CN105868328B - Method and apparatus for log correlation analysis - Google Patents
Method and apparatus for log correlation analysis Download PDFInfo
- Publication number
- CN105868328B CN105868328B CN201610181715.0A CN201610181715A CN105868328B CN 105868328 B CN105868328 B CN 105868328B CN 201610181715 A CN201610181715 A CN 201610181715A CN 105868328 B CN105868328 B CN 105868328B
- Authority
- CN
- China
- Prior art keywords
- item
- path
- log recording
- bit vector
- directed graph
- 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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/242—Query formulation
- G06F16/2425—Iterative querying; Query formulation based on the results of a preceding query
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/50—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
- G06F21/55—Detecting local intrusion or implementing counter-measures
- G06F21/554—Detecting local intrusion or implementing counter-measures involving event detection and direct action
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/50—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
- G06F21/55—Detecting local intrusion or implementing counter-measures
- G06F21/56—Computer malware detection or handling, e.g. anti-virus arrangements
- G06F21/562—Static detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2221/00—Indexing scheme relating to security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F2221/03—Indexing scheme relating to G06F21/50, monitoring users, programs or devices to maintain the integrity of platforms
- G06F2221/034—Test or assess a computer or a system
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Virology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Debugging And Monitoring (AREA)
Abstract
The present invention discloses the method and apparatus for log correlation analysis.In one embodiment, the association rules mining algorithm based on DFS tree changes the generation sequence of candidate, preferably searching maximum frequent itemsets, converts discovery maximum frequent itemsets for the Mining Problems of frequent item set.
Description
Technical field
The present invention relates generally to data mining technology, it particularly relates to the method and apparatus for being used for log correlation analysis.
Background technique
Association analysis is a kind of practical data mining technology, finds the connection between the different item in data acquisition system.Example
Such as, association analysis can be the connection in discovery transaction data base between different commodity.Thus, it is possible to pass through its purchase of discovery customer
The connection between different commodity in object basket, analyzes the buying habit of customer.
Apriori algorithm is a kind of frequent item set (Frequent of Mining Association Rules (Association Rule)
Itemset) algorithm uses the iterative search method of breadth First.The algorithm finds out all frequencies according to support (Support)
Numerous item collection (frequency) and according to confidence level (Confidence) generate correlation rule (intensity).Support (A- > B) indicates institute
There is existing A in event to have the probability of B, P (AB) again;Confidence level (A- > B) indicates in the event for A occur while the general of B occurs
Rate, and P (B | A)=P (AB)/P (A).The target of Apriori algorithm is to find to meet minimum support threshold value and min confidence threshold
The rule of value, i.e., strong rule.The process of algorithm includes finding out frequent 1 item collection set L1 to look for frequent 2 set L2 with L1, then
L3 is looked for L2, is circuited sequentially, until cannot find frequent k item collection.Here, frequent k item collection refers to that support is greater than minimum
The item collection including k item of support threshold.In above process, by Connection Step from frequent k-1 item collection Lk-1 and itself
Connection generates candidate k item collection Ck, and the candidate it is not possible that in Lk is got rid of by beta pruning step.
But Apriori algorithm needs Multiple-Scan database in the iterative process of execution " connection-beta pruning ", increases
I/O load.
Summary of the invention
A method of for log correlation analysis, comprising: bit vector generation unit, for being directed to each as event
The log recording of generation generates the position for indicating that whether this occurs in corresponding log recording for each of log recording
Vector, wherein the i-th bit instruction 0 of the bit vector of an item indicates that the Xiang Wei occurs in i-th of log recording, and instruction 1 indicates
This occurs in i-th of log recording;Non-directed graph generation unit, it is undirected for being constructed using the item in log recording as node
Scheme, all items occurred in one of log recording have side, path searching unit, for finding the non-directed graph between any two
In longest path, which includes k, judging unit, the bit vector of the item on the longest path is carried out and operation,
And judge whether the k item collection on the longest path constitutes Maximum Frequent according to the quantity of the numerical value 1 in the result with operation
Item collection.
