with an embodiment of the present disclosure, described to be grouped the multiple object according to the cryptographic Hash, comprising:
The multiple object is divided into N parts according to the cryptographic Hash, N >=2；
It is divided into the multiple object grouping for described N parts.
In accordance with an embodiment of the present disclosure, described that the multiple object is divided into N parts according to the cryptographic Hash, including according to institute
Cryptographic Hash is stated divided by the resulting remainder of N, the multiple object is divided into N parts.
In accordance with an embodiment of the present disclosure, described be grouped to each object applies different models, including is grouped to the first object
Using new model, the second object is grouped and applies old model, the number of objects of the first object grouping is less than described second pair
As the number of objects of grouping；
In the case where the test result of the new model is better than the test result of the old model, third object is grouped
Using the new model, the number of objects of the third object grouping is more than the number of objects of first object grouping.
In accordance with an embodiment of the present disclosure, the first object grouping is identical including at least part as third object grouping
Object；Or the first object grouping and third object grouping do not include same object.
In accordance with an embodiment of the present disclosure, the test result according to the test result of the new model compared to the old model
Promotion degree, determine the number of objects of third object grouping；
The promotion degree is bigger, and the number of objects of the third object grouping is more.
In accordance with an embodiment of the present disclosure, the multiple object is based at least one of following attribute and belongs to the same category: city
City, age, gender, occupation, work address, inhabitation address, browsing habit, consumption habit.
The disclosure also discloses a kind of electronic equipment, and Fig. 6 shows the structure of electronic equipment according to an embodiment of the present disclosure
Block diagram.
As shown in fig. 6, the electronic equipment 600 includes memory 601 and processor 602；Wherein,
The memory 601 is for storing one or more computer instruction, wherein one or more computer refers to
It enables and being executed by the processor 602 to realize following methods step:
Calculate the corresponding cryptographic Hash of object according to the respective identification information of multiple objects, the object include user or
User equipment；
The multiple object is grouped according to the cryptographic Hash；
Each object is grouped and applies different models；
Obtain the test result obtained to each object grouping using corresponding model.
In accordance with an embodiment of the present disclosure, described to obtain the test knot obtained to each object grouping using corresponding model
Fruit, comprising:
Bury a little for the object in each object grouping；
After to the object application corresponding model in each object grouping, obtain in each object grouping
Object buries a log；
The determining test result obtained to each object grouping using corresponding model of a log is buried according to described.
In accordance with an embodiment of the present disclosure, one or more computer instruction also by the processor execute with realize with
Lower method and step:
The test result is visualized.
In accordance with an embodiment of the present disclosure, the test result is visualized, including by it is following any one
Or various ways visualize the test result: table, curve graph, column diagram.
It is in accordance with an embodiment of the present disclosure, described to be grouped the multiple object according to the cryptographic Hash, comprising:
The multiple object is divided into N parts according to the cryptographic Hash, N >=2；
It is divided into the multiple object grouping for described N parts.
In accordance with an embodiment of the present disclosure, described that the multiple object is divided into N parts according to the cryptographic Hash, including according to institute
Cryptographic Hash is stated divided by the resulting remainder of N, the multiple object is divided into N parts.
In accordance with an embodiment of the present disclosure, described be grouped to each object applies different models, including is grouped to the first object
Using new model, the second object is grouped and applies old model, the number of objects of the first object grouping is less than described second pair
As the number of objects of grouping；
In the case where the test result of the new model is better than the test result of the old model, third object is grouped
Using the new model, the number of objects of the third object grouping is more than the number of objects of first object grouping.
In accordance with an embodiment of the present disclosure, the first object grouping is identical including at least part as third object grouping
Object；Or the first object grouping and third object grouping do not include same object.
In accordance with an embodiment of the present disclosure, the test result according to the test result of the new model compared to the old model
Promotion degree, determine the number of objects of third object grouping；
The promotion degree is bigger, and the number of objects of the third object grouping is more.
In accordance with an embodiment of the present disclosure, the multiple object is based at least one of following attribute and belongs to the same category: city
City, age, gender, occupation, work address, inhabitation address, browsing habit, consumption habit.
Fig. 7, which is shown, to be suitable for being used to realizing that the structure of the computer system of the data analysing method according to the embodiment of the present disclosure is shown
It is intended to.
