CN106250481A - Data digging methods based on big data and device - Google Patents
Data digging methods based on big data and device Download PDFInfo
- Publication number
- CN106250481A CN106250481A CN201610619432.XA CN201610619432A CN106250481A CN 106250481 A CN106250481 A CN 106250481A CN 201610619432 A CN201610619432 A CN 201610619432A CN 106250481 A CN106250481 A CN 106250481A
- Authority
- CN
- China
- Prior art keywords
- order data
- data set
- client
- sub
- order
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Databases & Information Systems (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Technology Law (AREA)
- Probability & Statistics with Applications (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of data digging method based on big data, the method includes: obtain the first order data set of the client having bought specified services, based on default client segmentation rule, the order data of this first order data set Zhong Ge section is divided into multiple second sub-order data set, and respectively the plurality of second sub-order data set is carried out data mining according to preset fuzzy data mining algorithm, obtain the user model feature that multiple second sub-order data set is the most corresponding.The invention also discloses a kind of data mining device based on big data, can effectively realize the process to big data, and from the user model feature of big extracting data each second sub-order data set based on client segmentation regular partition correspondence respectively, it is possible to provide for insurance company and formulate company management and the reference of sales tactics and be easy to develop more client.
Description
Technical field
The present invention relates to big data analysis field, particularly relate to a kind of data digging method based on big data and device.
Background technology
The particularity of the service due to insurance and products thereof so that client seems the heaviest for insurance company
Want, client be exactly insurance company existence this, along with buy insurance client get more and more, in insurance company in data base protect
The data deposited the most more are come the more, are formed and comprise mass efficient information in big data, and big data, and insurance company can be instructed real
The most such as price fixing, customer account management etc. work.
But, at present, a kind of effective manner can be from big extracting data user model feature.
Foregoing is only used for auxiliary and understands technical scheme, does not represent and recognizes that foregoing is existing skill
Art.
Summary of the invention
Present invention is primarily targeted at a kind of data digging methods based on big data of offer and device, it is intended to solve existing
Have in technology that do not exist effectively can be from the technical problem of the scheme of big extracting data user model feature.
For achieving the above object, a kind of based on big data the data digging methods that the present invention provides, described method includes:
Obtain the first order data set of the client having bought specified services；
Based on default client segmentation rule, the order data of each client in described first order data set is divided into
Multiple second sub-order data set；
Respectively multiple described second sub-order data set are carried out data according to preset fuzzy data mining algorithm to dig
Pick, obtains the user model feature that multiple described second sub-order data set is the most corresponding.
Preferably, described based on default client segmentation rule, the order data of described client is divided into multiple order collection
Close, including:
Extract the purchase number of times of each client in described first order data set, to other lead referral number of success and purchase
Buy total value；
Based on the purchase number of times of each client in described first order data set, to other lead referral number of success, purchase
Buy total value and the weight coefficient of each parameter that pre-sets calculates the weighted value of each client in described first order data set；
Based on the weight sector pre-set and the weighted value of described each client, by each in described first order data set
The order data of client is divided in multiple described second sub-order data set.
Preferably, the step of the first order data set of the client that specified services has been bought in described acquisition includes:
The order data of all clients having bought specified services is obtained from data base；
The order data of described all clients is carried out data cleansing, obtains described first order data set.
Preferably, described according to preset fuzzy data mining algorithm respectively to multiple described second sub-order data set
Carry out data mining, obtain the user model feature that multiple described second sub-order data set is the most corresponding, including:
For any one the second sub-order data set, obtain each second sub-order data collection as follows
The user model feature that conjunction is corresponding:
The order data of each client comprised from described second sub-order data set extracts at least one specified type
Client parameter value, constitute the first matrix of described second sub-order data set；
First matrix of described second sub-order data set is normalized, obtains described second sub-order numbers
According to the second matrix；
Based on the fuzzy data mining algorithm pre-set, described second matrix is carried out data mining, obtain described second
The user model feature that sub-order data set is corresponding.
Preferably, described based on the fuzzy data mining algorithm pre-set, described second matrix is carried out data mining,
Obtain the user model feature that described second sub-order data set is corresponding, including:
Utilize the maxmini algorithm in fuzzy data mining algorithm to obtain described second matrix norm and stick with paste similar matrix；
Utilize Maximum Tree Algorithm that described fuzzy similarity matrix is carried out cluster analysis process, obtain maximal tree, described maximum
Tree is the user model feature that described second sub-order data set is corresponding.
In order to solve the problems referred to above, the present invention also provides for a kind of data mining device based on big data, described device bag
Include:
Acquisition module, for obtaining the first order data set of the client having bought specified services；
Divide module, for based on default client segmentation rule by the ordering of each client in described first order data set
Forms data is divided into multiple second sub-order data set；
Excavate module, be used for according to preset fuzzy data mining algorithm respectively to multiple described second sub-order data collection
Conjunction carries out data mining, obtains the user model feature that multiple described second sub-order data set is the most corresponding.
Preferably, described division module includes:
Extraction module, for extracting the purchase number of times of each client in described first order data set, pushing away to other clients
Recommend number of success and the total value of purchases；
First computing module, for based on the purchase number of times of each client in described first order data set, to other visitors
Family recommends number of success, the total value of purchases and the weight coefficient of each parameter that pre-sets to calculate described first order data set
In the weighted value of each client；
Data divide module, for based on the weight sector pre-set and the weighted value of described each client, by described the
During in one order data set, the order data of each client is divided to multiple described second sub-order data set.
Preferably, described acquisition module includes:
Data acquisition module, for obtaining the order data of all clients having bought specified services from data base；
Cleaning module, for the order data of described all clients is carried out data cleansing, obtains described first order numbers
According to set.
Preferably for any one the second sub-order data set, described excavation module includes:
Parameter extraction module, extracts for the order data of each client comprised from described second sub-order data set
The client parameter value of at least one specified type, constitutes the first matrix of described second sub-order data set；
Normalization module, for being normalized the first matrix of described second sub-order data set, obtains
Second matrix of described second sub-order data；
Data-mining module, for carrying out data based on the fuzzy data mining algorithm pre-set to described second matrix
Excavate, obtain the user model feature that described second sub-order data set is corresponding.
