- Data Structure
- Networking
- RDBMS
- Operating System
- Java
- MS Excel
- iOS
- HTML
- CSS
- Android
- Python
- C Programming
- C++
- C#
- MongoDB
- MySQL
- Javascript
- PHP
- Physics
- Chemistry
- Biology
- Mathematics
- English
- Economics
- Psychology
- Social Studies
- Fashion Studies
- Legal Studies
- Selected Reading
- UPSC IAS Exams Notes
- Developer's Best Practices
- Questions and Answers
- Effective Resume Writing
- HR Interview Questions
- Computer Glossary
- Who is Who
Data Mining multidimensional association rule
Association rule mining helps us to find relationships among large dataset.
In Multidimensional association,
Multidimensional association rule comprises of more than one aspect
Numeric attributes should be discretized.
Attributes can be unmitigated or quantitative.
Quantitative characteristics are numeric and consolidate pecking order.
Three approaches in mining multidimensional association rules are −
Using static discretization of quantitative attributes
Discretization happens earlier to mining and is static. Discretized attributes are treated as absolute and use an algorithm called apriori algorithm to search for all k-frequent predicate sets(k or k+1 table scans are required). Each subset of a frequent predicate set should be continuous. Data cubes are best for mining as they are helping make mining accelerate. The predicate cells are related to the cells of an n-dimensional data cuboid.
Example − In a data cube, the 3D cuboid (id, name, class) are frequent, then it suggests that(id, name),(name, class),(id, class) are also frequent.
Using dynamic discretization of quantitative attributes
It is also known as mining Quantitative Association Rules and the attributes that are numeric are dynamically discretized.
Example
age(A,"12..25")Λrank(A,"1..4")Λgets(A,"laptop computer")
Output
rank1 |
rank2 |
rank3 |
rank4 |
|
---|---|---|---|---|
age,22-25 |
||||
age,18-21 |
||||
age,12-17 |
Here, the Attributes are classified in bins and are based on distribution of data. The bins can be further combined for analysis, therefore having the discretization dynamic.
Grid For Tuples
Distance based discretization with clustering
This is a dynamic discretization process that considers the distance between data of interest.
There are two steps involved in the mining process.
Interval of attributes involved are found by performing clustering.
Association rules are acquired by searching for groups of clusters that occur together.
NOTE − Clusters in rule antecedent and consequent are strongly associated and occur together.
Conclusion
The article consists of data mining using multidimensional association rule which help find relationship among the dataset and then use different approach(discretization) to perform mining. First approach is by using static discretization which uses apriori algorithm to locate predicate sets. Second is dynamic discretization which helps numeric attributes to dynamically discretized. Third is distance based discretization which measures distance between data points.