CN110096629B - Memory optimization method for transaction processing - Google Patents
Memory optimization method for transaction processing Download PDFInfo
- Publication number
- CN110096629B CN110096629B CN201910401314.5A CN201910401314A CN110096629B CN 110096629 B CN110096629 B CN 110096629B CN 201910401314 A CN201910401314 A CN 201910401314A CN 110096629 B CN110096629 B CN 110096629B
- Authority
- CN
- China
- Prior art keywords
- item
- weight
- transaction
- sets
- minws
- 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.)
- Active
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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for mining frequent item sets based on an effective weighted tree, which comprises the following steps: s1: reading a database D from a network, wherein the database D comprises N transactions, each transaction comprises different items, the number of the items and the weight w occupied by each item; s2: calculating the transaction weight tw of each transaction in the database D, and generating a transaction weight table; s3: calculating the weight W of the item set S, and presetting a threshold minws; if W is more than or equal to minws, the item set S is a frequent item set; if W is less than minws, the item set S is an infrequent item set; s4: an efficient weighted tree model is constructed for mining frequent item sets. The invention constructs the effective weighted tree model by giving weights to the item sets and calculating the weights, improves the mining efficiency of frequent item sets and reduces the memory application, and is suitable for databases of big data.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method for mining frequent weighting item sets based on an effective weighting tree.
Background
With the rapid development of computer technology, humans have entered the big data age. The processing, analysis and use of data are becoming increasingly important, and data analysis and data mining are becoming the leading research areas in the era. Data mining is a key component of the knowledge discovery field. As the name suggests, data mining is the acquisition of knowledge from big data.
Association rule mining (ARM, association rule mining) plays an important role in the field of data mining. ARM is used for relationship identification among items in a transaction database, and mining of frequent item sets plays an important role therein. Mining of frequent item sets is one of the key technologies for data mining.
In classical association rule mining, we typically use a support threshold to identify whether a set of items is frequently used. In many practical applications, however, such as shopping basket data analysis, web page prefetching, cross shopping, and network intrusion detection, the support threshold does not find the item sets we want because it only considers the frequency of the item sets and does not consider other implicit factors such as their number, interest, risk, or profit.
CH Cai, AWC Fu, CH Cheng and WW Kwong propose a model describing the concept of Weighted Association Rules (WAR) and an algorithm based on a priori mining Frequent Weighted Item Sets (FWIs). In 2009, B le, H iguyen and B Vo proposed the WIT-tree method to mine frequently weighted item sets, but each of them frequently weighted item sets is transacted, so the memory usage of the algorithm is high. In 2013, B Vo, F Coenen and B Le proposed the WIT-diff algorithm, and a difference set policy was adopted to reduce the memory usage and speed up the process, but in databases with larger project amounts, the operation failed due to memory overflow, so that frequent weighted term sets could not be obtained.
Disclosure of Invention
Aiming at the problem of high algorithm memory use rate in the prior art, the invention provides a method for mining frequent weighted item sets based on an effective weighted tree (EWT, effective weighted tree), which is used for mining FWI in databases with large item quantity (the number of the frequent item sets exceeds 30) through a weighted support threshold value, so as to reduce the memory use rate.
In order to achieve the above object, the present invention provides the following technical solutions:
a method of mining frequent item sets based on an effective weighted tree, comprising the steps of:
s1: reading a database D from a network, wherein the database D comprises N transactions, each transaction comprises different items, the number of the items and the weight w occupied by each item;
s2: calculating the transaction weight tw of each transaction in the database D, and generating a transaction weight table;
s3: calculating the weight W of the item set S, and presetting a threshold minws; if W is more than or equal to minws, the item set S is a frequent item set; if W is less than minws, the item set S is an infrequent item set;
s4: an efficient weighted tree model is constructed for mining frequent item sets.
Preferably, in the step S2, the calculation formula of the transaction weight tw is:
in the formula (1), t k Represents the kth transaction, tw (t k ) Representing transaction t k Transaction weight, I ki Representing the ith item in the kth transaction,representing item I ki Quantity of->Representation I ki N represents the number of items.
