CN107870936A - The related effective item set mining method, apparatus of data item and data processing equipment - Google Patents
The related effective item set mining method, apparatus of data item and data processing equipment Download PDFInfo
- Publication number
- CN107870936A CN107870936A CN201610854714.8A CN201610854714A CN107870936A CN 107870936 A CN107870936 A CN 107870936A CN 201610854714 A CN201610854714 A CN 201610854714A CN 107870936 A CN107870936 A CN 107870936A
- Authority
- CN
- China
- Prior art keywords
- item
- item collection
- collection
- pending
- data
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the present invention, which provides a kind of related effective item set mining method, apparatus of data item and data processing equipment, this method, to be included：Determine support of the pending item collection in transaction database, and support of each data item in the transaction database of the pending item collection；According to support of the pending item collection in transaction database, and support of each data item in the transaction database of the pending item collection, the item collection degree of correlation of the pending item collection is determined；Determine the item collection value of utility of the pending item collection；If the item collection value of utility of the pending item collection is not less than the minimum effectiveness threshold value of setting, and the item collection degree of correlation of the pending item collection is not less than predetermined minimum relevance threshold, it is determined that the pending item collection is the related effective item collection of data item.The embodiment of the present invention can realize the excavation of the related effective item collection of data item, lift the accuracy of the Result of effective item collection.
Description
Technical field
The present invention relates to technical field of data processing, and in particular to a kind of related effective item set mining side of data item
Method, device and data processing equipment.
Background technology
Transaction database is a kind of database that can record the affairs such as transaction, news, and transaction database has been usually noted
At least one affairs, every affairs include at least one data item, and are to characterize the association in transaction database between data item
Rule, at least one data item can be gathered to form an item collection again；Because the transaction database of type of transaction etc. can reflect use
The preference at family, therefore when to user's recommendation information, excavate in the multiple item collections often formed from transaction database to
The item collection that family is recommended；And during item collection is excavated, generally require to consider item collection (the abbreviation effective item that value of utility is higher
Collection).
Effective item collection is the higher item collection of value of utility, is by calculating each item collection at present when excavating effective item collection
Item collection value of utility, then the item collection value of utility of each item collection is compared with the minimum effectiveness threshold value set, so as to by item collection
Value of utility is more than or equal to the item collection of the minimum effectiveness threshold value of setting, as effective item collection.
The inventors found that：At present when excavating effective item collection, each data item in item collection is mainly considered
Value of utility accumulation result, the data item in this effective item collection that may cause to excavate is incoherent, influences Result
Accuracy；If the incoherent effective item collection of data item is after user is recommended, it is unfavorable for user in the affairs such as transaction
Guiding, in some instances it may even be possible to there is situation about misleading, influence recommend accuracy；
For example, (notebook, disc) and (notebook, toothpaste) is the effective item collection excavated based on item collection value of utility,
Because the value of utility of notebook is higher, therefore the item collection combined with notebook is probably effective item collection；But in (notes
This, disc) and (notebook, toothpaste) the two item collections in, be between notebook and the data item of disc it is related, and notebook with
It is incoherent between the data item of toothpaste, if recommending the item collection of (notebook, toothpaste), the data item that will cause in recommendation itemset
It is uncorrelated, it is unfavorable for guiding of the user in the affairs such as transaction；
Therefore, how to excavate data item is related effective item collection, lifts the accuracy of Result, becomes this
The problem of art personnel need to consider.
The content of the invention
In view of this, the embodiment of the present invention provides a kind of related effective item set mining method, apparatus of data item and place
Equipment is managed, to excavate the related effective item collection of data item, lifts the accuracy of the Result of effective item collection.
To achieve the above object, the embodiment of the present invention provides following technical scheme：
A kind of related effective item set mining method of data item, including：
Determine support of the pending item collection in transaction database, and the pending item collection each data item described
Support in transaction database；
Existed according to support of the pending item collection in transaction database, and each data item of the pending item collection
Support in the transaction database, determine the item collection degree of correlation of the pending item collection；
Determine the item collection value of utility of the pending item collection；
If the item collection value of utility of the pending item collection is not less than the minimum effectiveness threshold value of setting, and the pending item collection
The item collection degree of correlation be not less than predetermined minimum relevance threshold, it is determined that the pending item collection is related efficient of data item
Use item collection.
The embodiment of the present invention also provides a kind of data item related effective item set mining device, including：
Item collection support determining module, for determining support of the pending item collection in transaction database；
Item-support determining module, for determining each data item of the pending item collection in the transaction database
In support；
Degree of correlation determining module, for the support according to the pending item collection in transaction database, and described treat
Support of each data item of item collection in the transaction database is handled, determines the item collection degree of correlation of the pending item collection；
Item collection value of utility determining module, for determining the item collection value of utility of the pending item collection；
Related effective item collection determining module, if the item collection value of utility for the pending item collection is not less than setting most
It is poorly efficient to use threshold value, and the item collection degree of correlation of the pending item collection is not less than predetermined minimum relevance threshold, it is determined that it is described
Pending item collection is the related effective item collection of data item.
The embodiment of the present invention also provides a kind of data processing equipment, including the effective item that data item described above is related
Collect excavating gear.
Based on above-mentioned technical proposal, the related effective item set mining method bag of data item provided in an embodiment of the present invention
Include：Determine support of the pending item collection in transaction database, and the pending item collection each data item in the affairs
Support in database；According to support of the pending item collection in transaction database, and the pending item collection
Support of each data item in the transaction database, determine the item collection degree of correlation of the pending item collection；It is determined that described treat
Handle the item collection value of utility of item collection；If the item collection value of utility of the pending item collection is not less than the minimum effectiveness threshold value of setting, and
The item collection degree of correlation of the pending item collection is not less than predetermined minimum relevance threshold, it is determined that the pending item collection is number
According to the related effective item collection of item.It can be seen that the embodiment of the present invention is when excavating effective item collection, except the item by pending item collection
Collect minimum this index of effectiveness threshold value of value of utility not less than setting, weigh whether pending item collection is effective item collection, may be used also
Predetermined minimum relevance threshold is not less than by the item collection degree of correlation of pending item collection, weighs the data item phase of pending item collection
Guan Du, so as to excavate item collection value of utility not less than setting minimum effectiveness threshold value, and the item collection degree of correlation not less than it is predetermined most
The related effective item collection of the data item of low relevance threshold, realize the excavation of the related effective item collection of data item, lifting height
The accuracy of the Result of effectiveness item collection.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
The embodiment of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
The accompanying drawing of offer obtains other accompanying drawings.