A kind of device for log correlation analysis, comprising: bit vector generation unit, for being directed to each as event
The log recording of generation generates the position for indicating that whether this occurs in corresponding log recording for each of log recording
Vector, wherein the i-th bit instruction 0 of the bit vector of an item indicates that the Xiang Wei occurs in i-th of log recording, and instruction 1 indicates
This occurs in i-th of log recording;Non-directed graph generation unit, it is undirected for being constructed using the item in log recording as node
Scheme, all items occurred in one of log recording have side, path searching unit, for finding the non-directed graph between any two
In longest path, which includes k, judging unit, the bit vector of the item on the longest path is carried out and operation,
And judge whether the k item collection on the longest path constitutes Maximum Frequent according to the quantity of the numerical value 1 in the result with operation
Item collection.
According to one embodiment of present invention, have to the traversal in audit log data library and only once, can reduce big
It measures unnecessary repetition to compare, improves the efficiency of log correlation analysis.
According to one embodiment of present invention, it when audit log data library and minimum support update, does not need again
Scan database can complete the update in audit log correlation rule library, to improve the digging efficiency of correlation rule.
According to one embodiment of present invention, the association rules mining algorithm based on DFS tree can change
Candidate generation sequence, greatly reduce the quantity of the candidate of generation, and simplify transaction database update and
The replacement problem of frequent item set when minimum support changes.
According to one embodiment of present invention, rule digging can be associated to a large amount of audit log data, it can be with
It was found that valuable information in cloud resource management platform, effective to realize system intrusion detection, the safety in timely discovery system
Loophole and internal malicious act.
The other feature and advantage that also will be understood that the embodiment of the present invention when being read in conjunction with the figure and being described below, wherein attached
Figure shows the principle of the embodiment of the present invention by means of example.
Detailed description of the invention
Fig. 1 is the schematic diagram of the method according to an embodiment of the invention for log correlation analysis.
Fig. 2 is the non-directed graph generated according to the example in table 1.
Fig. 3 is showing for the method according to an embodiment of the invention that frequent item set is searched based on DFS tree
It is intended to.
Fig. 4 is the schematic diagram of the device according to an embodiment of the invention for log correlation analysis.
Specific embodiment
Hereinafter, the principle that invention will be described in conjunction with the embodiments.It should be understood that the embodiment provided is only
Those skilled in the art more fully understand and practice the present invention, are not intended to limit the scope of the invention.Therefore, this specification
In include implementation detail be not necessarily to be construed as limitation to the range or the range that may be claimed of invention, but should
It is considered as the description specific to embodiment.
Fig. 1 is the schematic diagram of the method according to an embodiment of the invention for log correlation analysis.The embodiment
Attempt to find maximum frequent itemsets.Obtained maximum frequent itemsets can be used to close the item in log recording with particular event
Connection, dynamically detection system is invaded.As shown in Figure 1, that steps are as follows is described for method each.
In step 110, the bit vector about the item in log recording is generated.The day occurred for each as event
Will record generates the bit vector for indicating that whether this occurs in corresponding log recording for each of log recording,
In, the i-th bit instruction 0 of the bit vector of an item indicates that the Xiang Wei occurs in i-th of log recording, and instruction 1 indicates that this exists
Occur in i-th of log recording.
As an example, the following table 1 shows database journal record sheet according to an embodiment of the invention.In the table,
One log recording indicates an event.