Preferably, in the step S3, the calculation formula of the weight W is:
in the formula (2), W represents the weight of the term set S, t j Representing the intersection transaction to which the item in the item set S belongs, tw (t j ) Representing transaction t j Is used to determine the transaction weight of (1),representing the sum of the transaction weights of the intersection transactions to which the items in the item set S belong,representing the sum of all transaction weights.
Preferably, the step S4 includes the steps of:
s4-1: calculating the weight of a 1-time item set in the database D, and taking the 1-time item set corresponding to the weight not smaller than the minimum weight support threshold minws as a 1-time node;
s4-2: and calculating the weight of m times of item sets in the database D, and constructing child nodes of 1 time of nodes, thereby forming an effective weighted tree model.
Preferably, the method for constructing the child node of the 1 st node is as follows:
a. firstly, obtaining 2 times of item sets by respectively merging the 1 times of item sets in the first layer, and calculating the weight of the 2 times of item sets through a formula (2); taking the 2 times item set corresponding to the weight not smaller than the minimum weight support threshold minws as a child node of the 1 times node, namely the 2 times node;
b. and (c) recursively calling the step a until the weight of the obtained m times of item sets is smaller than a minimum weight support threshold minws, and constructing an effective weighted tree model, namely indicating that the frequent item sets in the database D are completely mined.
In summary, due to the adoption of the technical scheme, compared with the prior art, the invention has at least the following beneficial effects:
according to the invention, the attribute of the item in the transaction, which most represents the importance of the item, is selected as the preset weight, the item table of the recorded item and the transaction containing the item and the weight table of the recorded transaction weight can be obtained by only scanning the database once, and the item set weight is rapidly calculated, so that the EWT tree is established for frequent item set mining. In the database with larger project quantity, the running memory is reduced, the problem that the database cannot be used on a plurality of application programs due to heap overflow is solved, and meanwhile, frequent project sets can be quickly and effectively mined.
Description of the drawings:
fig. 1 is a flow chart of a method for mining frequently weighted term sets based on an efficient weighted tree according to an exemplary embodiment of the present invention.
Fig. 2 is a schematic diagram of an efficient weighted tree model structure according to an exemplary embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples and embodiments. It should not be construed that the scope of the above subject matter of the present invention is limited to the following embodiments, and all techniques realized based on the present invention are within the scope of the present invention.
Fig. 1 is a method for mining frequent weight term sets based on an effective weighted tree according to an exemplary embodiment of the present invention, specifically including the following steps:
s1: a database D is read from the network, the database D containing N transactions, each transaction containing a different item, the number of items and the weight w occupied by each item.
In the embodiment, a database D is read from a network, wherein the database D comprises N transactions, and N is more than or equal to 1 and is a positive integer; each transaction contains a different item and the weight of the item. The same item may belong to different transactions and a collection of multiple different items is referred to as an item set. The set containing only one item is a 1-time item set, the set containing m items is a m-time item set, m is more than or equal to 2, and m is an integer.
For example, database D contains 5 transactions numbered 1,2,3,4,5; and the items in database D are I 1 ,I 2 ,I 3 ,I 4 ,I 5 Table 1 is a weight table of items, the weight w represents the importance of the item in the database, for example in the mark database, a commodity represents a corresponding item, and the weight represents the profit of the commodity. Table 2 is a transaction item quantity table, the item quantity representing the quantity of an item in a transaction, which can be represented by C, and in the mark database, the item quantity representing the quantity of merchandise.
Table 1 weight table of items
| Project (I) | Weight (w) |
| I 1 | 0.2 |
| I 2 | 0.5 |
| I 3 | 0.4 |
| I 4 | 0.1 |
| I 5 | 0.6 |
Table 2 transaction item quantity table
In this embodiment, it can be seen from Table 2 that the same item can belong to different transactions, e.g. item I 1 And belongs to the transaction 1, the transaction 3, the transaction 4 and the transaction 5, so that the item table of the item in the database D, namely the item table, such as the table 3, can be obtained.