Fig. 1 is the flow chart of the related effective item set mining method of data item provided in an embodiment of the present invention；
Fig. 2 is the structured flowchart of the related effective item set mining device of data item provided in an embodiment of the present invention；
Fig. 3 is the structured flowchart of item collection support determining module provided in an embodiment of the present invention；
Fig. 4 is the structured flowchart of Item-support determining module provided in an embodiment of the present invention；
Fig. 5 is the structured flowchart of degree of correlation determining module provided in an embodiment of the present invention；
Fig. 6 is the hardware block diagram of data processing equipment provided in an embodiment of the present invention.
Embodiment
For ease of understand the embodiment of the present invention description technical scheme, below first to the present embodiments relate to title it is general
Thought is introduced.
1st, affairs：A record in transaction database；For example what is recorded in the transaction database of type of transaction is commodity
Transaction record, then each affairs in transaction database can correspond to the transaction record of a commodity.
2nd, affairs numbering (English：TID)：The numbering of different affairs in transaction database；Optionally, affairs generally according to when
State serial number.
3rd, data item：The information project recorded in affairs, at least one data item is included in an affairs；Such as transaction
In the Transaction Information of type, the data item of the commodity comprising transaction in each affairs, and the inside value of utility of each commodity (are such as handed over
Easy quantity)；Number of transaction is a kind of avatar of the internal value of utility under scene of merchandising, in the transaction database of other scenes
In, the form of internal value of utility can adjust accordingly；
It is as shown in table 1 below, 10 affairs are included in the transaction database of type of transaction, every affairs indicate a transaction note
Record, the data item of the trade name comprising each transaction in every affairs, and number of transaction (the inside effect in affairs of each commodity
With a kind of form of value)；
| Affairs are numbered | Affairs (trade name：Number of transaction) |
| T1 | A:1,C:2,D:3 |
| T2 | A:2,D:1,E:2 |
| T3 | B:3,C:5 |
| T4 | A:1,C:3,D:1,E:2 |
| T5 | B:1,D:3,E:2 |
| T6 | B:2,D:2 |
| T7 | B:3,C:2,D:1,E:1 |
| T8 | A:2,C:3 |
| T9 | C:2,D:2,E:1 |
| T10 | A:2,C:2,D:1 |
Table 1
From table 1 it follows that in the transaction database of type of transaction, the data item in affairs can be trade name
Claim, internal value of utility can be the number of transaction of each commodity in affairs；In table 1, transaction database includes A, B, C, D and E this 5
Data item, wherein, the practical significance of T1 affairs can be：One instruction purchase, 1 A commodity, 2 C commodity and 3 D commodity
Transaction record；And the practical significance of T7 affairs can be：One instruction purchase, 3 B commodity, 2 C commodity, 1 D commodity and 1
The shopping record of E commodity；
And each affairs in News Field, table 1 can include at least one news, each affairs can record each newly
Interest value, the susceptibility size, freshness size etc. of news；In fields such as stocks, each affairs in table 1 can include at least one
Stock, each affairs can record the risk size of each stock, income size etc..
4th, item collection：The set that at least one data item is formed, for characterize in transaction database a kind of correlation rule；
The affairs point different from item collection is that affairs are typically that the note in transaction database of generation is triggered by the event of reality
Record, and item collection is typically what is gone out from database mining, might not there is actual implication.
5th, k- item collections：Include set of the k according to item；For example 1- item collections can include the item collection of a data item,
Item collection A as only included data item A；2- item collections can include the item collection of two data item, such as only include data item A and B item
Collect AB, by that analogy.
6th, outside value of utility table (such as profit flow table, Profit Table)：Record in transaction database corresponding to each data item
The form of unit external value of utility；In the transaction database of type of transaction, profit flow table can be one kind of outside value of utility table
Avatar, i.e., outside value of utility table can record the unit profit value of each data item in transaction database；Table 2 shows profit
Table, it can refer to；
| Data item | A | B | C | D | E |
| Unit profit value | 6 | 12 | 1 | 9 | 3 |
Table 2
From table 2 it can be seen that profit flow table expression is to sell the unit profit that a commodity can obtain, for example sell one
Part commodity A can earn a profit 6 yuan；Selling a commodity B can earn a profit 12 yuan；Accordingly, outside value of utility table can be with
Represent, unit external value of utility corresponding to each data item.
7th, value of utility (Utility of an item in a transaction) of the data item in affairs：One number
Can be that inside value of utility of a certain data item in an affairs is multiplied by the data item according to the value of utility in Xiang Yi bar affairs
Unit external value of utility；Such as in the transaction database of type of transaction, value of utility of a certain data item in an affairs can be,
Number of transaction of the data item in the affairs is multiplied by the unit profit value of the data item；So that shown in Tables 1 and 2, data item B exists
Value of utility in T3 affairs can be 3 × 12=36.
8th, value of utility (Utility of an itemset in a transaction) of the item collection in affairs：A certain item
Concentrate value of utility of each data item in a certain affairs plus and；So that shown in Tables 1 and 2, item collection BC is (only comprising data item B
With C item collection) value of utility in T3 affairs is 3 × 12+5 × 1=41.
9th, item collection value of utility (Itemset utility in Database)：Effect of a certain item collection in transaction database
With value, i.e., value of utility of a certain item collection in each affairs of all data item comprising the item collection plus and.
10th, the value of utility (Transaction Utility) of affairs：The value of utility of a certain affairs is to form the affairs
Value of utility of each data item in the affairs plus and；Shown in table 1, to include data item B, D and E in affairs T5, the present invention
Embodiment can determine that affairs T5 value of utility is 1 × 12+3 × 9+2 × 3=45.