Log recording | ? |
1 | I2、I3 |
2 | I1、I2、I5 |
3 | I1、I2、I4 |
4 | I1、I3 |
5 | I1、I2、I3 |
6 | I2、I4 |
7 | I2、I3 |
8 | I1、I2、I3、I5 |
9 | I1、I3 |
Table 1
In the step 110, for each log recording 1-9, respective bit vector is generated for item I1-I5.According to
It indicates, can produce bit vector as shown in table 2 below:
? | Bit vector |
I1 | 011110011 |
I2 | 111011110 |
I3 | 100110111 |
I4 | 001001000 |
I5 | 010000010 |
Table 2
For example, the bit vector 011110011 of I1 indicates that it occurs in journal record 2-5,8-9.
In the step 120, non-directed graph is constructed by node of the item in log recording.It defines and occurs in a log recording
There is side between any two in all items.Accordingly, for the example in table 1, item I2 and I3 in log recording 1 have side, day
Item I1, I2, I5 in will record 2 have side, etc. between any two.Fig. 2 is the non-directed graph generated according to the example in table 1.
In step 130, the longest path of the non-directed graph is found, which includes k.In one example, root
Longest path is found according to depth-first traversal algorithm.
In step 140, according to the bit vector of the item on longest path, judge whether its item collection constitutes maximum frequent set
Collection.The bit vector of item on the longest path is carried out and operation, and according to the quantity of the numerical value 1 in the result with operation come
Judge whether the k item collection on the longest path constitutes maximum frequent itemsets.
In the illustrated example shown in fig. 2, finding out longest path is I2-I4-I1-I3-I6, to the position of the item in the longest path
Vector carries out and operation, result 000000000.This means that all items on this paths are not remembered in a log simultaneously
Occur in record.That is, the k item collection on the longest path can not constitute maximum frequent itemsets.
In one example, when the support for judging the item collection on the longest path is less than threshold value, according to the non-directed graph
Find vice-minister path with the longest path, which includes k-1, to the bit vector of the item on the vice-minister path carry out with
Operation, and judge whether the k-1 item collection on the vice-minister path is constituted most according to the quantity of the numerical value 1 in the result with operation
Big frequent item set.The candidate that vice-minister path can be excessive to avoid generation is searched by non-directed graph and the longest path, into
And improve the efficiency for searching maximum frequent itemsets.And so on, until k is equal to 1.
Fig. 3 is showing for the method according to an embodiment of the invention that frequent item set is searched based on DFS tree
It is intended to.In this embodiment, it is based on DFS tree (Frequent Items mining based on Depth-
First tree, is abbreviated as FIDF-tree) association rules mining algorithm change change candidate generation sequence, preferably
Maximum frequent itemsets are found, convert discovery maximum frequent itemsets for the Mining Problems of frequent item set.
As shown in figure 3, that steps are as follows is described for method each.
In the step 310, initial parameter, such as support threshold, confidence threshold value etc. are set.
In step 320, every bit vector is generated.It can generate as described in abovementioned steps 110 about in log recording
The bit vector of item.Each characteristic value in item log recording in log recording, such as source IP address, purpose IP address, service class
Type, connection status etc..
In a step 330, according to the relevance between items, double loop networks are constructed.
In step 340, k tree is generated.
According to one embodiment of present invention, the structure of FIDF-tree is defined as follows: tree root is defined as " NULL ", candidate
Item collection { I1, I2, I3 ..., IS } is expressed as from the node on the child V1 to the path of leaf node VS of root node in tree.It is non-
Leaf node includes an attribute: the title of item.Leaf node includes two attributes: the title of item is corresponding with the leaf node
The support of candidate.The item collection path of candidate { I1, I2, I3 ..., IS } refers to be traversed from the child I1 of root node
Leaf node Is in tree.Path from I1 to Is is connected by side (I1, I2), (I2, I3) ..., (IS-1, IS)
The node sequence I1, I2, I3 ... come, IS, and duplicate node is not present in sequence.If in FIDF-tree represented by path
The maximum length of candidate is k, then the FIDF-tree is known as k-FIDF-tree.