Table 3 item table
| Project (I) | Transaction number (t) k ) |
| I 1 | 1,3,4,5 |
| I 2 | 2,3,4 |
| I 3 | 1,2,3 |
| I 4 | 1,4 |
| I 5 | 2,4,5 |
S2: the transaction weight tw of each transaction in the database D is calculated, and a transaction weight table is generated.
In this embodiment, the set of all the items in the database D is i= { I 1 ,I 2 ,I 3 ,…,I n N represents n items in database D, I n Representing the nth item, each transaction contains a different item and the weight that the item occupies. The calculation formula of the weight occupied by each transaction in the database D is:
in the formula (1), t k Represents the kth transaction, tw (t k ) Representing transaction t k Transaction weight, I ki Representing the ith item in the kth transaction,representing item I ki Quantity of->Representation I ki Is a weight of (2).
For example, transaction t with number 1 can be obtained from tables 1 and 2 1 The weight of (2) is:
tw(t 1 )＝2*0.2+1*0.4+1*0.1＝0.9，
then analogize to a transaction weight table.
Table 4 transaction weight table
| Transaction number (t) k ) | Transaction weight (tw) |
| 1 | 0.9 |
| 2 | 4.7 |
| 3 | 2.2 |
| 4 | 3.5 |
| 5 | 1.2 |
S3: calculating the weight W of the item set S, and comparing the weight W with a preset minimum weight support threshold minws; if W is more than or equal to minws, the item set S is a frequent item set; if W < minws, then item set S is an infrequent item set.
In this embodiment, the set of transactions in database D is t= { T 1 ,t 2 ,…,t k ,…,t n The number of the sets T is N, T k Representing the kth transaction. The item sets are composed of different items, and the set containing only one item is 1 item set, and the set containing m items is m item sets. The items in the item set S belong to different transactions t k The weight W of the item set S is the ratio of the sum of the transaction weights of the intersection transactions to which the item belongs to the sum of all the transaction weights, i.e
In the formula (2), W represents the weight of the term set S, t j Representing the intersection transaction to which the item in the item set S belongs, tw (t j ) Representing transaction t j Is used to determine the transaction weight of (1),representing the sum of the transaction weights of the intersection transactions to which the items in the item set S belong,representing the sum of all transaction weights.
For example, in this embodiment, the item set S is { I } 1 ，I 2 As can be seen from Table 3, item I 1 The belonged transactions are transactions with numbers 1,3,4 and 5; item I 2 The belonged transaction is the transaction with the numbers of 2,3 and 4; item I 1 And I 2 Intersection of the belonging transactions is transaction 3 and transaction 4, then
In this embodiment, the minimum weight supports a threshold minws=0.4.
S4: an efficient weighted tree model is constructed for mining frequent item sets.
In this embodiment, the term sets are composed of different terms, and may be divided into 1 term set and m (m is greater than or equal to 2, and m is a positive integer) term sets, but not every term set is a frequent term set, so that the invention facilitates visual display by constructing an effective weighted tree (EWT, effective weighted tree) model for mining frequent term sets in the database D.
S4-1: and calculating the weight of the 1 st item set, and constructing 1 st node to form a first layer.
In this embodiment, 1-time item sets are item sets including only one item, such as item set { I } 1 }，{I 2 }. The invention calculates the weights of all the 1 st-degree term sets through the formula (2). In this embodiment, a 1 st item set corresponding to a weight equal to or greater than a minimum weight support threshold minws is used as a 1 st node; and removing the 1-time item set corresponding to the weight smaller than the minimum weight support threshold minws, namely not taking the 1-time item set as the 1-time node, so that the running process can be quickened, and the memory usage amount is reduced.
In this embodiment, the minimum weight supports a threshold minws=0.4.
In this embodiment, the relationship between the 1 st node of the first layer is a sibling node, the relationship between the adjacent 1 st node is an adjacent sibling node, for example, the first 1 st node and the second 1 st node are adjacent sibling nodes.