Preferably, described data-mining module includes:
Second computing module, for utilizing the maxmini algorithm in fuzzy data mining algorithm to obtain described second matrix
Fuzzy similarity matrix；
3rd computing module, is used for utilizing Maximum Tree Algorithm that described fuzzy similarity matrix is carried out cluster analysis process,
To maximal tree, described maximal tree is the user model feature that described second sub-order data set is corresponding.
The present invention provides a kind of data digging method based on big data, and the method includes: obtains and has bought specified services
The first order data set of client, based on default client segmentation rule ordering this first order data set Zhong Ge section
Forms data is divided into multiple second sub-order data set, and according to preset fuzzy data mining algorithm respectively to the plurality of
Two sub-order data set carry out data mining, and the user model obtaining multiple second sub-order data set the most corresponding is special
Levy, by the way, it is possible to effectively realize the process to big data, and from big extracting data based on client segmentation rule
The user model feature of each second sub-order data set correspondence respectively divided, it is possible to provide formulation company pipe for insurance company
Reason and the reference of sales tactics and be easy to develop more client.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of data digging methods based on big data in first embodiment of the invention；
Fig. 2 is the schematic flow sheet of the refinement step of step 102 in first embodiment shown in Fig. 1 of the present invention；
Fig. 3 is the schematic flow sheet of the refinement step of step 101 in first embodiment shown in Fig. 1 of the present invention；
Fig. 4 is the schematic flow sheet of the refinement step of step 103 in first embodiment shown in Fig. 1 of the present invention；
Fig. 5 is the schematic flow sheet of the refinement step of step 403 in embodiment illustrated in fig. 4；
Fig. 6 is the schematic diagram of the functional module of data mining devices based on big data in second embodiment of the invention；
Fig. 7 is the schematic diagram of the refinement functional module dividing module 602 in the second embodiment shown in Fig. 6 of the present invention；
Fig. 8 is the schematic diagram of the refinement functional module of acquisition module 601 in the second embodiment shown in Fig. 6 of the present invention；
Fig. 9 is the schematic diagram of the refinement functional module excavating module 603 in the second embodiment shown in Fig. 6 of the present invention；
Figure 10 is the schematic diagram of the refinement functional module of data-mining module 803 in embodiment illustrated in fig. 8 of the present invention.
The realization of the object of the invention, functional characteristics and advantage will in conjunction with the embodiments, are described further referring to the drawings.
Detailed description of the invention
Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
Owing to prior art not having a kind of effective manner can cause protecting from big extracting data user model feature
Danger company cannot specify more excellent management strategy and sales tactics based on the user model feature of the client buying insurance, does not sends out
Wave the effect of big data.
In order to solve above-mentioned technical problem, the present invention proposes a kind of data digging method based on big data, it is possible to effectively
Realize process to big data, and from big extracting data each second sub-order data collection based on client segmentation regular partition
Close the most corresponding user model feature, it is possible to provide for insurance company and formulate company management and the reference of sales tactics and be easy to
Develop more client.
Referring to Fig. 1, for data digging methods based on big data a kind of in first embodiment of the invention, its feature exists
In, described method includes:
Step 101, acquisition have bought the first order data set of the client of specified services；
In embodiments of the present invention, insurance company is all arranged by the server of its correspondence, and arranges use on that server
In the data base preserving customer data.
Wherein, above-mentioned data digging method based on big data is by data mining devices based on big data (letter below
Claim: data mining device) realize, this data mining device can timing or reach default value in the data volume of big data
The above-mentioned data digging method of Shi Qidong, or in the case of receiving data mining instruction, start above-mentioned data digging method.
In embodiments of the present invention, when needs carry out data mining, appointment industry has been bought in acquisition by data mining device
The first order data set of the client of business.Wherein, during this specified services can be all business that insurance company provides extremely
Few a kind of business.Such as, an insurance company can provide one or more in following business: life insurance business, health
Insurance business, accident/injury insurance business, group insurance business, endowment insurance business, juvenile education gold insurance business.
Wherein, the first order data set comprises the order data of the client buying this specified services, this order
Data comprise the personal information of client and the purchase information of business.
Wherein, the personal information of client comprises account and encrypted message, the ID card information of customer authentication, the family of client
The information such as address, mailbox, wherein, the purchase information of business includes: buys the sequence information of business, buy number of times, to other visitors
The information such as number of success, the total value of purchases is recommended at family.Wherein, successfully refer to that client is by the purchase chain of certain business to other lead referral
Other clients are given in sending and receiving, and other clients link based on this purchase and successfully have purchased this business.
Step 102, based on default client segmentation rule by the order numbers of each client in described first order data set
According to being divided into multiple second sub-order data set；
Step 103, respectively multiple described second sub-order data set are entered according to preset fuzzy data mining algorithm
Row data mining, obtains the user model feature that multiple described second sub-order data set is the most corresponding.
In embodiments of the present invention, data mining device is obtaining buying the first order data collection of the client of specified services
After conjunction, put up with default client segmentation rule and the order data of each client in the first order data set is divided into multiple the
Two sub-order data set.
It is understood that this client segmentation rule pre-set can be arranged by system default, it is also possible to by managing
Personnel are configured as required, and this client segmentation rule can be to divide according to geographical position, or according to purchase
The type of business divide, or can be based on buying number of times, to visitors such as other lead referral number of success, total value of purchases
The comprehensive weight of family parameter divides, it should be noted that in actual applications, can divide according to concrete needs,
Do not repeat.
Wherein, the order data of each client in the first order data set is being divided into multiple second sub-order data collection
After conjunction, the plurality of second sub-order data set is entered respectively by data mining device according to preset fuzzy data mining algorithm
Row data mining, obtains the user model feature that multiple second sub-order data set is the most corresponding, enabling effectively real
Existing data mining.