11st, the total utility value of database：In database the value of utility of each affairs plus and；With shown in table 1, database it is total
The value of utility for each affairs that value of utility is T1 to T10 plus and be：
35+27+41+24+45+42+50+15+23+23=325.
12nd, the item collection degree of correlation：Correlation in item collection between each data item, all data of the item collection are combined for weighing
The degree of correlation between, is represented with Kulc values；
Optionally, the support that the item collection degree of correlation of a certain item collection can be according to the item collection in database, the item collection
Support of each data item in database determines；A kind of optional calculation formula can be as follows：
Wherein, Kulc (X) is the item collection X item collection degree of correlation, and k is included by item collection X
The number of data item, sup (X) are supports of the item collection X in database, and sup (j) is j-th of data item in item collection X in number
According to the support in storehouse；
Such as item collection ABC (item collection for including data item A, B and C), Kulc (ABC)=1/3* (sup (ABC)/
sup(A)+sup(ABC)/sup(B)+sup(ABC)/sup(C))；That is the determination process of the item collection degree of correlation of an item collection can be：
The quotient of support of support of the item collection in database respectively with each data item of the item collection in database is determined, will
Identified each quotient is added, obtain quotient plus and, by quotient plus and divided by item collection data item number, obtain the item collection
The item collection degree of correlation.
13rd, effective item collection (High Utility Itemset, HUI)：When item collection item collection value of utility >=setting most
Poorly efficient to use threshold value, then the item collection is effective item collection；
Optionally, the minimum effectiveness threshold value of setting can be pre-defined fixed minimum effectiveness threshold value；
Optionally, the embodiment of the present invention also can be corresponding minimum for the setting of each item collection according to data item different in item collection
Effectiveness threshold value；The minimum effectiveness threshold value table of definable of the embodiment of the present invention, it is as shown in table 3 below
| Data item | A | B | C | D | E |
| Minimum effectiveness threshold value | 56 | 65 | 53 | 50 | 70 |
Table 3
Minimum effectiveness threshold value table shown in table 3, instruction have minimum effectiveness threshold value corresponding to each data item, so as to a certain in setting
During the minimum effectiveness threshold value of item collection, the embodiment of the present invention can be by the minimum effectiveness threshold value corresponding to data item that the item collection includes
In, minimum minimum effectiveness threshold value is as the minimum effectiveness threshold value set by the item collection；For each item collection, the embodiment of the present invention can
The data item that minimum effectiveness threshold value is minimum in item collection is determined, the minimum effectiveness threshold value of identified data item is determined it is for this
The set minimum effectiveness threshold value of collection.
14th, the affairs weighting effectiveness (Transaction Weighted Utility, TWU) of item collection：Include a certain item collection
Affairs value of utility sum；Exemplified by shown in Tables 1 and 2, when specified item collection is B (item collection for only including data item B), then
Affairs comprising item collection B are T3, T5, T6 and T7, corresponding T3, T5, the value of utility of T6 and T7 affairs plus and be 41+45+42+
50=178, then item collection B affairs weighting effectiveness is 178.
15th, high affairs weighting effectiveness item collection (High Transaction Weighted Utilization Itemset,
HTWUI)：When TWU >=setting of item collection minimum effectiveness threshold value (uniquely fixed minimum effectiveness threshold value, or, for the item collection institute
The minimum effectiveness threshold value set) when, then the item collection is that high affairs weight effectiveness item collection；For example item collection B affairs weighting effectiveness is
178, and item collection B minimum effectiveness threshold value is 65, item collection B affairs weighting effectiveness is more than minimum effectiveness threshold value, determines that item collection B is
High affairs weight effectiveness item collection.
16th, the related effective item collection (Correlated High Utility Itemset, CoHUI) of data item：When certain
The minimum effectiveness threshold value of item collection value of utility >=setting of item collection, the minimum degree of correlation threshold of the and item collection degree of correlation of the item collection >=predetermined
Value, then the item collection is the related effective item collection of data item.
17th, related effective upper bound item collection (Correlated high utility upper-bound itemset,
CHUUBI)：When the minimum effectiveness threshold value of TWU >=setting of item collection, the minimum related threshold of and Kulc values of the item collection >=predetermined,
The item collection is referred to as the effective upper bound item collection of correlation.
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
Fig. 1 is the flow chart of the related effective item set mining method of data item provided in an embodiment of the present invention, this method
The data processing equipment with data-handling capacity is can be applied to, is such as applied to the data processing server of network side, optionally,
According to the difference of data mining scene, the excavation of the related effective item collection of data item is also likely to be in computer of user side etc.
Carried out in equipment；Reference picture 1, the related effective item set mining method of data item provided in an embodiment of the present invention can wrap
Include：
Step S100, support of the pending item collection in transaction database, and each number of the pending item collection are determined
According to support of the item in the transaction database；
Pending item collection can be any item collection for being excavated from the uncertain data storehouse of trade transactions type, an item collection
Including at least one data item；For each pending item collection, the embodiment of the present invention can determine that pending item collection in Transaction Information
Support in storehouse, and support of each data item in the transaction database of the pending item collection；
In data mining process, generally use support (support) quantifies the correlation rule between object, passes through branch
Degree of holding can reflect the serviceability for finding rule；
Optionally, in embodiments of the present invention, support of the item collection in transaction database can represent, comprising this
Probability of occurrence of the affairs of item collection in transaction database；The quantity and number of transactions of target transaction corresponding to the item collection can be passed through
Determine that target transaction corresponding to an item collection can include all data item of the item collection according to the ratio of total transactions in storehouse
Affairs, accordingly, target transaction set corresponding to an item collection refer to comprising the item collection all data item office's group
Into set；
Accordingly, it can determine that pending item collection is corresponding to determine support of the pending item collection in transaction database
Target transaction quantity, the ratio with total transactions in transaction database, identified ratio is defined as described waiting to locate
Support of the item collection in transaction database is managed, wherein, target transaction corresponding to the pending item collection is to wait to locate comprising described
Manage the affairs of all data item of item collection；
Support of the data item in transaction database in item collection can represent that the affairs comprising the data item exist
Probability of occurrence in transaction database；Total number of transactions in transactions and transaction database comprising the data item can be passed through
The ratio of amount determines；
Accordingly, determining support of each data item of the pending item collection in the transaction database can be,
For each data item of the pending item collection, it is determined that total affairs in transactions and transaction database comprising the data item
The ratio of quantity, identified ratio is defined as branch of the data item in the transaction database in the pending item collection
Degree of holding.