S4-2: and calculating the weight of the m times of item sets, and constructing child nodes, thereby forming an effective weighted tree model.
In this embodiment, the m-degree item set is an item set including a plurality of different items, e.g., { I } 1 ，I 2 }，{I 1 ，I 3 }，{I 1 ，I 2 ，I 3 }。
In this embodiment, the number of layers of the effective weighted tree model corresponds to the number of items included in the item set, that is, the number of items included in the item set in the mth layer is m.
a. Firstly, the 1 st item set of the first 1 st node and the 1 st item set of the following brother node in the first layer are respectively combined two by two to obtain 2 nd item sets, and the weight of the 2 nd item sets is calculated through a formula (2). And taking the 2 times item sets corresponding to the weight values greater than or equal to the minimum weight support threshold minws as child nodes of the first 1 times node, namely 2 times nodes, and removing the 2 times item sets corresponding to the weight values smaller than the minimum weight support threshold minws.
b. And c, repeating the step a until the weights of the 2 times item sets obtained by the pairwise union of all 1 time nodes are calculated, and constructing 2 times nodes to form a second layer.
c. And c, recursively calling the steps a and b, and constructing an effective weighted tree model until the weight of the obtained m times of item sets is smaller than a minimum weight support threshold minws, namely, removing the m times of item sets, wherein the number of layers of the effective weighted tree model is m-1, namely, indicating that the frequent item sets in the database D are completely mined.
Referring to fig. 2, the item set i= { I in table 1 1 ,I 2 ,I 3 ,I 4 ,I 5 Then 1 st item set is { I }, respectively 1 }，{I 2 }，{I 3 }，{I 4 }，{I 5 Then the 1 st order item set { I } can be calculated by equation (2) 1 }，{I 2 }，{I 3 }，{I 4 }，{I 5 The weights W of the term set { I } are 0.624,0.832,0.624,0.352,0.752, respectively, and the minimum weight supports a threshold minws=0.4 4 Weight less than minimum weight support threshold minwsItem set { I } 4 The term set { I } cannot be used as a 1 st-order node 1 }，{I 2 }，{I 3 }，{I 5 The 1 st node in the first layer of the active weighted tree model. Then the 1 st-order item set { I } 1 }，{I 2 }，{I 3 }，{I 5 The 2 times item sets are obtained by two-by-two union sets, the 2 times item sets with the weight smaller than the minimum weight support threshold minws are removed after the weight of the 2 times item sets is calculated, and then the 2 times node { I ] in the second layer of the effective weighted tree model is obtained 1 ，I 2 }，{I 2 ，I 3 }，{I 2 ，I 5 }. 2 times item sets of 2 times nodes are combined pairwise to obtain 3 times item sets, the weights of the 3 times item sets are smaller than a minimum weight support threshold minws through calculation, the model construction is completed, and the item sets corresponding to all constructed nodes are frequent item sets, namely { I } 2 }，{I 3 }，{I 4 }，{I 5 }，{I 1 ，I 2 }，{I 2 ，I 3 }，{I 2 ，I 5 And is a frequent item set.
For verification, experiments were performed on a ntel Core I5,3.2GHz processor with a memory of 4GB, a hard disk of 500GB, and a Windows 8.1 system installed, all algorithms were implemented in Java, the database used was an Accident dataset, the number of transactions contained was 340183, the number of items was 468, and the size was 34.1M. The experimental results are shown in Table 5.
TABLE 5 Account dataset experiments
As can be seen from table 5, the number of frequent item sets increases with the decrease of the minimum weight support threshold minws, and when the number of frequent item sets is greater than 31, the memory of the WIT algorithm and the WIT-diff algorithm overflows, and the mining of a large number of frequent item sets cannot be performed. The memory is stable in use, does not change greatly along with the change of the number of frequent item sets, and can effectively excavate a large number of frequent item sets in the database.