In embodiments of the present invention, data mining device obtains the first order data collection of the client having bought specified services
Close, based on default client segmentation rule, the order data of this first order data set Zhong Ge section is divided into multiple second son
Order data set, and according to preset fuzzy data mining algorithm respectively to the plurality of second sub-order data set number
According to excavation, obtain the user model feature that multiple second sub-order data set is the most corresponding, by the way, it is possible to effectively
Realize process to big data, and from big extracting data each second sub-order data collection based on client segmentation regular partition
Close the most corresponding user model feature, it is possible to provide for insurance company and formulate company management and the reference of sales tactics and be easy to
Develop more client.
For the technical scheme being better understood from the embodiment of the present invention, refer to Fig. 2, for shown in Fig. 1 of the present invention first
The schematic flow sheet of the refinement step of step 102 in embodiment, this step 102 is: based on default client segmentation rule by institute
The order data stating client is divided into multiple order set, and the refinement step of this step 102 includes:
Step 201, extract the purchase number of times of each client in described first order data set, to the success of other lead referral
Number of times and the total value of purchases；
Step 202, based on the purchase number of times of each client in described first order data set, to other lead referral successes
Number of times, the total value of purchases and the weight coefficient of each parameter pre-set calculate each client in described first order data set
Weighted value；
Step 203, based on the weight sector pre-set and the weighted value of described each client, by described first order data
During in set, the order data of each client is divided to multiple described second sub-order data set.
In embodiments of the present invention, data mining device is obtaining buying the first order data collection of the client of specified services
After conjunction, by extracting the purchase number of times of each client in this first order data set, to other lead referral number of success and purchase
Buy total value, and by based on the purchase number of times of each client in this first order data set, to other lead referral number of success, purchase
Buy total value and the weight coefficient of each parameter that pre-sets calculates the weighted value of each client in this first order data set.
Wherein, the computing formula of the weighted value calculating client is as follows:
M=a*x1 ,+b*x2+c*x3
Wherein, M represents the weighted value of client, and a represents the purchase number of times of client, and x1 represents the weight buying number of times of client
Coefficient, b represents that, to other lead referral number of success, x2 represents client's weight coefficient to other lead referral number of success, c
Representing the total value of purchases of client, x3 represents the weight coefficient of the total value of purchases of client.
By above-mentioned formula, it is possible to be effectively calculated the weighted value of each client in the first order data set.
In embodiments of the present invention, in obtaining the first order data set after the weighted value of each client, will be based in advance
The weighted value of each client in the weight sector first arranged and the first order data set, by each client in the first order data set
Order data be divided in multiple second sub-order data set.Concrete: data mining device is by the first order data collection
In conjunction, the weighted value of each client is normalized, and obtains the weighted value after each client's normalization, the above-mentioned power pre-set
Weight interval can be [0,0.1), [0.1,0.2), [0.2,0.3), [0.3,0.4), [0.4,0.5), [0.5,0.6), [0.7,
0.8), [0.8,0.9), [0.9,1], and based on above-mentioned weight sector, each client can be divided to the second sub-order of correspondence
In data acquisition system.
In embodiments of the present invention, data mining device will extract the purchase of each client time in the first order data set
Several, to other lead referral number of success and the total value of purchases and secondary based on the purchase of each client in this first order data set
Number, calculate the first order numbers to other lead referral number of success, the total value of purchases and the weight coefficient of each parameter that pre-sets
According to the weighted value of each client in set, and based on the weight sector pre-set and the weighted value of each client, by the first order numbers
It is divided in multiple second sub-order data set, by the way according to the order data of each client in set, it is possible to based on
Realize based on buying number of times, realizing in the first order data set to other parameters such as lead referral number of success and the total value of purchases
The division of each client.
Based on first embodiment shown in Fig. 1, refer to Fig. 3, for step 101 in first embodiment shown in Fig. 1 of the present invention
The schematic flow sheet of refinement step, this step 101 is: obtain the first order data set of the client having bought specified services,
And the refinement step of this step 101 includes:
Step 301, from data base, obtain the order data of all clients having bought specified services；
Step 302, order data to described all clients carry out data cleansing, obtain described first order data collection
Close.
In embodiments of the present invention, the order data of client is all stored in data base, and this data base can be
Hadoop data base, or the other kinds of data base being suitable for the storage of big data.
In embodiments of the present invention, data mining device will obtain all clients buying specified services from data base
Order data, and the order data of these all clients is carried out data cleansing, to obtain the first order data set.Wherein,
Some the invalid order data in the order data of all clients can be removed by the way of data cleansing.
Based on the first embodiment shown in Fig. 1, refer to Fig. 4, for step 103 in first embodiment shown in Fig. 1 of the present invention
The schematic flow sheet of refinement step, this step 103 is: according to preset fuzzy data mining algorithm respectively to multiple described
Two sub-order data set carry out data mining, obtain the user model that multiple described second sub-order data set is the most corresponding
Feature, the refinement step of this step 103 includes:
Step 401, each client comprised from described second sub-order data set order data extract at least one
The client parameter value of specified type, constitutes the first matrix of described second sub-order data set；
Step 402, the first matrix to described second sub-order data set are normalized, and obtain described second
Second matrix of sub-order data；
Step 403, based on the fuzzy data mining algorithm pre-set, described second matrix is carried out data mining, obtain
The user model feature that described second sub-order data set is corresponding.
In embodiments of the present invention, for any one the second sub-order data set, data mining device all will be according to
Step 401 to the content that step 403 describes obtains the user model feature that each the second sub-order data set is corresponding.
In embodiments of the present invention, the order data of each client in the first order data set is being drawn by data mining device
After dividing to multiple second sub-order data set, for each the second sub-order data set, data mining device will
The client parameter extracting at least one specified type the order data of each client comprised is combined from this second sub-order data
Value, constitutes the first matrix of this second sub-order data set, wherein, the client parameter value of at least one specified type of extraction
Can be purchase number of times, retain situation, continuation of insurance number of times at present, recommend number of times etc. to other people.Wherein, if reservation situation is at present
It is that then this value retaining situation at present is 1, if reservation situation is no at present, the value retaining situation the most at present is 0.