Step S110, the support according to the pending item collection in transaction database, and the pending item collection
Support of each data item in the transaction database, determine the item collection degree of correlation of the pending item collection；
The item collection degree of correlation of identified pending item collection, it can be used for weighing between the data item of the pending item collection
The degree of correlation.
Step S120, the item collection value of utility of the pending item collection is determined；
The item collection value of utility of pending item collection represents, value of utility of the item collection in transaction database, can pass through
Value of utility of the pending item collection in corresponding each target transaction plus and determine；
Value of utility of the pending item collection in a certain target transaction, each data item of pending item collection can be represented in the mesh
Mark affairs in value of utility plus and；
Value of utility of one data item in an affairs can be that inside value of utility of the data item in an affairs is multiplied by
The unit external value of utility of the data item.
If the item collection value of utility of step S130, described pending item collection is not less than the minimum effectiveness threshold value of setting, and described
The item collection degree of correlation of pending item collection is not less than predetermined minimum relevance threshold, it is determined that the pending item collection is data item
Related effective item collection.
Predetermined minimum relevance threshold can have phase between the data item of the item collection pre-defined according to priori
Guan Du degree of correlation lower limit；The item collection degree of correlation of item collection >=(being not less than) predetermined minimum relevance threshold, then can determine that
It is related between the data item of the item collection；
Optionally, the minimum effectiveness threshold value of setting can be the minimum effectiveness threshold value of predefined fixation, the item collection of item collection
The minimum effectiveness threshold value of value of utility >=(being not less than) predefined fixation, then can determine that the item collection is effective item collection；
Optionally, because minimum effectiveness threshold value corresponding to different pieces of information item is possible different (as shown in table 3), this causes difference
Minimum effectiveness threshold value is also likely to be different corresponding to item collection, therefore is the accuracy of effective item collection determined by lifting, this
Inventive embodiments are directed to each item collection, can be according to the data item included in item collection, for the minimum effectiveness threshold value of item collection setting adaptation；Tool
Body, for each item collection, the embodiment of the present invention can determine that the data item that minimum effectiveness threshold value is minimum in item collection, by identified number
The minimum effectiveness threshold value set according to the minimum effectiveness threshold value of item as the item collection；
Accordingly, the determination process for the minimum effectiveness threshold value of the corresponding setting of pending item collection can be：According to most poorly efficient
With threshold value table, the minimum effectiveness threshold value of each data item in the pending item collection is determined；By each data in the pending item collection
Minimum minimum effectiveness threshold value in the minimum effectiveness threshold value of item, it is defined as the minimum effectiveness threshold of the corresponding setting of the pending item collection
Value；
So as to be directed to each item collection, the minimum effectiveness threshold value of the item collection value of utility of item collection setting corresponding with the item collection is compared
Right, when the item collection value of utility of item collection corresponds to the minimum effectiveness threshold value of setting not less than the item collection, it is effective to determine the item collection
Item collection；
It can be seen that the minimum effectiveness threshold value that definable of the embodiment of the present invention is fixed, or, it is minimum for setting corresponding to items collection
Effectiveness threshold value, realize the setting of minimum effectiveness threshold value；
To realize the excavation of the related effective item collection of data item, for each pending item collection, the embodiment of the present invention can lead to
The item collection value of utility for crossing pending item collection judges whether pending item collection is effective item collection, and passes through the item collection of pending item collection
The degree of correlation judges whether the data item of pending item collection is related, so as to excavate the related effective item collection of data item.
The related effective item set mining method of data item provided in an embodiment of the present invention includes：Determine that pending item collection exists
Support in transaction database, and support of each data item in the transaction database of the pending item collection；Root
According to support of the pending item collection in transaction database, and the pending item collection each data item in the number of transactions
According to the support in storehouse, the item collection degree of correlation of the pending item collection is determined；Determine the item collection value of utility of the pending item collection；
If the item collection value of utility of the pending item collection is not less than the minimum effectiveness threshold value of setting, and the item collection phase of the pending item collection
Guan Du is not less than predetermined minimum relevance threshold, it is determined that the pending item collection is the related effective item collection of data item.
It can be seen that the embodiment of the present invention is when excavating effective item collection, except the item collection value of utility by pending item collection is not less than setting
Minimum this index of effectiveness threshold value, weigh whether pending item collection is effective item collection, can also pass through the item collection of pending item collection
The degree of correlation is not less than predetermined minimum relevance threshold, the data item degree of correlation of pending item collection is weighed, so as to excavate item collection
Value of utility is not less than the minimum effectiveness threshold value set, and the item collection degree of correlation is not less than the data item of predetermined minimum relevance threshold
Related effective item collection, the excavation of the related effective item collection of data item is realized, lift the Result of effective item collection
Accuracy.
Optionally, it is determined that support of the pending item collection in transaction database, and each number of the pending item collection
After support of the item in the transaction database, the embodiment of the present invention can be existed by equation below based on pending item collection
Support in transaction database, and support of each data item in the transaction database of the pending item collection, really
The item collection degree of correlation of the fixed pending item collection；
According to formulaThe pending item collection X item collection degree of correlation is determined, wherein, Kulc (X)
For the item collection X item collection degree of correlation, for k by the number of the item collection X data item included, sup (X) is branch of the item collection X in database
Degree of holding, sup (j) are support of j-th of data item in database in item collection X.