The present invention provides a method for mining frequently weighted term sets based on an efficient weighted tree, and the method can be implemented in a plurality of ways, and the above is only a preferred embodiment of the present invention. It should be noted that modifications can be made by those skilled in the art without departing from the principles of the present invention, which modifications should also be considered as being within the scope of the present invention. The part not explicitly described in this embodiment can be implemented by the prior art.
Claims (2)
1. A memory optimization method for transaction processing, comprising the steps of:
s1: reading a mark database D from a network, wherein the mark database D comprises N transactions, each transaction comprises different items, the number of the items and the weight w occupied by each item; in the mark database, one commodity represents a corresponding item, and the weight represents the profit of the commodity;
s2: calculating the transaction weight tw of each transaction in the mark database D, and generating a transaction weight table;
in the formula (1), t k Represents the kth transaction; tw (t) k ) Representing transaction t k Transaction weights of (2); i ki Representing an ith item in a kth transaction;representing item I ki Is the number of (3); />Representing item I ki Weights of (I) i.e. item I ki The weight of (2) is w; n represents the number of items;
s3: calculating the weight W of the item set S, and presetting a threshold minws; if W is more than or equal to minws, the item set S is a frequent item set; if W is less than minws, the item set S is an infrequent item set;
in the formula (2), W represents the weight of the term set S, t j Representing the intersection transaction to which the item in the item set S belongs, tw (t j ) Representing transaction t j Is used to determine the transaction weight of (1),representing the sum of the transaction weights of the intersection transactions to which the items in item set S belong, +.>Representing the sum of all transaction weights, tw (t k ) Representing transaction t k Transaction weights of (2);
the mining method of the frequent item set comprises the following steps:
s4: an effective weighted tree model is constructed, frequent item sets in a mark database D are mined, the running memory is reduced, and the memory optimization is completed;
s4-1: calculating the weight of a 1-time item set in the mark database D, taking the 1-time item set corresponding to the weight not smaller than the minimum weight support threshold minws as a 1-time node, and removing the 1-time item set corresponding to the weight smaller than the minimum weight support threshold minws;
s4-2: calculating the weight of m times of item sets in a mark database D, constructing sub-nodes of 1 time of nodes, and forming an effective weighted tree model until the weight of the obtained m times of item sets is smaller than a minimum weight support threshold minws, namely, removing the m times of item sets, wherein the item sets corresponding to all constructed nodes are frequent item sets.
2. The memory optimization method for transaction processing according to claim 1, wherein the method for constructing child nodes of the 1 st-order node is as follows:
a. firstly, obtaining 2 times of item sets by respectively merging the 1 times of item sets in the first layer, and calculating the weight of the 2 times of item sets through a formula (2); taking the 2 times item set corresponding to the weight not smaller than the minimum weight support threshold minws as a child node of the 1 times node, namely the 2 times node;
b. and (c) recursively calling the step (a) until the weight of the obtained m times of item sets is smaller than a minimum weight support threshold minws, and constructing an effective weighted tree model, namely indicating that the frequent item sets in the mark database D are completely mined.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910401314.5A CN110096629B (en) | 2019-05-15 | 2019-05-15 | Memory optimization method for transaction processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910401314.5A CN110096629B (en) | 2019-05-15 | 2019-05-15 | Memory optimization method for transaction processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110096629A CN110096629A (en) | 2019-08-06 |
| CN110096629B true CN110096629B (en) | 2023-07-28 |
Family
ID=67448095
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910401314.5A Active CN110096629B (en) | 2019-05-15 | 2019-05-15 | Memory optimization method for transaction processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110096629B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120136677A (en) * | 2011-06-09 | 2012-12-20 | 성균관대학교산학협력단 | Method and tree structure of database for extracting data steams frequent pattern based on weighted support and structure of database |