And after obtaining the first matrix of above-mentioned second sub-order data set, will be to this second sub-order data set
The first matrix be normalized, obtain the second matrix of this second sub-order data.Concrete: in view of the first matrix
In data be not the number of [0,1] closed interval, so should be by these initial data standardization, so that each index
Value is unified in the numerical characteristic scope that certain is common.First will calculate in the second sub-order data, the parameter of each type
Meansigma methods, such as, calculates the meansigma methods buying number of times of all clients in the second sub-order data set, calculates the second sub-order
The meansigma methods of the situation that retains at present of all clients in data acquisition system, in calculating the second sub-order data set, all clients' is continuous
Protect the meansigma methods of number of times, and recommend the meansigma methods of number of times to other people.Mining data device also will calculate the second sub-order numbers simultaneously
According to, the standard deviation of the parameter of each type, and the meansigma methods of parameter based on above-mentioned each type and each type
The standard deviation of parameter calculate the standard deviation of each data in the first matrix, to obtain normalized matrix.And owing to now obtaining
Normalized matrix is not interval interior in [0,1], the extreme value standardized algorithm pair that employing is also pre-set by data mining device
This normalized matrix processes, and to obtain normalized matrix, is the second above-mentioned matrix.
And in embodiments of the present invention, data mining device also by based on the fuzzy data algorithm pre-set to this second
Matrix carries out data mining, obtains the user model feature that this second sub-order data set is corresponding.
In embodiments of the present invention, the order numbers of each client that data mining device comprises from the second sub-order data set
According to the client parameter value of middle at least one specified type of extraction, constitute the first matrix of the second sub-order data set, and to this
First matrix of the second sub-order data set is normalized, and obtains the second matrix of this second sub-order data, and
Based on the fuzzy data mining algorithm pre-set, this second matrix is carried out data mining, obtain this second sub-order data collection
Close corresponding user model feature, by the way, it is possible to be effectively obtained the client that the second sub-order data set is corresponding
Pattern feature.
Further, refer to Fig. 5, for the flow process signal of the refinement step of step 403 in embodiment illustrated in fig. 4 of the present invention
Figure, this step 403 is: based on the fuzzy data mining algorithm pre-set, described second matrix is carried out data mining, obtains
The user model feature that described second sub-order data set is corresponding, and the refinement step of this step 403 includes:
Step 501, utilize the maxmini algorithm in fuzzy data mining algorithm obtain described second matrix norm stick with paste phase
Like matrix；
Step 502, utilize Maximum Tree Algorithm that described fuzzy similarity matrix is carried out cluster analysis process, obtain maximal tree,
Described maximal tree is the user model feature that described second sub-order data set is corresponding.
In embodiments of the present invention, data mining device will utilize the maxmini algorithm in fuzzy data mining algorithm to obtain
Similar matrix is stuck with paste to the second matrix norm.And utilize Maximum Tree Algorithm that fuzzy similarity matrix is carried out cluster analysis process, obtain
Maximal tree, this maximal tree is the user model feature that the second sub-order data set is corresponding.
Wherein, maxmini algorithm is the one in fuzzy data mining algorithm, for each number in the second matrix
According to maxmini algorithm can be used to carry out Fuzzy Processing, to obtain the second matrix norm paste similar matrix, and it is being somebody's turn to do
After fuzzy similarity matrix, use Maximum Tree Algorithm to carry out cluster analysis process and obtain maximal tree, by using Maximum Tree Algorithm
Carry out cluster analysis to process and can construct a figure with all objects being classified as summit, and (its when Rij is not equal to 0
Middle Rij is data in fuzzy similarity matrix), summit i and summit j just can be linked to be a limit, and its method is first to draw
Some i in vertex set, then connects limit by Rij order from big to small, it is desirable to do not produce loop successively, until all of
Summit all by till connection, most obtains a maximal tree, and each the limit of tree can give a certain numerical value, i.e. available visitor
Family pattern feature.
In embodiments of the present invention, by the way, it is possible to effectively determine that the second sub-order data set is corresponding
User model feature.
Refer to Fig. 6, for the showing of functional module of data mining devices based on big data in second embodiment of the invention
It is intended to, should include by data mining devices based on big data:
Acquisition module 601, for obtaining the first order data set of the client having bought specified services；
In embodiments of the present invention, when needs carry out data mining, specified services has been bought in acquisition by acquisition module 601
The first order data set of client.Wherein, during this specified services can be all business that insurance company provides at least
A kind of business.Such as, an insurance company can provide one or more in following business: life insurance business, health are protected
Danger business, accident/injury insurance business, group insurance business, endowment insurance business, juvenile education gold insurance business.
Wherein, the first order data set comprises the order data of the client buying this specified services, this order
Data comprise the personal information of client and the purchase information of business.
Wherein, the personal information of client comprises account and encrypted message, the ID card information of customer authentication, the family of client
The information such as address, mailbox, wherein, the purchase information of business includes: buys the sequence information of business, buy number of times, to other visitors
The information such as number of success, the total value of purchases is recommended at family.Wherein, successfully refer to that client is by the purchase chain of certain business to other lead referral
Other clients are given in sending and receiving, and other clients link based on this purchase and successfully have purchased this business.
Divide module 602, be used for each client in described first order data set based on default client segmentation rule
Order data be divided into multiple second sub-order data set；
Excavate module 603, be used for according to preset fuzzy data mining algorithm respectively to multiple described second sub-order numbers
Carry out data mining according to set, obtain the user model feature that multiple described second sub-order data set is the most corresponding.
In embodiments of the present invention, after obtaining the first order data set of client of purchase specified services, divide
Module 602 is put up with default client segmentation rule and the order data of each client in the first order data set is divided into multiple the
Two sub-order data set.
It is understood that this client segmentation rule pre-set can be arranged by system default, it is also possible to by managing
Personnel are configured as required, and this client segmentation rule can be to divide according to geographical position, or according to purchase
The type of business divide, or can be based on buying number of times, to visitors such as other lead referral number of success, total value of purchases
The comprehensive weight of family parameter divides, it should be noted that in actual applications, can divide according to concrete needs,
Do not repeat.
Wherein, the order data of each client in the first order data set is being divided into multiple second sub-order data collection
After conjunction, excavate module 603 and respectively the plurality of second sub-order data set is entered according to preset fuzzy data mining algorithm
Row data mining, obtains the user model feature that multiple second sub-order data set is the most corresponding, enabling effectively real
Existing data mining.
In embodiments of the present invention, data mining device obtains the first order data collection of the client having bought specified services
Close, based on default client segmentation rule, the order data of this first order data set Zhong Ge section is divided into multiple second son
Order data set, and according to preset fuzzy data mining algorithm respectively to the plurality of second sub-order data set number
According to excavation, obtain the user model feature that multiple second sub-order data set is the most corresponding, by the way, it is possible to effectively
Realize process to big data, and from big extracting data each second sub-order data collection based on client segmentation regular partition
Close the most corresponding user model feature, it is possible to provide for insurance company and formulate company management and the reference of sales tactics and be easy to
Develop more client.
Refer to Fig. 7, for the second embodiment shown in Fig. 6 of the present invention divides the signal of the refinement functional module of module 602
Figure, described division module 602 includes:
Extraction module 701, for extracting the purchase number of times of each client in described first order data set, to other clients
Recommend number of success and the total value of purchases；
First computing module 702, for based on the purchase number of times of each client in described first order data set, to other
Lead referral number of success, the total value of purchases and the weight coefficient of each parameter pre-set calculate described first order data collection
The weighted value of each client in conjunction；
Data divide module 703, for based on the weight sector pre-set and the weighted value of described each client, by described
During in first order data set, the order data of each client is divided to multiple described second sub-order data set.
In embodiments of the present invention, after obtaining the first order data set of client of purchase specified services, extract
Module 701 is by extracting the purchase number of times of each client in this first order data set, to other lead referral number of success and purchases
Buy total value, and the first computing module 702 is by based on the purchase number of times of each client in this first order data set, to other clients
Number of success, the total value of purchases and the weight coefficient of each parameter that pre-sets is recommended to calculate each visitor in this first order data set
The weighted value at family.
Wherein, the computing formula of the weighted value calculating client is as follows:
M=a*x1 ,+b*x2+c*x3
Wherein, M represents the weighted value of client, and a represents the purchase number of times of client, and x1 represents the weight buying number of times of client
Coefficient, b represents that, to other lead referral number of success, x2 represents client's weight coefficient to other lead referral number of success, c
Representing the total value of purchases of client, x3 represents the weight coefficient of the total value of purchases of client.
By above-mentioned formula, it is possible to be effectively calculated the weighted value of each client in the first order data set.
In embodiments of the present invention, in obtaining the first order data set after the weighted value of each client, data divide
Module 703 is by based on the weighted value of each client in the weight sector pre-set and the first order data set, by the first order
During in data acquisition system, the order data of each client is divided to multiple second sub-order data set.Concrete: data mining device
The weighted value of each client in first order data set is normalized, obtains the weighted value after each client's normalization,
The above-mentioned weight sector pre-set can be [0,0.1), [0.1,0.2), [0.2,0.3), [0.3,0.4), [0.4,0.5),
[0.5,0.6), [0.7,0.8), [0.8,0.9), [0.9,1], and based on above-mentioned weight sector, each client can be divided to
In the second corresponding sub-order data set.
In embodiments of the present invention, data mining device will extract the purchase of each client time in the first order data set
Several, to other lead referral number of success and the total value of purchases and secondary based on the purchase of each client in this first order data set
Number, calculate the first order numbers to other lead referral number of success, the total value of purchases and the weight coefficient of each parameter that pre-sets
According to the weighted value of each client in set, and based on the weight sector pre-set and the weighted value of each client, by the first order numbers
It is divided in multiple second sub-order data set, by the way according to the order data of each client in set, it is possible to based on
Realize based on buying number of times, realizing in the first order data set to other parameters such as lead referral number of success and the total value of purchases
The division of each client.
Refer to Fig. 8, for the signal of the refinement functional module of acquisition module 601 in the second embodiment shown in Fig. 6 of the present invention
Figure, this acquisition module 601 includes:
Data acquisition module 801, for obtaining the order data of all clients having bought specified services from data base；
Cleaning module 802, for the order data of described all clients is carried out data cleansing, obtains described first order
Data acquisition system.
In embodiments of the present invention, the order data of client is all stored in data base, and this data base can be
Hadoop data base, or the other kinds of data base being suitable for the storage of big data.
In embodiments of the present invention, data acquisition module 801 buys all of specified services by obtaining from data base
The order data of client, and cleaning module 802 carries out data cleansing to the order data of these all clients, to obtain the first order
Data acquisition system.Wherein, some the invalid orders in the order data of all clients can be removed by the way of data cleansing
Data.
Refer to Fig. 9, for the second embodiment shown in Fig. 6 of the present invention is excavated the signal of the refinement functional module of module 603
Figure, this excavation module 603 includes:
Parameter extraction module 901, for the order data of each client comprised from described second sub-order data set
Extract the client parameter value of at least one specified type, constitute the first matrix of described second sub-order data set；
Normalization module 902, for the first matrix of described second sub-order data set is normalized,
The second matrix to described second sub-order data；
Data-mining module 903, for carrying out described second matrix based on the fuzzy data mining algorithm pre-set
Data mining, obtains the user model feature that described second sub-order data set is corresponding.
In embodiments of the present invention, the order data of each client in the first order data set is being drawn by data mining device
After dividing to multiple second sub-order data set, for each the second sub-order data set, parameter extraction module 901
The order data of each client comprised from the combination of this second sub-order data will be extracted client's ginseng of at least one specified type
Numerical value, constitutes the first matrix of this second sub-order data set, wherein, the client parameter of at least one specified type of extraction
Value can be purchase number of times, retains situation, continuation of insurance number of times at present, recommend number of times etc. to other people.Wherein, if retaining situation at present
Be yes, then this value retaining situation at present is 1, if reservation situation is no at present, the value retaining situation the most at present is 0.
And after obtaining the first matrix of above-mentioned second sub-order data set, normalization module 902 will to this second
First matrix of sub-order data set is normalized, and obtains the second matrix of this second sub-order data.Concrete:
The number of [0,1] closed interval in view of the data in the first matrix, thus should by these initial data standardization, thus
Each desired value is made to be unified in certain common numerical characteristic scope.First will calculate in the second sub-order data, each
The meansigma methods of the parameter of type, such as, calculates the meansigma methods buying number of times of all clients in the second sub-order data set,
Calculate the meansigma methods of the situation that retains at present of all clients in the second sub-order data set, calculate the second sub-order data set
In the meansigma methods of continuation of insurance number of times of all clients, and recommend the meansigma methods of number of times to other people.Mining data device also will meter simultaneously
Calculate in the second sub-order data, the standard deviation of the parameter of each type, and parameter based on above-mentioned each type is average
The standard deviation of the parameter of value and each type calculates the standard deviation of each data in the first matrix, to obtain normalized matrix.And
Owing to the normalized matrix now obtained is not the most in [0,1] is interval, the pole that employing is also pre-set by data mining device
This normalized matrix is processed by value standardized algorithm, to obtain normalized matrix, is the second above-mentioned matrix.
And in embodiments of the present invention, data-mining module 903 also by based on the fuzzy data algorithm pre-set to this
Second matrix carries out data mining, obtains the user model feature that this second sub-order data set is corresponding.
In embodiments of the present invention, the order numbers of each client that data mining device comprises from the second sub-order data set
According to the client parameter value of middle at least one specified type of extraction, constitute the first matrix of the second sub-order data set, and to this
First matrix of the second sub-order data set is normalized, and obtains the second matrix of this second sub-order data, and
Based on the fuzzy data mining algorithm pre-set, this second matrix is carried out data mining, obtain this second sub-order data collection
Close corresponding user model feature, by the way, it is possible to be effectively obtained the client that the second sub-order data set is corresponding
Pattern feature.
Refer to Figure 10, for the signal of the refinement functional module of data-mining module 803 in embodiment illustrated in fig. 8 of the present invention
Figure, this data-mining module 803 includes:
Second computing module 1001, for utilizing the maxmini algorithm in fuzzy data mining algorithm to obtain described second
Matrix norm sticks with paste similar matrix；
3rd computing module 1002, is used for utilizing Maximum Tree Algorithm to carry out described fuzzy similarity matrix at cluster analysis
Reason, obtains maximal tree, and described maximal tree is the user model feature that described second sub-order data set is corresponding.
In embodiments of the present invention, the minimax utilized in fuzzy data mining algorithm is calculated by the second computing module 1001
Method obtains the second matrix norm and sticks with paste similar matrix.And the 3rd computing module 1002 utilize Maximum Tree Algorithm that fuzzy similarity matrix is entered
Row cluster analysis processes, and obtains maximal tree, and this maximal tree is the user model feature that the second sub-order data set is corresponding.
Wherein, maxmini algorithm is the one in fuzzy data mining algorithm, for each number in the second matrix
According to maxmini algorithm can be used to carry out Fuzzy Processing, to obtain the second matrix norm paste similar matrix, and it is being somebody's turn to do
After fuzzy similarity matrix, use Maximum Tree Algorithm to carry out cluster analysis process and obtain maximal tree, by using Maximum Tree Algorithm
Carry out cluster analysis to process and can construct a figure with all objects being classified as summit, and (its when Rij is not equal to 0
Middle Rij is data in fuzzy similarity matrix), summit i and summit j just can be linked to be a limit, and its method is first to draw
Some i in vertex set, then connects limit by Rij order from big to small, it is desirable to do not produce loop successively, until all of
Summit all by till connection, most obtains a maximal tree, and each the limit of tree can give a certain numerical value, i.e. available visitor
Family pattern feature.
In embodiments of the present invention, by the way, it is possible to effectively determine that the second sub-order data set is corresponding
User model feature.
Through the above description of the embodiments, those skilled in the art is it can be understood that arrive above-described embodiment side
Method can add the mode of required general hardware platform by software and realize, naturally it is also possible to by hardware, but a lot of in the case of
The former is more preferably embodiment.Based on such understanding, prior art is done by technical scheme the most in other words
The part going out contribution can embody with the form of software product, and this computer software product is stored in a storage medium
In (such as ROM/RAM, magnetic disc, CD), including some instructions with so that a station terminal equipment (can be mobile phone, computer, take
Business device, air-conditioner is, or the network equipment etc.) method that performs each embodiment of the present invention.
These are only the preferred embodiments of the present invention, not thereby limit the scope of the claims of the present invention, every utilize this
Equivalent structure or equivalence flow process that bright description and accompanying drawing content are made convert, or are directly or indirectly used in other relevant skills
Art field, is the most in like manner included in the scope of patent protection of the present invention.
Claims (10)
1. a data digging method based on big data, it is characterised in that described method includes:
Obtain the first order data set of the client having bought specified services；
Based on default client segmentation rule, the order data of each client in described first order data set is divided into multiple
Second sub-order data set；
Respectively multiple described second sub-order data set are carried out data mining according to preset fuzzy data mining algorithm,
To the user model feature that multiple described second sub-order data set are the most corresponding.
Method the most according to claim 1, it is characterised in that described regular by described client based on default client segmentation
Order data be divided into multiple order set, including:
Extract the purchase number of times of each client in described first order data set, to other lead referral number of success and buy total
Volume；
Based on the purchase number of times of each client in described first order data set, to other lead referral number of success, buy total
Volume and the weight coefficient of each parameter pre-set calculate the weighted value of each client in described first order data set；
Based on the weight sector pre-set and the weighted value of described each client, by each client in described first order data set
Order data be divided in multiple described second sub-order data set.