Optionally, support sups (X) of the item collection X in database can be the quantity of target transaction corresponding to item collection X
With the ratio of total transactions in transaction database；
Support sup (j) of j-th of data item in database in item collection X can include j-th of data item
Transactions and transaction database in total transactions ratio；
Accordingly, the support according to the pending item collection in transaction database, and the pending item collection are each
Support of the data item in the transaction database, determining the item collection degree of correlation of the pending item collection can be：Determine institute
Support of the pending item collection in transaction database is stated, each data item with the pending item collection is in database respectively
The quotient of support；Will determined by each quotient be added, obtain quotient plus and；By obtained quotient plus and divided by described treat
The data item number of item collection is handled, obtains the item collection degree of correlation of the pending item collection.
Optionally, the related effective item set mining method of data item provided in an embodiment of the present invention can be applied to network side
Data processing server, and source data used is excavated to data processing server transmission by terminal device, realizes data
The excavation of the related effective item collection of item；
By taking the grouping of commodities of the related effective item collection of display data item as an example, when user wishes to show on the terminal device
During the grouping of commodities of the related effective item collection of data item, user can be (outer by the unit profit value of each commodity by terminal device
A kind of form of portion's value of utility) data processing server is sent to, profit value represents to sell an available profit of commodity；
Meanwhile user can be sent to by terminal device by the transaction record in the historical period of preset duration before current point in time
Data processing server；The time of every transaction, the number of transaction of each data item and each transaction can be recorded in transaction record
The information such as the profit value of acquisition, after data processing server receives these transaction records, the Transaction Information of type of transaction can be formed
Storehouse；
So as to which each item collection formed for the data item in transaction database, data processing server are executable such as Fig. 1 institutes
Show method, realize the excavation of the related effective item collection of data item, and Result is recommended into terminal device；
Specifically, data processing server is directed to each item collection that grouping of commodities is formed in transaction database, it may be determined that item collection
Support in transaction database, and support of each commodity in transaction database of item collection, so that it is determined that going out item collection
The item collection degree of correlation；And by determining the item collection value of utility of item collection, by minimum effectiveness threshold value of the item collection value of utility not less than setting, and
The item collection degree of correlation is not less than the item collection of predetermined minimum relevance threshold, is defined as the related effective item collection of data item.
Optionally, the embodiment of the present invention can be based on pseudo- projection (projection) and ordering techniques, realize data item correlation
Effective item collection excavation；Projected based on puppet in (projection) and the mining process of ordering techniques, the item of each item collection
The sequence of the collection degree of correlation has downward closure property (sorted downward closure property of Kulc)；
Such as assume sortord that each data item in every affairs of database uses for：Support according to data item
Spend size and carry out ascending sort, i.e., sorted from small to large according to the support of data item, then the item collection phase for the item collection being calculated
The downward closure property based on sequence be present in Guan Du；
As it is assumed that data item a1,a2,...,ak,ak+1Sortord be based on support size carry out ascending order row
Sequence, i.e. sup (a1)≤sup(a2)≤...≤sup(ak)≤sup(ak+1)；Following relation then be present：
To sum up, it may be determined that if 1- item collections are ranked up from small to large according to the size of support, the item collection of 1- item collections
There is downward closure property in the degree of correlation, i.e., when certain (K-1) item collection is unsatisfactory for the minimum relevance threshold of the item collection degree of correlation >=predetermined
When, all supersets (K- item collections) based on (K-1) item collection are unlikely to be the related effective item collection of data item, and reason is
The item collection relevance degree for being somebody's turn to do (K-1) item collection X all supersets (K- item collections) is respectively less than predetermined minimum relevance threshold；One item
The superset of collection includes all data item and other data item of the item collection；Accordingly, in the database of this example table 1, because { sup
(A)=5；Sup (B)=4；Sup (C)=7；Sup (D)=8；Sup (E)=5 }, so the support size according to data item is entered
Row ascending sort, it is as a result B ＜ A ＜ E ＜ C ＜ D (B ＜ A represent B sequences before A).
Optionally, the embodiment of the present invention can first determine the related effective item collection of the data item comprising a data item
(i.e. the related effective 1- item collections of data item), obtains set CHUUBI1(in the related effective i.e. comprising a data item
Boundary's item collection) and CoHUI1(the effective item collection of the data item correlation i.e. comprising a data item)；It is then based on pseudo- projection
(projection) technology is one by one to including the related effective upper bound item collection CHUUBI of a data item1Handled, root
According to outside value of utility corresponding to the data item included in the related effective upper bound 1- item collections of each data item, each data item phase is calculated
Affairs weighting value of utility TWU and the item collection of the effective 1- item collections of pass in subdata base is projected degree of correlation Kulc values；Tool
Gymnastics is made shown in following algorithm 1；The word of following algorithm heels can be the literal interpretation to respective code；
It can be seen that the information that the scan database first of algorithm 1 carries out each 1- item collections calculates, including each item collection is in database
Support, and each item collection affairs weighting effectiveness, then find out correlation effective upper bound 1- item collections CHUUBI1And data item
Related effective item collection 1- item collections CoHUI1(Lines 2-7)。
In force, the embodiment of the present invention can determine putting in order for each data item in database, the embodiment of the present invention
Each data item in database can be ranked up at random, each data item in database can also be arranged by calculating
Sequence；Specifically, as shown in Line 8, the embodiment of the present invention can use the ascending order order support of the support of 1- item collections
Ascending order, i.e. set CHUUBI1In each item be ranked up from small to large by their support size.
Afterwards, the embodiment of the present invention can iteratively call function Project-CHUUBI (ij,db|ij, k), constantly it is based on
Projection technology minings go out all extension item collections using each data item as prefix, by it is each extension item collection with excavate order according to
Secondary is defined as pending item collection, determines the set CoHUI of the related effective k- item collections of data itemk, and related effective
The set CHUUBI of upper bound k- item collectionsk；
Finally, as the complete set CHUUBI of algorithm process1In all 1- item collections when, return related in whole database
The set CHUUBIs of effective upper bound item collection；This model scans raw data base (Line 12) for the second time, calculates set
The actual utility value (Lines13-15) of each upper bound item collection in CHUUBIs, actual utility value is found out more than or equal to minimum
The item collection of effectiveness threshold value, finally return to the related effective item collection (Line 16) of data item all in whole database.
