A new domain specific scripting language for automated machine learning pipeline
This paper concerns on two difficulties faced by non-experts� users in the utilization of machine learning; design of the model and the programming task for implementation. From the varieties of the machine learning algorithms, selecting the best model with the best configurations is a critical an...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Blue Eyes Intelligence Engineering and Sciences Publication
2019
|
| Online Access: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074531804&doi=10.35940%2fijrte.B1082.0982S1119&partnerID=40&md5=5b698dacf7267737db7a4da5a91fb53a http://eprints.utp.edu.my/24965/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This paper concerns on two difficulties faced by non-experts� users in the utilization of machine learning; design of the model and the programming task for implementation. From the varieties of the machine learning algorithms, selecting the best model with the best configurations is a critical and complex design issue. In light of this situation, automated machine learning pipeline is highly beneficial. Research has proved that Genetic Programming is highly useful to find the best pipeline of an automated machine learning model. However, in respond to the implementation difficulty, there exists a limited software tool that support easy implementation for automated machine learning based on Genetic Programming. This paper presents the specifications of a domain specific scripting language for the easy development of an automated machine learning with the underlying Genetic Programming. The scripting language has a very minimal characters of codes, hence easier to understand and more concise than the Python programming language. © BEIESP. |
|---|