In the step 330, a root node root is created first, NULL is identified as, is then calculated according to depth-first
Method traverses double loop networks, finds out longest path, constructs the first stalk tree of root.If nodes all at this time have all accessed
It crosses, then k-FIDF-tree construction complete, the node otherwise not accessed from other again continues to search vice-minister path, as root node
Another subtree, until all nodes are all accessed.Finally all occurred in item collection path bit vectors are carried out
AND operation obtains the support of the item collection, and second attribute of the leaf node labeled as the path.
In step 350, judge whether the support of leaf node is greater than minimum support threshold value.If it is, in step
Maximum frequent itemsets are exported in rapid 360.The maximum frequent itemsets are greater than the leaf node of minimum support threshold value by support
All items in the path at place are constituted.In the DFS tree of construction, the nonvoid subset of any frequent item set is certain
It is frequent item set, and the superset of any nonmatching grids must be nonmatching grids.In this embodiment, in run-down data
On the basis of library, maximum frequent itemsets are preferentially gone for, therefore, when database and support threshold change, are not needed again
Scan database, improves the efficiency of association rule mining, and can satisfy the demand of real-time.
In step 350, judge whether the support of leaf node is greater than minimum support threshold value.If it is not, then in step
In rapid 370, (k-1)-FIDF-tree is generated.In step 370, for the item collection path of k-FIDF-tree, subset k-1 is obtained
Candidate,.Wherein, for each candidate, (k-1) subtree is constructed, is occurred according to i-th of element in k-1 candidate
Position establish i-th layer of tree node of subtree, if there is new node on layer position a certain during traversal new tree
Tree is linked in sequence into the middle surplus element of this candidate to be added in subtree.
Step 350 is entered back into, judges whether the support of the leaf node in (k-1)-FIDF-tree is greater than minimum support
Spend threshold value.If it is, exporting maximum frequent itemsets in step 360.If it is not, then it is again introduced into step 370, (k-1)-
Longest item collection path in FIDF-tree, seeking length is the Son item set of (k-2).Step 370 and 350 is repeated, until there is leaf knot
The support of point meets support threshold.
Fig. 4 is the schematic diagram of the device according to an embodiment of the invention for log correlation analysis.In the implementation
In example, device includes bit vector generation unit 410, non-directed graph generation unit 420, path searching unit 430, judging unit 440.
Bit vector generation unit 410 is configured to the log recording occurred for each as event, is in log recording
Each of generate and to indicate the bit vector that whether occurs in corresponding log recording of this, wherein the bit vector of an item
I-th bit instruction 0 indicates that the Xiang Wei occurs in i-th of log recording, and instruction 1 indicates that this occurs in i-th of log recording.
Non-directed graph generation unit 420 is configured to construct non-directed graph, one of log by node of the item in log recording
There is side between any two in all items occurred in record.
Path searching unit 430 is configured to find the longest path in the non-directed graph, which includes k.
Judging unit 440 is configured to carry out the bit vector of the item on the longest path and operation, and according to operation
Result in the quantity of numerical value 1 whether constitute maximum frequent itemsets come the k item collection judged on the longest path.
In another embodiment, when judging unit 440 judges that the support of the item collection on the longest path is less than threshold value
When, path searching unit 430 is further configured to find vice-minister path according to the non-directed graph and the longest path, the vice-minister path
Including k-1, the judging unit is further configured to that the bit vector of the item on the vice-minister path is carried out and operated, and
Judge whether the k-1 item collection on the vice-minister path constitutes maximum frequent set according to the quantity of the numerical value 1 in the result with operation
Collection.
Fig. 1 and each frame shown in Fig. 3 can be considered as method and step, and/or be considered as due to running computer program generation
It is operated caused by code, and/or is considered as being configured to implement the logic circuit component of multiple couplings of correlation function.Although operation
It is depicted in figure in a specific sequence, but this is understood not to require the particular order shown in or come in sequential order
These operations are executed, or the operation of all illustrations is required to be performed, to do the trick.In some cases, multitask
Parallel processing may be advantageous.