Project-CHUUBI(ij,db|ij, k) algorithm can be as follows shown in algorithm 2；
Optionally, the embodiment of the present invention is before subdata base projection operation is carried out, and can use a kind of based on sequence
The downward closure of Kulc values (Sorted downward closure property) is judged in advance, when certain (K-1) item collection
When being unsatisfactory for the minimum relevance threshold of the item collection degree of correlation >=predetermined, the superset of the item collection is filtered.
The embodiment of the present invention can excavate the related effective item set mining of data item, lift the excavation knot of effective item collection
The accuracy of fruit.
The effective item set mining device related to data item provided in an embodiment of the present invention is introduced below, hereafter retouches
The related effective item set mining device of the data item stated can be related to above-described data item effective item set mining side
Method is mutually to should refer to.
Fig. 2 is the structured flowchart of the related effective item set mining device of data item provided in an embodiment of the present invention, reference
Fig. 2, the device can include：
Item collection support determining module 100, for determining support of the pending item collection in transaction database；
Item-support determining module 200, for determining each data item of the pending item collection in the number of transactions
According to the support in storehouse；
Degree of correlation determining module 300, for the support according to the pending item collection in transaction database, and it is described
Support of each data item of pending item collection in the transaction database, determine that the item collection of the pending item collection is related
Degree；
Item collection value of utility determining module 400, for determining the item collection value of utility of the pending item collection；
Related effective item collection determining module 500, if the item collection value of utility for the pending item collection is not less than setting
Minimum effectiveness threshold value, and the item collection degree of correlation of the pending item collection is not less than predetermined minimum relevance threshold, it is determined that
The pending item collection is the related effective item collection of data item.
Optionally, Fig. 3 shows the alternative construction of item collection support determining module 100 provided in an embodiment of the present invention, ginseng
According to Fig. 3, item collection support determining module 100 can include：
First ratio determining unit 110, for determining the quantity of target transaction corresponding to pending item collection, with Transaction Information
The ratio of total transactions in storehouse, identified ratio is defined as support of the pending item collection in transaction database
Degree, wherein, target transaction corresponding to the pending item collection is the affairs of all data item comprising the pending item collection.
Optionally, Fig. 4 shows the alternative construction of Item-support determining module 200 provided in an embodiment of the present invention,
Reference picture 4, the Item-support determining module 200 can include：
Second ratio determining unit 210, for each data item for the pending item collection, it is determined that including the data item
Transactions and transaction database in total transactions ratio, identified ratio is defined as the pending item collection
In support of the data item in the transaction database.
Optionally, Fig. 5 shows the alternative construction of degree of correlation determining module 300 provided in an embodiment of the present invention, reference picture
5, degree of correlation determining module 300 can include：
Quotient determining unit 310, for determining support of the pending item collection in transaction database, respectively with institute
State the quotient of support of each data item of pending item collection in database；
Quotient add with determining unit 320, for will determined by each quotient be added, obtain quotient plus and；
Divided by unit 330, for by obtained quotient plus and divided by the pending item collection data item number, obtain
The item collection degree of correlation of the pending item collection.
Optionally, item collection value of utility determining module 400 is particularly used in, and determines the pending item collection corresponding each
Value of utility in target transaction；The addition of value of utility of the pending item collection in corresponding each target transaction and, is obtained
To the item collection value of utility of the pending item collection；Wherein, value of utility of the pending item collection in a target transaction represents, institute
State value of utility of each data item of pending item collection in the target transaction plus and；One data item of the pending item collection
Value of utility in a target transaction represents that inside value of utility of the data item in the target transaction is multiplied by the list of the data item
The outside value of utility in position.
Optionally, the related effective item set mining device of data item it is determined that setting minimum effectiveness threshold value when, can use
According to minimum effectiveness threshold value table, determining the minimum effectiveness threshold value of each data item in the pending item collection, the minimum effectiveness
Threshold value table record has minimum effectiveness threshold value corresponding to each data item；By the minimum effectiveness threshold of each data item in the pending item collection
Minimum minimum effectiveness threshold value in value, it is defined as the minimum effectiveness threshold value of the corresponding setting of the pending item collection.
Optionally, the related effective item set mining device of data item can be additionally used in, if including the item collection of a data item
Sorted from small to large according to support, and the item collection degree of correlation of an item collection is less than predetermined minimum relevance threshold, it is determined that
The superset of the item collection is not the related effective item collection of data item；Wherein, the superset of an item collection includes all of the item collection
Data item and other data item.
Optionally, the related effective item set mining device of data item can be additionally used in, and a data item is included excavating
Related effective upper bound 1- item collections after, all expansions using each upper bound 1- item collections as prefix are excavated based on pseudo- shadow casting technique
Exhibition item collection, by each extension item collection to excavate order, i.e., sequentially, being defined as successively is pending for the support ascending order of each data item
Item collection.
The embodiment of the present invention is also provided with a kind of data processing equipment, and the data processing equipment can include described above
The related effective item set mining device of data item.
Optionally, Fig. 6 shows the hardware block diagram of data processing equipment, reference picture 6, and the data processing equipment can be with
Including：Processor 1, communication interface 2, memory 3 and communication bus 4；
Wherein processor 1, communication interface 2, memory 3 complete mutual communication by communication bus 4；
Optionally, communication interface 2 can be the interface of communication module, such as the interface of gsm module；
Processor 1, for configuration processor；
Memory 3, for depositing program；
Program can include program code, and described program code includes computer-managed instruction.
Processor 1 is probably a central processor CPU, or specific integrated circuit ASIC (Application
Specific Integrated Circuit), or it is arranged to implement the integrated electricity of one or more of the embodiment of the present invention
Road.
Memory 3 may include high-speed RAM memory, it is also possible to also including nonvolatile memory (non-volatile
Memory), a for example, at least magnetic disk storage.
Wherein, program can be specifically used for：
Determine support of the pending item collection in transaction database, and the pending item collection each data item described
Support in transaction database；
Existed according to support of the pending item collection in transaction database, and each data item of the pending item collection
Support in the transaction database, determine the item collection degree of correlation of the pending item collection；
Determine the item collection value of utility of the pending item collection；
If the item collection value of utility of the pending item collection is not less than the minimum effectiveness threshold value of setting, and the pending item collection
The item collection degree of correlation be not less than predetermined minimum relevance threshold, it is determined that the pending item collection is related efficient of data item
Use item collection.