| CN105740245A (en) * | 2014-12-08 | 2016-07-06 | 北京邮电大学 | Frequent item set mining method |
| CN106033424A (en) * | 2015-03-11 | 2016-10-19 | 哈尔滨工业大学深圳研究生院 | A data mining method and device |
| CN106250549A (en) * | 2016-08-14 | 2016-12-21 | 重庆大学 | A kind of Frequent Pattern Mining method based on internal memory |
| CN108346085A (en) * | 2018-01-30 | 2018-07-31 | 南京邮电大学 | Electric business platform personalized recommendation method based on weighted frequent items mining algorithm |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7801924B2 (en) * | 2006-12-28 | 2010-09-21 | Infosys Technologies Ltd. | Decision tree construction via frequent predictive itemsets and best attribute splits |
| KR100899144B1 (en) * | 2007-10-26 | 2009-05-27 | 연세대학교 산학협력단 | Method and apparatus for finding maximal frequent itemsets over data streams |
| CN102306183B (en) * | 2011-08-30 | 2014-05-21 | 王洁 | Transaction data stream closed weighted frequent pattern (DS_CWFP) mining method |
| CN107908776A (en) * | 2017-11-30 | 2018-04-13 | 国网浙江省电力公司经济技术研究院 | Frequent mode Web Mining algorithm and system based on affairs project incidence matrix |
-
2019
- 2019-05-15 CN CN201910401314.5A patent/CN110096629B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120136677A (en) * | 2011-06-09 | 2012-12-20 | 성균관대학교산학협력단 | Method and tree structure of database for extracting data steams frequent pattern based on weighted support and structure of database |
| CN105740245A (en) * | 2014-12-08 | 2016-07-06 | 北京邮电大学 | Frequent item set mining method |
| CN106033424A (en) * | 2015-03-11 | 2016-10-19 | 哈尔滨工业大学深圳研究生院 | A data mining method and device |
| CN106250549A (en) * | 2016-08-14 | 2016-12-21 | 重庆大学 | A kind of Frequent Pattern Mining method based on internal memory |
| CN108346085A (en) * | 2018-01-30 | 2018-07-31 | 南京邮电大学 | Electric business platform personalized recommendation method based on weighted frequent items mining algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 委鑫.《优秀硕士论文》.2012,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110096629A (en) | 2019-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Density-based clustering for real-time stream data | |
| Lin et al. | Mining high utility itemsets in big data | |
| CN104881424B (en) | A kind of acquisition of electric power big data, storage and analysis method based on regular expression | |
| US9361343B2 (en) | Method for parallel mining of temporal relations in large event file | |
| KR101003842B1 (en) | Method and system of clustering for multi-dimensional data streams | |
| US20170359361A1 (en) | Selecting representative metrics datasets for efficient detection of anomalous data | |
| CN111612041A (en) | Abnormal user identification method and device, storage medium and electronic equipment | |
| CN111090780B (en) | Method and device for determining suspicious transaction information, storage medium and electronic equipment | |
| CN101119302A (en) | Method for digging frequency mode in the lately time window of affair data flow | |
| CN112925859A (en) | Data storage method and device | |
| You et al. | Accelerated frequent closed sequential pattern mining for uncertain data | |
| Mohbey | Memory-optimized distributed utility mining for big data | |
| Pei et al. | Clustering of temporal event processes | |
| Dinh et al. | A survey of privacy preserving utility mining | |
| Alsaeedi et al. | An Incremental Interesting Maximal Frequent Itemset Mining Based on FP‐Growth Algorithm | |
| CN113506113B (en) | Credit card cash-registering group-partner mining method and system based on associated network | |
| CN110096629B (en) | Memory optimization method for transaction processing | |
| CN109933575A (en) | The storage method and device of monitoring data | |
| Olech et al. | Hierarchical gaussian mixture model with objects attached to terminal and non-terminal dendrogram nodes | |
| Portela et al. | The search of conditional outliers | |
| CN114298245A (en) | Anomaly detection method and device, storage medium and computer equipment | |
| KR100907283B1 (en) | Method and apparatus for finding cluster over data streams | |
| Manikandan et al. | Design of Transactional Prediction using Plan Mine and Genetic Algorithms | |
| Yu et al. | Core decomposition and maintenance in bipartite graphs | |
| CN112750047A (en) | Behavior relation information extraction method and device, storage medium and electronic equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |