Knowledge of extraction from trained neural network by using decision tree
Inside the sets of data, hidden knowledge can be acquired by using neural network. These knowledge are described within topology, using activation function and connection weight at hidden neurons and output neurons. Is hardly to be understanding since neural networks act as a black box. The black bo...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
IEEE
2017
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/18263/ http://umpir.ump.edu.my/id/eprint/18263/1/Knowledge%20of%20Extraction%20from%20Trained%20Neural%20Network%20by%20Using%20Decision%20Tree.pdf http://umpir.ump.edu.my/id/eprint/18263/2/Knowledge%20of%20Extraction%20from%20Trained%20Neural%20Network%20by%20Using%20Decision%20Tree%201.pdf |
| _version_ | 1848820630477078528 |
|---|---|
| author | Soleh, Ardiansyah Mazlina, Abdul Majid Jasni, Mohamad Zain |
| author_facet | Soleh, Ardiansyah Mazlina, Abdul Majid Jasni, Mohamad Zain |
| author_sort | Soleh, Ardiansyah |
| building | UMP Institutional Repository |
| collection | Online Access |
| description | Inside the sets of data, hidden knowledge can be acquired by using neural network. These knowledge are described within topology, using activation function and connection weight at hidden neurons and output neurons. Is hardly to be understanding since neural networks act as a black box. The black box problem can be solved by extracting knowledge (rule) from trained neural network. Thus, the aim of this paper is to extract valuable information from trained neural networks using decision. Further, the Levenberg Marquardt algorithm was applied to training 30 networks for each datasets, using learning parameters and basis weights differences. As the number of hidden neurons increase, mean squared error and mean absolute percentage error decrease, and more time they need to deal with the dataset, that is result of investigation from neural network architectures. Decision tree induction generally performs better in knowledge extraction result with accuracy and precision level from 84.07 to 93.17 percent. The extracted rule can be used to explaining the process of the neural network systems and also can be applied in other systems like expert systems. |
| first_indexed | 2025-11-15T02:12:30Z |
| format | Conference or Workshop Item |
| id | ump-18263 |
| institution | Universiti Malaysia Pahang |
| institution_category | Local University |
| language | English English |
| last_indexed | 2025-11-15T02:12:30Z |
| publishDate | 2017 |
| publisher | IEEE |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | ump-182632018-07-19T07:16:28Z http://umpir.ump.edu.my/id/eprint/18263/ Knowledge of extraction from trained neural network by using decision tree Soleh, Ardiansyah Mazlina, Abdul Majid Jasni, Mohamad Zain QA76 Computer software Inside the sets of data, hidden knowledge can be acquired by using neural network. These knowledge are described within topology, using activation function and connection weight at hidden neurons and output neurons. Is hardly to be understanding since neural networks act as a black box. The black box problem can be solved by extracting knowledge (rule) from trained neural network. Thus, the aim of this paper is to extract valuable information from trained neural networks using decision. Further, the Levenberg Marquardt algorithm was applied to training 30 networks for each datasets, using learning parameters and basis weights differences. As the number of hidden neurons increase, mean squared error and mean absolute percentage error decrease, and more time they need to deal with the dataset, that is result of investigation from neural network architectures. Decision tree induction generally performs better in knowledge extraction result with accuracy and precision level from 84.07 to 93.17 percent. The extracted rule can be used to explaining the process of the neural network systems and also can be applied in other systems like expert systems. IEEE 2017-02 Conference or Workshop Item PeerReviewed pdf en http://umpir.ump.edu.my/id/eprint/18263/1/Knowledge%20of%20Extraction%20from%20Trained%20Neural%20Network%20by%20Using%20Decision%20Tree.pdf pdf en http://umpir.ump.edu.my/id/eprint/18263/2/Knowledge%20of%20Extraction%20from%20Trained%20Neural%20Network%20by%20Using%20Decision%20Tree%201.pdf Soleh, Ardiansyah and Mazlina, Abdul Majid and Jasni, Mohamad Zain (2017) Knowledge of extraction from trained neural network by using decision tree. In: 2nd International Conference on Science in Information Technology (ICSITech 2016) , 26-27 October 2016 , Balikpapan, Indonesia. pp. 220-225.. ISBN 978-1-5090-1721-8 (Published) https://ieeexplore.ieee.org/document/7852637/ |
| spellingShingle | QA76 Computer software Soleh, Ardiansyah Mazlina, Abdul Majid Jasni, Mohamad Zain Knowledge of extraction from trained neural network by using decision tree |
| title | Knowledge of extraction from trained neural network by using decision tree |
| title_full | Knowledge of extraction from trained neural network by using decision tree |
| title_fullStr | Knowledge of extraction from trained neural network by using decision tree |
| title_full_unstemmed | Knowledge of extraction from trained neural network by using decision tree |
| title_short | Knowledge of extraction from trained neural network by using decision tree |
| title_sort | knowledge of extraction from trained neural network by using decision tree |
| topic | QA76 Computer software |
| url | http://umpir.ump.edu.my/id/eprint/18263/ http://umpir.ump.edu.my/id/eprint/18263/ http://umpir.ump.edu.my/id/eprint/18263/1/Knowledge%20of%20Extraction%20from%20Trained%20Neural%20Network%20by%20Using%20Decision%20Tree.pdf http://umpir.ump.edu.my/id/eprint/18263/2/Knowledge%20of%20Extraction%20from%20Trained%20Neural%20Network%20by%20Using%20Decision%20Tree%201.pdf |