Each embodiment is described by the way of progressive in this specification, what each embodiment stressed be and other
The difference of embodiment, between each embodiment identical similar portion mutually referring to.For device disclosed in embodiment
For, because it is corresponded to the method disclosed in Example, so description is fairly simple, related part is said referring to method part
It is bright.
Professional further appreciates that, with reference to the unit of each example of the embodiments described herein description
And algorithm steps, can be realized with electronic hardware, computer software or the combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software, the composition and step of each example are generally described according to function in the above description.These
Function is performed with hardware or software mode actually, application-specific and design constraint depending on technical scheme.Specialty
Technical staff can realize described function using distinct methods to each specific application, but this realization should not
Think beyond the scope of this invention.
Directly it can be held with reference to the step of method or algorithm that the embodiments described herein describes with hardware, processor
Capable software module, or the two combination are implemented.Software module can be placed in random access memory (RAM), internal memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (11)
1. a kind of related effective item set mining method of data item, it is characterised in that including：
Determine support of the pending item collection in transaction database, and the pending item collection each data item in the affairs
Support in database；
According to support of the pending item collection in transaction database, and the pending item collection each data item described
Support in transaction database, determine the item collection degree of correlation of the pending item collection；
Determine the item collection value of utility of the pending item collection；
If the item collection value of utility of the pending item collection is not less than the minimum effectiveness threshold value of setting, and the item of the pending item collection
The collection degree of correlation is not less than predetermined minimum relevance threshold, it is determined that the pending item collection is the related effective item of data item
Collection.
2. the related effective item set mining method of data item according to claim 1, it is characterised in that the determination is treated
Handling support of the item collection in transaction database includes：
Determine the quantity of target transaction corresponding to pending item collection, the ratio with total transactions in transaction database, by really
Fixed ratio is defined as support of the pending item collection in transaction database, wherein, corresponding to the pending item collection
Target transaction is the affairs of all data item comprising the pending item collection；
Support of each data item for determining the pending item collection in the transaction database includes：
For each data item of the pending item collection, it is determined that total in transactions and transaction database comprising the data item
The ratio of transactions, identified ratio is defined as in the pending item collection data item in the transaction database
Support.
3. the related effective item set mining method of data item according to claim 1 or 2, it is characterised in that described
According to support of the pending item collection in transaction database, and the pending item collection each data item in the number of transactions
According to the support in storehouse, determining the item collection degree of correlation of the pending item collection includes：
Support of the pending item collection in transaction database is determined, each data item with the pending item collection exists respectively
The quotient of support in database；
Will determined by each quotient be added, obtain quotient plus and；
By obtained quotient plus and divided by the pending item collection data item number, obtain the item collection of the pending item collection
The degree of correlation.
4. the related effective item set mining method of data item according to claim 1, it is characterised in that the determination institute
Stating the item collection value of utility of pending item collection includes：
Determine value of utility of the pending item collection in corresponding each target transaction；Wherein, the pending item collection is one
Value of utility in target transaction represents, value of utility in the target transaction of each data item of the pending item collection plus and；
Value of utility of one data item of the pending item collection in a target transaction represents that the data item is in the target transaction
Internal value of utility is multiplied by the unit external value of utility of the data item；
The addition of value of utility of the pending item collection in corresponding each target transaction and, is obtained into the pending item collection
Item collection value of utility.
5. the related effective item set mining method of data item according to claim 4, it is characterised in that the setting
The determination process of minimum effectiveness threshold value includes：
According to minimum effectiveness threshold value table, the minimum effectiveness threshold value of each data item in the pending item collection is determined, it is described most poorly efficient
There is minimum effectiveness threshold value corresponding to each data item with threshold value table record；
By the minimum minimum effectiveness threshold value in the minimum effectiveness threshold value of each data item in the pending item collection, it is defined as described treat
The minimum effectiveness threshold value of the corresponding setting of processing item collection.
6. the related effective item set mining method of data item according to claim 1, it is characterised in that methods described is also
Including：
If the item collection comprising a data item sorts from small to large according to support, and the item collection degree of correlation of an item collection is less than in advance
Fixed minimum relevance threshold, it is determined that the superset of the item collection is not the related effective item collection of data item；Wherein, an item
The superset of collection includes all data item of the item collection.
7. the related effective item set mining method of data item according to claim 1, it is characterised in that methods described is also
Including：
After the related effective upper bound item collection comprising a data item is excavated, excavated based on pseudo- shadow casting technique with each
Related effective upper bound item collection comprising a data item is all extension item collections of prefix, and each extension item collection is suitable to excavate
Sequence successively be defined as pending item collection；Wherein, when item collection affairs weighting effectiveness not less than setting minimum effectiveness threshold value, and
The item collection degree of correlation of the item collection is not less than predetermined minimum relevance threshold, then the item collection is related effective upper bound item collection.
A kind of 8. related effective item set mining device of data item, it is characterised in that including：
Item collection support determining module, for determining support of the pending item collection in transaction database；
Item-support determining module, for determining each data item of the pending item collection in the transaction database
Support；
Degree of correlation determining module, for the support according to the pending item collection in transaction database, and it is described pending
Support of each data item of item collection in the transaction database, determine the item collection degree of correlation of the pending item collection；
Item collection value of utility determining module, for determining the item collection value of utility of the pending item collection；
Related effective item collection determining module, if the item collection value of utility for the pending item collection is most poorly efficient not less than setting
With threshold value, and the item collection degree of correlation of the pending item collection is not less than predetermined minimum relevance threshold, it is determined that described to wait to locate
It is the related effective item collection of data item to manage item collection.
9. the related effective item set mining device of data item according to claim 8, it is characterised in that the item collection branch
Degree of holding determining module includes：
First ratio determining unit, it is and total in transaction database for determining the quantity of target transaction corresponding to pending item collection
The ratio of transactions, identified ratio is defined as support of the pending item collection in transaction database, wherein,
Target transaction corresponding to the pending item collection is the affairs of all data item comprising the pending item collection；
The Item-support determining module includes：
Second ratio determining unit, for each data item for the pending item collection, it is determined that the affairs comprising the data item
The ratio of quantity and total transactions in transaction database, is defined as the number in the pending item collection by identified ratio
According to support of the item in the transaction database.
10. the related effective item set mining device of data item according to claim 8 or claim 9, it is characterised in that the phase
Pass degree determining module includes：
Quotient determining unit, for determining support of the pending item collection in transaction database, wait to locate with described respectively
Manage the quotient of support of each data item of item collection in database；
Quotient adds and determining unit, for will determined by each quotient be added, obtain quotient plus and；
Divided by unit, for by obtained quotient plus and divided by the pending item collection data item number, obtain described treat
Handle the item collection degree of correlation of item collection.
11. a kind of data processing equipment, it is characterised in that including the related height of the data item described in claim any one of 8-10
Effectiveness item set mining device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610854714.8A CN107870936A (en) | 2016-09-27 | 2016-09-27 | The related effective item set mining method, apparatus of data item and data processing equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610854714.8A CN107870936A (en) | 2016-09-27 | 2016-09-27 | The related effective item set mining method, apparatus of data item and data processing equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107870936A true CN107870936A (en) | 2018-04-03 |
Family
ID=61752082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610854714.8A Pending CN107870936A (en) | 2016-09-27 | 2016-09-27 | The related effective item set mining method, apparatus of data item and data processing equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107870936A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408563A (en) * | 2018-11-07 | 2019-03-01 | 哈尔滨工业大学（深圳） | High average utility item set mining method, apparatus and computer equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5794209A (en) * | 1995-03-31 | 1998-08-11 | International Business Machines Corporation | System and method for quickly mining association rules in databases |
| CN102789615A (en) * | 2011-05-17 | 2012-11-21 | 阿里巴巴集团控股有限公司 | Book information correlation recommendation method, server and system |
| CN103020256A (en) * | 2012-12-21 | 2013-04-03 | 电子科技大学 | Association rule mining method of large-scale data |
| CN103093369A (en) * | 2011-11-03 | 2013-05-08 | 阿里巴巴集团控股有限公司 | Method and device for offering matched product based on correlation degree between products |
| CN105608182A (en) * | 2015-12-23 | 2016-05-25 | 一兰云联科技股份有限公司 | Uncertain data model oriented utility item set mining method |
-
2016
- 2016-09-27 CN CN201610854714.8A patent/CN107870936A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5794209A (en) * | 1995-03-31 | 1998-08-11 | International Business Machines Corporation | System and method for quickly mining association rules in databases |
| CN102789615A (en) * | 2011-05-17 | 2012-11-21 | 阿里巴巴集团控股有限公司 | Book information correlation recommendation method, server and system |
| CN103093369A (en) * | 2011-11-03 | 2013-05-08 | 阿里巴巴集团控股有限公司 | Method and device for offering matched product based on correlation degree between products |
| CN103020256A (en) * | 2012-12-21 | 2013-04-03 | 电子科技大学 | Association rule mining method of large-scale data |
| CN105608182A (en) * | 2015-12-23 | 2016-05-25 | 一兰云联科技股份有限公司 | Uncertain data model oriented utility item set mining method |
Non-Patent Citations (3)
| Title |
|---|
| 王敬华，罗相洲，吴倩: "基于投影的高效用项集挖掘算法", 《小型微型计算机系统》 * |
| 王立俊: "基于多重最小支持度的高效用频繁项集挖掘算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王金: "基于多最小支持度的关联规则挖掘研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408563A (en) * | 2018-11-07 | 2019-03-01 | 哈尔滨工业大学（深圳） | High average utility item set mining method, apparatus and computer equipment |
| CN109408563B (en) * | 2018-11-07 | 2021-06-22 | 哈尔滨工业大学（深圳） | High average utility item set mining method and device and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200218737A1 (en) | Method, system and program product for matching of transaction records | |
| JP2001522495A (en) | How to Aggregate a Dynamically Growing Database Data Set | |
| JPH10232804A (en) | Method and device for executing assembly collation for data base system | |
| CN109961357A (en) | User data processing method, device, electronic equipment and storage medium | |
| CN103646341B (en) | A kind of website provides the recommendation method and apparatus of object | |
| CN106789338B (en) | Method for discovering key people in dynamic large-scale social network | |
| CN107870956A (en) | A kind of effective item set mining method, apparatus and data processing equipment | |
| US20100239084A1 (en) | Memorable Telephone Number Identification | |
| CN104239335B (en) | User-specific information acquisition methods and device | |
| CN107436914A (en) | Recommend method and device | |
| CN108256113A (en) | The method for digging and device of data genetic connection | |
| CN106408154A (en) | Method and apparatus for processing workload information | |
| JP4914164B2 (en) | Business information management system, business information management method, and business information management program | |
| CN110275903A (en) | Improve the method and system of the feature formation efficiency of machine learning sample | |
| CN104573132B (en) | Song lookup method and device | |
| CN109918385A (en) | Tripartite's account checking method, electronic device and readable storage medium storing program for executing | |
| CN107870936A (en) | The related effective item set mining method, apparatus of data item and data processing equipment | |
| CN107870913A (en) | The high of effective time it is expected weight item collection method for digging, device and processing equipment | |
| CN110717318A (en) | Intent-driven contextualized and collaborative-intent-adapted content-populating system | |
| CN109636437A (en) | Cell average price predictor method, electronic device and storage medium | |
| CN106445975B (en) | Item set mining method and device | |
| US20050125346A1 (en) | Method and system for calculating and presenting bankruptcy preference payment defenses | |
| JP3131142B2 (en) | Map data linkage system | |
| CN103425688B (en) | Method and server for providing information | |
| CN108052595A (en) | A kind of processing method and processing device of data query |
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 |
Application publication date: 20180403 |
|
| RJ01 | Rejection of invention patent application after publication |