As shown in fig. 7, computer system 700 includes central processing unit (CPU) 701, it can be read-only according to being stored in
Program in memory (ROM) 702 or be loaded into the program in random access storage device (RAM) 703 from storage section 708 and
Execute the various processing in above-described embodiment.In RAM 703, also it is stored with system 700 and operates required various program sum numbers
According to.CPU 701, ROM 702 and RAM 703 are connected with each other by bus 704.Input/output (I/O) interface 707 also connects
To bus 704.
I/O interface 705 is connected to lower component: the importation 706 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 707 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 708 including hard disk etc.；
And the communications portion 709 of the network interface card including LAN card, modem etc..Communications portion 709 via such as because
The network of spy's net executes communication process.Driver 710 is also connected to I/O interface 705 as needed.Detachable media 711, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 710, in order to read from thereon
Computer program be mounted into storage section 708 as needed.
Particularly, in accordance with an embodiment of the present disclosure, method as described above may be implemented as computer software programs.Example
Such as, embodiment of the disclosure includes a kind of computer program product comprising be tangibly embodied in and its readable medium on meter
Calculation machine program, the computer program include the program code that method is determined for executing above-mentioned object type.In such reality
It applies in example, which can be downloaded and installed from network by communications portion 709, and/or from detachable media
711 are mounted.
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in course diagram or block diagram can generation
A part of one module, section or code of table, a part of the module, section or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in the embodiment of the present disclosure involved unit or module can be realized by way of software, can also be with
It is realized by way of programmable hardware.Described unit or module also can be set in the processor, these units or
The title of module does not constitute the restriction to the unit or module itself under certain conditions.
As on the other hand, the disclosure additionally provides a kind of computer readable storage medium, the computer-readable storage medium
Matter can be computer readable storage medium included in electronic equipment or computer system in above-described embodiment；It can also be with
It is individualism, without the computer readable storage medium in supplying equipment.Computer-readable recording medium storage have one or
More than one program of person, described program is used to execute by one or more than one processor is described in disclosed method.
Above description is only the preferred embodiment of the disclosure and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the disclosure, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from the inventive concept, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed in the disclosure
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (10)
1. a kind of data analysing method characterized by comprising
The corresponding cryptographic Hash of object is calculated according to the respective identification information of multiple objects, the object includes user or user
Equipment；
The multiple object is grouped according to the cryptographic Hash；
Each object is grouped and applies different models；
Obtain the test result obtained to each object grouping using corresponding model.
2. the method according to claim 1, wherein described obtain applies respective mode to each object grouping
The test result that type obtains, comprising:
Bury a little for the object in each object grouping；
After to the object application corresponding model in each object grouping, the object in each object grouping is obtained
Bury a log；
The determining test result obtained to each object grouping using corresponding model of a log is buried according to described.
3. the method according to claim 1, wherein further including being visualized to the test result.
4. according to the method described in claim 3, it is characterized in that, visualized to the test result, including it is logical
It crosses any one or more following mode to visualize the test result: table, curve graph, column diagram.
5. the method according to claim 1, wherein described divide the multiple object according to the cryptographic Hash
Group, comprising:
The multiple object is divided into N parts according to the cryptographic Hash, N >=2；
It is divided into the multiple object grouping for described N parts.
6. according to the method described in claim 5, it is characterized by:
It is described that the multiple object is divided into N parts according to the cryptographic Hash including resulting remaining divided by N according to the cryptographic Hash
Number, is divided into N parts for the multiple object.
7. according to the method described in claim 1, it is characterized by:
Described be grouped to each object applies different models, including applies new model to the grouping of the first object, to the second object point
Group applies old model, and the number of objects of the first object grouping is less than the number of objects of second object grouping；
In the case where the test result of the new model is better than the test result of the old model, third object is grouped and is applied
The number of objects of the new model, the third object grouping is more than the number of objects of first object grouping.
8. a kind of data analysis set-up, which is characterized in that described device includes:
Computing module is configured as calculating the corresponding cryptographic Hash of object according to the respective identification information of multiple objects, described
Object includes user or user equipment；
Grouping module is configured as being grouped the multiple object according to the cryptographic Hash；
Application module is configured as being grouped each object and applies different models；
Module is obtained, is configured as obtaining the test result for obtaining each object grouping using corresponding model.
9. a kind of electronic equipment, which is characterized in that including memory and processor；Wherein, the memory is for storing one
Or a plurality of computer instruction, wherein one or more computer instruction is executed by the processor to realize claim
The described in any item method and steps of 1-7.
10. a kind of readable storage medium storing program for executing, is stored thereon with computer instruction, which is characterized in that the computer instruction is by processor
Claim 1-7 described in any item method and steps are realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910290693.5A CN110033323A (en) | 2019-04-11 | 2019-04-11 | Data analysing method, device, electronic equipment and readable storage medium storing program for executing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910290693.5A CN110033323A (en) | 2019-04-11 | 2019-04-11 | Data analysing method, device, electronic equipment and readable storage medium storing program for executing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110033323A true CN110033323A (en) | 2019-07-19 |
Family
ID=67238120
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910290693.5A Pending CN110033323A (en) | 2019-04-11 | 2019-04-11 | Data analysing method, device, electronic equipment and readable storage medium storing program for executing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110033323A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110532226A (en) * | 2019-08-06 | 2019-12-03 | 厦门网宿有限公司 | A kind of file comparison method, device and server |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105243006A (en) * | 2015-09-30 | 2016-01-13 | 百度在线网络技术（北京）有限公司 | Flow layer setting method and apparatus based on flow experiment and flow experiment implementing method and apparatus |
| CN105610654A (en) * | 2016-03-02 | 2016-05-25 | 合一网络技术(北京)有限公司 | Server, and policy online test method and system |
| CN106066826A (en) * | 2016-05-24 | 2016-11-02 | 北京金山安全软件有限公司 | Automatic testing method and device for buried point and terminal |
-
2019
- 2019-04-11 CN CN201910290693.5A patent/CN110033323A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105243006A (en) * | 2015-09-30 | 2016-01-13 | 百度在线网络技术（北京）有限公司 | Flow layer setting method and apparatus based on flow experiment and flow experiment implementing method and apparatus |
| CN105610654A (en) * | 2016-03-02 | 2016-05-25 | 合一网络技术(北京)有限公司 | Server, and policy online test method and system |
| CN106066826A (en) * | 2016-05-24 | 2016-11-02 | 北京金山安全软件有限公司 | Automatic testing method and device for buried point and terminal |
Non-Patent Citations (1)
| Title |
|---|
| 叶青青: "从新闻人到产品经理", 《网络与新媒体传播核心教材丛书 网络与新媒体财经报道》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110532226A (en) * | 2019-08-06 | 2019-12-03 | 厦门网宿有限公司 | A kind of file comparison method, device and server |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ellis | Effect sizes and the interpretation of research results in international business | |
| Zebende et al. | Detrended multiple cross-correlation coefficient | |
| US20080168376A1 (en) | Visual designer for non-linear domain logic | |
| Collins et al. | An improvement selection methodology for key performance indicators | |
| Wielebski et al. | Graphically supported evaluation of mapping techniques used in presenting spatial accessibility | |
| Pradana et al. | The implication of digital organisational culture on firm performance | |
| Kock | Using data labels to discover moderating effects in PLS-based structural equation modeling | |
| CN110502277A (en) | A kind of bad taste detection method of code based on BP neural network | |
| CN108268624A (en) | User data method for visualizing and system | |
| Yeon et al. | A dynamic diffusion model for managing customer's expectation and satisfaction | |
| Geng et al. | Empirical research on the spatial distribution and determinants of regional e-commerce in China: Evidence from Chinese provinces | |
| CN108230005A (en) | Online cloud service processing system, evaluation method and computer program product thereof | |
| CN108665513A (en) | Drawing practice based on user behavior data and device | |
| Luigi et al. | THE USE OF REGRESSION ANALYSIS IN MARKETING RESEARCH. | |
| Chiemeke et al. | An assessment of the impact of digital technology adoption on economic growth and labour productivity in Nigeria | |
| CN107958346A (en) | The recognition methods of abnormal behaviour and device | |
| CN110033323A (en) | Data analysing method, device, electronic equipment and readable storage medium storing program for executing | |
| Castro et al. | Bottlenecks or inefficiency? An assessment of first instance Italian courts’ performance | |
| Li et al. | Applying time-series analysis to detect scale drift | |
| Bekkers et al. | Complexity, tacit knowledge and the scope for technological catch‐up | |
| JMP® 9 | Basic analysis and graphing | |
| Chien | The Influences of Balanced Scorecard upon Intellectual Capital Accumulation Implemented by Listed LED Companies in Taiwan: Amoderator of organizational commitment | |
| Olga | Criterion of identification of test steps in training and educational systems. | |
| Kvaløy et al. | A class of tests for trend in time censored recurrent event data | |
| CN107357847A (en) | Data processing method and its device |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190719 |