Exemplary embodiment can be implemented in hardware, software, or a combination thereof.For example, certain aspects of the invention can be hard
Implement in part, and other aspects can then be implemented in software.Although the aspect of exemplary embodiment of the present invention can be shown and
It is described as block diagram, flow chart, but is well understood that, these devices described herein or method can be as non-limiting reality
Functional module is implemented as in the system of example.In addition, above-mentioned apparatus is understood not to require to carry out in all of the embodiments illustrated
This separation, and should be understood that described program assembly and system and generally can be integrated in single software product
Or it is packaged into multiple software product.
Those skilled in the relevant art implement aforementioned exemplary of the invention when aforementioned specification is read in conjunction with the figure
The various modifications of example and deformation can become obvious for those skilled in the relevant art.Therefore, the embodiment of the present invention is not limited to
Disclosed specific embodiment, and variation and other embodiments are intended within the scope of the appended claims.
Claims (4)
1. a kind of device for log correlation analysis characterized by comprising
Bit vector generation unit, the log recording for occurring for each as event are each item in log recording
Generate the bit vector for indicating that whether this occurs in corresponding log recording, wherein the i-th bit of the bit vector of an item indicates
0 indicates that the Xiang Wei occurs in i-th of log recording, and instruction 1 indicates that this occurs in i-th of log recording;
Non-directed graph generation unit goes out in one of log recording for constructing non-directed graph by node of the item in log recording
There is side between any two in existing all items,
Path searching unit, for finding the longest path in the non-directed graph, which includes k,
Judging unit is carried out and is operated to the bit vector of the item on the longest path, and according to the number in the result with operation
Whether the quantity of value 1 constitutes maximum frequent itemsets come the k item collection judged on the longest path.
2. device as described in claim 1, which is characterized in that
When the judging unit judge the item collection on the longest path support be less than threshold value when, the path searching unit into
One step is configured to find vice-minister path according to the non-directed graph and the longest path, which includes k-1, the judgement
Unit is further configured to that the bit vector of the item on the vice-minister path is carried out and operated, and according in the result with operation
The quantity of numerical value 1 judge whether the k-1 item collection on the vice-minister path constitutes maximum frequent itemsets.
3. a kind of method for log correlation analysis characterized by comprising
It is whether each of log recording generation indicates this in phase for the log recording that each occurs as event
The bit vector occurred in the log recording answered, wherein the i-th bit instruction 0 of the bit vector of an item indicates the Xiang Wei on i-th
Occur in will record, instruction 1 indicates that this occurs in i-th of log recording;
Non-directed graph is constructed by node of the item in log recording, all items occurred in one of log recording are deposited between any two
On side,
The longest path in the non-directed graph is found, which includes k,
The bit vector of item on the longest path is carried out and operated, and according to the quantity of the numerical value 1 in the result with operation
Whether the k item collection to judge on the longest path constitutes maximum frequent itemsets.
