Data mining reduction methods and performances of rules
In data mining the accuracy of models are associated with the strength of the rules.However, most machine learning techniques produce a large number of rules.The consequence is with large number of rules generated,processing time is much longer. This study examines rules of different lengths of attr...
| Main Authors: | , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2009
|
| Subjects: | |
| Online Access: | https://repo.uum.edu.my/id/eprint/13596/1/PID263.pdf |
| _version_ | 1825803215160999936 |
|---|---|
| author | Ahmad, Faudziah Basir, Mohammad Aizat |
| author_facet | Ahmad, Faudziah Basir, Mohammad Aizat |
| author_sort | Ahmad, Faudziah |
| collection | UUM |
| description | In data mining the accuracy of models are associated with the strength of the rules.However, most machine learning techniques produce a large number of rules.The consequence is with large number of rules generated,processing time is much longer. This study examines rules of different lengths of attributes in terms of performance based on percentage of accuracy. The research adopts the Knowledge Discovery in Databases “KDD” methodology for analysis and applies various data mining techniques in the experiments.Data of 50 hardware dataset companies which, contains 31 attributes and 400 records have been used. In summary, results show that in terms of performance of rules, Genetic Algorithm has produced the highest number of rules followed by Johnson’s Algorithm and Holte’s 1R.The best classifier for extracting rules in this study is VOT (Voting of
Object Tracking).In terms of performance of rules, best results comes from rules with 30 attributes, followed by rules with 1 intersection attribute and lastly rules with 3 intersection attributes. Among the three sets of attributes, the 3 intersection attributes are considered as the attributes that can be used as predictor attributes. |
| first_indexed | 2024-07-04T05:53:12Z |
| format | Conference or Workshop Item |
| id | uum-13596 |
| institution | Universiti Utara Malaysia |
| language | English |
| last_indexed | 2024-07-04T05:53:12Z |
| publishDate | 2009 |
| record_format | eprints |
| spelling | uum-135962015-04-07T03:17:59Z https://repo.uum.edu.my/id/eprint/13596/ Data mining reduction methods and performances of rules Ahmad, Faudziah Basir, Mohammad Aizat QA76 Computer software In data mining the accuracy of models are associated with the strength of the rules.However, most machine learning techniques produce a large number of rules.The consequence is with large number of rules generated,processing time is much longer. This study examines rules of different lengths of attributes in terms of performance based on percentage of accuracy. The research adopts the Knowledge Discovery in Databases “KDD” methodology for analysis and applies various data mining techniques in the experiments.Data of 50 hardware dataset companies which, contains 31 attributes and 400 records have been used. In summary, results show that in terms of performance of rules, Genetic Algorithm has produced the highest number of rules followed by Johnson’s Algorithm and Holte’s 1R.The best classifier for extracting rules in this study is VOT (Voting of Object Tracking).In terms of performance of rules, best results comes from rules with 30 attributes, followed by rules with 1 intersection attribute and lastly rules with 3 intersection attributes. Among the three sets of attributes, the 3 intersection attributes are considered as the attributes that can be used as predictor attributes. 2009-06-24 Conference or Workshop Item PeerReviewed application/pdf en https://repo.uum.edu.my/id/eprint/13596/1/PID263.pdf Ahmad, Faudziah and Basir, Mohammad Aizat (2009) Data mining reduction methods and performances of rules. In: International Conference on Computing and Informatics 2009 (ICOCI09), 24-25 June 2009, Legend Hotel, Kuala Lumpur. http://www.icoci.cms.net.my |
| spellingShingle | QA76 Computer software Ahmad, Faudziah Basir, Mohammad Aizat Data mining reduction methods and performances of rules |
| title | Data mining reduction methods and performances of rules |
| title_full | Data mining reduction methods and performances of rules |
| title_fullStr | Data mining reduction methods and performances of rules |
| title_full_unstemmed | Data mining reduction methods and performances of rules |
| title_short | Data mining reduction methods and performances of rules |
| title_sort | data mining reduction methods and performances of rules |
| topic | QA76 Computer software |
| url | https://repo.uum.edu.my/id/eprint/13596/1/PID263.pdf |
| work_keys_str_mv | AT ahmadfaudziah dataminingreductionmethodsandperformancesofrules AT basirmohammadaizat dataminingreductionmethodsandperformancesofrules |