Method the most according to claim 1, it is characterised in that what the client of specified services had been bought in described acquisition first orders
The step of forms data set includes:
The order data of all clients having bought specified services is obtained from data base；
The order data of described all clients is carried out data cleansing, obtains described first order data set.
4. according to the method described in claims 1 to 3 any one, it is characterised in that described dig according to preset fuzzy data
Pick algorithm carries out data mining to multiple described second sub-order data set respectively, obtains multiple described second sub-order data
The user model feature that set is the most corresponding, including:
For any one the second sub-order data set, obtain each second sub-order data set pair as follows
The user model feature answered:
The order data of each client comprised from described second sub-order data set is extracted the visitor of at least one specified type
Family parameter value, constitutes the first matrix of described second sub-order data set；
First matrix of described second sub-order data set is normalized, obtains described second sub-order data
Second matrix；
Based on the fuzzy data mining algorithm pre-set, described second matrix is carried out data mining, obtain described second son and order
The user model feature that forms data set is corresponding.
Method the most according to claim 4, it is characterised in that described based on the fuzzy data mining algorithm pair pre-set
Described second matrix carries out data mining, obtains the user model feature that described second sub-order data set is corresponding, including:
Utilize the maxmini algorithm in fuzzy data mining algorithm to obtain described second matrix norm and stick with paste similar matrix；
Utilizing Maximum Tree Algorithm that described fuzzy similarity matrix is carried out cluster analysis process, obtain maximal tree, described maximal tree is i.e.
For the user model feature that described second sub-order data set is corresponding.
6. a data mining device based on big data, it is characterised in that described device includes:
Acquisition module, for obtaining the first order data set of the client having bought specified services；
Divide module, for regular by the order numbers of each client in described first order data set based on default client segmentation
According to being divided into multiple second sub-order data set；
Excavate module, for respectively multiple described second sub-order data set being entered according to preset fuzzy data mining algorithm
Row data mining, obtains the user model feature that multiple described second sub-order data set is the most corresponding.
Device the most according to claim 6, it is characterised in that described division module includes:
Extraction module, for extracting the purchase number of times of each client in described first order data set, becoming to other lead referral
Merit number of times and the total value of purchases；
First computing module, for based on the purchase number of times of each client in described first order data set, push away to other clients
Recommend number of success, the total value of purchases and the weight coefficient of each parameter that pre-sets calculates in described first order data set each
The weighted value of client；
Data divide module, for based on the weight sector pre-set and the weighted value of described each client, order described first
During in forms data set, the order data of each client is divided to multiple described second sub-order data set.
Device the most according to claim 6, it is characterised in that described acquisition module includes:
Data acquisition module, for obtaining the order data of all clients having bought specified services from data base；
Cleaning module, for the order data of described all clients is carried out data cleansing, obtains described first order data collection
Close.
9. according to the device described in claim 6 to 8 any one, it is characterised in that for any one the second sub-order numbers
According to set, described excavation module includes:
Parameter extraction module, extracts at least for the order data of each client comprised from described second sub-order data set
The client parameter value of one specified type, constitutes the first matrix of described second sub-order data set；
Normalization module, for being normalized the first matrix of described second sub-order data set, obtains described
Second matrix of the second sub-order data；
Data-mining module, digs for described second matrix being carried out data based on the fuzzy data mining algorithm pre-set
Pick, obtains the user model feature that described second sub-order data set is corresponding.
Device the most according to claim 9, it is characterised in that described data-mining module includes:
Second computing module, for utilizing the maxmini algorithm in fuzzy data mining algorithm to obtain described second matrix norm
Stick with paste similar matrix；
3rd computing module, is used for utilizing Maximum Tree Algorithm that described fuzzy similarity matrix is carried out cluster analysis process, obtains
Big tree, described maximal tree is the user model feature that described second sub-order data set is corresponding.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610619432.XA CN106250481A (en) | 2016-07-29 | 2016-07-29 | Data digging methods based on big data and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610619432.XA CN106250481A (en) | 2016-07-29 | 2016-07-29 | Data digging methods based on big data and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106250481A true CN106250481A (en) | 2016-12-21 |
Family
ID=57606493
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610619432.XA Pending CN106250481A (en) | 2016-07-29 | 2016-07-29 | Data digging methods based on big data and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106250481A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106897893A (en) * | 2016-12-30 | 2017-06-27 | 厦门南讯软件科技有限公司 | It is a kind of to be grouped into the system that customer provides personalized interactive activity according to customer attributes |
| CN106897894A (en) * | 2016-12-30 | 2017-06-27 | 厦门南讯软件科技有限公司 | It is a kind of based on the accurate layered approach of client analyzed customer order |
| CN107688901A (en) * | 2017-08-24 | 2018-02-13 | 北京小度信息科技有限公司 | Data adjustment method and device |
| CN108415908A (en) * | 2017-02-09 | 2018-08-17 | 腾讯科技（北京）有限公司 | A kind of processing method and server of multi-medium data |
| CN109255633A (en) * | 2018-09-05 | 2019-01-22 | 成都柚车科技有限公司 | Shared carwash client management method and system |
| CN112347214A (en) * | 2020-11-06 | 2021-02-09 | 平安科技（深圳）有限公司 | Target area dividing method and device, electronic equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101937541A (en) * | 2009-06-30 | 2011-01-05 | 商文彬 | Method and device for evaluating client credit |
| CN101944221A (en) * | 2010-09-07 | 2011-01-12 | 上海腾唐数码科技有限公司 | Price comparing network shopping system and method |
| CN102982084A (en) * | 2011-11-01 | 2013-03-20 | 微软公司 | Multi-dimentional data report |
| US20140379703A1 (en) * | 2013-06-20 | 2014-12-25 | Actuate Corporation | Generating a Venn Diagram Using a Columnar Database Management System |
| CN104392360A (en) * | 2014-11-11 | 2015-03-04 | 许明 | Resource distribution method and system |
| CN105354210A (en) * | 2015-09-23 | 2016-02-24 | 深圳市爱贝信息技术有限公司 | Mobile game payment account behavior data processing method and apparatus |
| CN105404986A (en) * | 2014-09-15 | 2016-03-16 | 深圳前海百递网络有限公司 | Method and device for obtaining evaluation data |
| CN105488697A (en) * | 2015-12-09 | 2016-04-13 | 焦点科技股份有限公司 | Potential customer mining method based on customer behavior characteristics |
-
2016
- 2016-07-29 CN CN201610619432.XA patent/CN106250481A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101937541A (en) * | 2009-06-30 | 2011-01-05 | 商文彬 | Method and device for evaluating client credit |
| CN101944221A (en) * | 2010-09-07 | 2011-01-12 | 上海腾唐数码科技有限公司 | Price comparing network shopping system and method |
| CN102982084A (en) * | 2011-11-01 | 2013-03-20 | 微软公司 | Multi-dimentional data report |
| US20140379703A1 (en) * | 2013-06-20 | 2014-12-25 | Actuate Corporation | Generating a Venn Diagram Using a Columnar Database Management System |
| CN105404986A (en) * | 2014-09-15 | 2016-03-16 | 深圳前海百递网络有限公司 | Method and device for obtaining evaluation data |
| CN104392360A (en) * | 2014-11-11 | 2015-03-04 | 许明 | Resource distribution method and system |
| CN105354210A (en) * | 2015-09-23 | 2016-02-24 | 深圳市爱贝信息技术有限公司 | Mobile game payment account behavior data processing method and apparatus |
| CN105488697A (en) * | 2015-12-09 | 2016-04-13 | 焦点科技股份有限公司 | Potential customer mining method based on customer behavior characteristics |
Non-Patent Citations (3)
| Title |
|---|
| 曾建勋 等: "《中国科学技术信息研究所 论文集》", 31 January 2013, 科学技术文献出版社 * |
| 李莉 等: "《管理定量分析 决策中常用的分析方法》", 31 August 2007, 上海交通大学出版社 * |
| 王智平: "数据挖掘在保险客户数据中的应用研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106897893A (en) * | 2016-12-30 | 2017-06-27 | 厦门南讯软件科技有限公司 | It is a kind of to be grouped into the system that customer provides personalized interactive activity according to customer attributes |
| CN106897894A (en) * | 2016-12-30 | 2017-06-27 | 厦门南讯软件科技有限公司 | It is a kind of based on the accurate layered approach of client analyzed customer order |
| CN108415908A (en) * | 2017-02-09 | 2018-08-17 | 腾讯科技（北京）有限公司 | A kind of processing method and server of multi-medium data |
| CN108415908B (en) * | 2017-02-09 | 2021-12-10 | 腾讯科技（北京）有限公司 | Multimedia data processing method and server |
| CN107688901A (en) * | 2017-08-24 | 2018-02-13 | 北京小度信息科技有限公司 | Data adjustment method and device |
| CN107688901B (en) * | 2017-08-24 | 2021-05-11 | 北京星选科技有限公司 | Data adjusting method and device |
| CN109255633A (en) * | 2018-09-05 | 2019-01-22 | 成都柚车科技有限公司 | Shared carwash client management method and system |
| CN112347214A (en) * | 2020-11-06 | 2021-02-09 | 平安科技（深圳）有限公司 | Target area dividing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106250481A (en) | Data digging methods based on big data and device | |
| US12086888B2 (en) | Retail deployment model | |
| CN103559636B (en) | Restaurant management system based on cloud computing and intelligent analysis | |
| JP4360759B2 (en) | Method and system for investigating advertising sales effects using correlation analysis | |
| Cadez et al. | Probabilistic modeling of transaction data with applications to profiling, visualization, and prediction | |
| CN106997549A (en) | The method for pushing and system of a kind of advertising message | |
| CN107146089A (en) | The single recognition methods of one kind brush and device, electronic equipment | |
| CN102073956A (en) | Data mining-based directional advertisement release method, system and equipment | |
| CN107341682A (en) | A kind of advertising message monitoring and evaluation method and its device | |
| CN110197390B (en) | Recommendation method and system based on association degree and economic value of association rule | |
| Cumby et al. | Predicting customer shopping lists from point-of-sale purchase data | |
| CN107239442A (en) | A kind of method and apparatus of calculating address similarity | |
| CN108153824A (en) | The determining method and device of targeted user population | |
| CN110009379A (en) | A kind of building of site selection model and site selecting method, device and equipment | |
| CN109087138A (en) | Data processing method and system, computer system and readable storage medium storing program for executing | |
| CN107480227A (en) | A kind of information-pushing method based on big data | |
| CN109919608A (en) | A kind of recognition methods, device and the server of high-risk transaction agent | |
| CN110163245A (en) | Class of service prediction technique and system | |
| CN109376287A (en) | House property map construction method, device, computer equipment and storage medium | |
| CN108268660A (en) | A kind of customer data recommends method, server and storage medium | |
| US7533095B2 (en) | Data mining within a message handling system | |
| Ali et al. | Influential billboard slot selection using pruned submodularity graph | |
| CN110162549A (en) | A kind of fire data analysis method, device, readable storage medium storing program for executing and terminal device | |
| CN106408483A (en) | Meteorology cloud intelligent business method and system | |
| Rossetti et al. | Simulation modeling of customer checkout configurations |
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 |
Address after: 518057 Shenzhen Software Park, Nanshan District high tech Industrial Park, Guangdong, China,, 6 401-402 Applicant after: Yongxing Shenzhen Polytron Technologies Inc Address before: 518057 Shenzhen Software Park, Nanshan District high tech Industrial Park, Guangdong, China,, 6 401-402 Applicant before: Shenzhen Longrise Technology Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161221 |
|
| RJ01 | Rejection of invention patent application after publication |