4. method as claimed in claim 3, which is characterized in that
When the support for judging the item collection on the longest path is less than threshold value, found according to the non-directed graph and the longest path secondary
Long path, the vice-minister path include k-1, are carried out to the bit vector of the item on the vice-minister path and operation, and according to behaviour
The quantity of numerical value 1 in the result of work judges whether the k-1 item collection on the vice-minister path constitutes maximum frequent itemsets.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610181715.0A CN105868328B (en) | 2016-03-28 | 2016-03-28 | Method and apparatus for log correlation analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610181715.0A CN105868328B (en) | 2016-03-28 | 2016-03-28 | Method and apparatus for log correlation analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105868328A CN105868328A (en) | 2016-08-17 |
CN105868328B true CN105868328B (en) | 2019-05-10 |
Family
ID=56626048
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610181715.0A Active CN105868328B (en) | 2016-03-28 | 2016-03-28 | Method and apparatus for log correlation analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105868328B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107835087B (en) * | 2017-09-14 | 2022-09-02 | 北京科东电力控制系统有限责任公司 | Automatic extraction method of alarm rule of safety equipment based on frequent pattern mining |
CN112199344B (en) * | 2020-10-14 | 2024-03-19 | 杭州安恒信息技术股份有限公司 | Log classification method and device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102043851A (en) * | 2010-12-22 | 2011-05-04 | 四川大学 | Multiple-document automatic abstracting method based on frequent itemset |
CN103593400A (en) * | 2013-12-13 | 2014-02-19 | 陕西省气象局 | Lightning activity data statistics method based on modified Apriori algorithm |
CN103678530A (en) * | 2013-11-30 | 2014-03-26 | 武汉传神信息技术有限公司 | Rapid detection method of frequent item sets |
-
2016
- 2016-03-28 CN CN201610181715.0A patent/CN105868328B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102043851A (en) * | 2010-12-22 | 2011-05-04 | 四川大学 | Multiple-document automatic abstracting method based on frequent itemset |
CN103678530A (en) * | 2013-11-30 | 2014-03-26 | 武汉传神信息技术有限公司 | Rapid detection method of frequent item sets |
CN103593400A (en) * | 2013-12-13 | 2014-02-19 | 陕西省气象局 | Lightning activity data statistics method based on modified Apriori algorithm |
Non-Patent Citations (1)
Title |
---|
一种基于图的关联规则挖掘改进算法;王政伟等;《计算机工程与科学》;20050512;第27卷(第5期);第48-51页 |
Also Published As
Publication number | Publication date |
---|---|
CN105868328A (en) | 2016-08-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Kumbhare et al. | An overview of association rule mining algorithms | |
CN108463973A (en) | Fingerprint recognition basic reason is analyzed in cellular system | |
US8954311B2 (en) | Arrangements for extending configuration management in large IT environments to track changes proactively | |
Duan et al. | Root cause analysis approach based on reverse cascading decomposition in QFD and fuzzy weight ARM for quality accidents | |
US11892904B2 (en) | Directed incremental clustering of causally related events using multi-layered small world networks | |
US11055631B2 (en) | Automated meta parameter search for invariant based anomaly detectors in log analytics | |
Song et al. | Uniwalk: Unidirectional random walk based scalable simrank computation over large graph | |
CN105868328B (en) | Method and apparatus for log correlation analysis | |
Robu et al. | Mining frequent patterns in data using apriori and eclat: A comparison of the algorithm performance and association rule generation | |
Bala et al. | Performance analysis of apriori and fp-growth algorithms (association rule mining) | |
US20230273924A1 (en) | Trimming blackhole clusters | |
Sinha et al. | Identification of best algorithm in association rule mining based on performance | |
US20230205618A1 (en) | Performing root cause analysis on data center incidents | |
Pandey et al. | Mining on relationships in big data era using improve apriori algorithm with MapReduce approach | |
Lakshmi et al. | Sensitivity analysis for safe grainstorage using big data | |
US20190294534A1 (en) | Program usability performance classification | |
Duggirala et al. | Mining Positive and Negative Association Rules Using CoherentApproach | |
LABZIOUI et al. | New Approach based on Association Rules for Building and Optimizing OLAP Cubes on Graphs | |
US20240086762A1 (en) | Drift-tolerant machine learning models | |
Sevim et al. | A link prediction framework for hotel recommendations | |
Reinthal et al. | Finding the densest common subgraph with linear programming | |
Girgis et al. | An approach for enhancing regression testing using genetic algorithm and data flow analysis | |
Fu et al. | K-clique community detection based on union-find | |
Muthukumaran et al. | Software defect prediction using augmented Bayesian networks | |
Fegade et al. | Mining Frequent Itemsets for Improving the Effectiveness of Marketing and Sales |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |