19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Decision Trees and Random Forests Modeling by using P Systems José M. Sempere Department of Information Systems and Computation Universitat Politècnica de València jsempere@dsic.upv.es http://personales.upv.es/jsempere
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Goals of this work: (1) Provide algorithms to define P systems from decision trees and random forests (2) Introduce machine learning techniques to obtain decision trees and random forests from given data through membrane and object rules based on P systems working in entropic/functional manner Previous Works Decision Tree Models Induced by Membrane Systems (2015) J. Wang, J. Hu, M.J. Pérez-Jiménez, A.Riscos-Núñez • ROMJIST Vol.18 No.3 pp 228-239 Self-constructing Recognizer P Systems. D. Díaz-Pernil, F. Peña-Cantillana, M.A. Gutiérrez-Naranjo. In Proceedings of • the Thirteenth Brainstorming Week on Membrane Computing , pp 137-154. Fenix Editora. 2014. Díaz-Pernil et al . Wang et al. Our approach Trees defined by the membrane structure Tree-like objects Trees defined by the membrane structure Non-deterministic search of structures External Induction Algorithm Algorithm runs by P rules within an entropic manner
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany A Decision Tree is a representation for a discrete-valued function !: # $ ×# & × ⋯×# ( → * # $ +,-./(# $ . 2, 2 $ ) +,-./(# $ . 2, 2 5 ) # & +,-./ # 6 . 2, 2 8 ∈ {# 6 . 2 = 2 8 , # 6 . 2 ≤ 2 8 , # 6 . 2 ≥ 2 8 , # 6 . 2 ≠ 2 8 , …} # 6 +,-./(# 6 . 2, 2 7 ) +,-./(# 6 . 2, 2 $ ) *. 2 6 *. 2 $
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany An example: protein-protein interactions prediction (From “What are decision trees ?”. Carl Kingsford & Steven L. Salzberg. Nature Biotechnology 26 No. 9, 1011 – 1013 (2008) )
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Implementing decision trees by membrane structures (I) Objects alphabet Membrane labels alphabet
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Implementing decision trees by membrane structures (I) An algorithm to translate decision trees to P systems
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Implementing decision trees by membrane structures (I) An algorithm to translate decision trees to P systems
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Implementing decision trees by membrane structures (I) An algorithm to translate decision trees to P systems
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Implementing decision trees by membrane structures (I) An algorithm to translate decision trees to P systems
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Implementing decision trees by membrane structures (II) SF Yes SF No EC Yes GDL5 Yes GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Implementing decision trees by membrane structures (III) SF Yes SF No EC Yes ∀ ", $, % ∈ '(), *+ ,- '() .-/ " .0 $ 12/3 % → [.0 $ ] ,- 7() ∀ ", $, % ∈ '(), *+ ,- *+ .-/ " .0 $ 12/3 % → [.-/ " 12/3 % ] ,- 8+ GDL5 Yes GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Implementing decision trees by membrane structures (IV) SF Yes SF No EC Yes ∀ ", $, % ∈ '(), *+ ,- '() .-/ " .0 $ 12/3 % → [.0 $ ] ,- 7() ∀ ", $, % ∈ '(), *+ ,- *+ .-/ " .0 $ 12/3 % → [.-/ " 12/3 % ] ,- 8+ GDL5 Yes 12/3 '() → [7,.] 12/3 7() GDL5 No 12/3 *+ → [89] 12/3 8+ SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Implementing decision trees by membrane structures (V) [YES] → [ ] YES [NO] → [ ] NO [YES] → [ ] YES SF No SF Yes [NO] → [ ] NO [NO] → [ ] NO [YES] → [ ] YES EC Yes ∀ ", $, % ∈ '(), *+ ,- '() .-/ " .0 $ 12/3 % → [.0 $ ] ,- 7() ∀ ", $, % ∈ '(), *+ ,- *+ .-/ " .0 $ 12/3 % → [.-/ " 12/3 % ] ,- 8+ [YES] → [ ] YES [NO] → [ ] NO GDL5 Yes 12/3 '() → [7,.] 12/ 7() [NO] → [ ] NO GDL5 No 12/3 *+ → [89] 12/ 8+ SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany A parsing example SF Yes SF No EC Yes ECno SCLyes SFyes GDL5no GDL5 Yes GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany A parsing example SF Yes SF No EC Yes SCLyes GDL5 Yes GDL5no GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany A parsing example SF Yes SF No EC Yes GDL5 Yes GDL5no GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany A parsing example SF Yes SF No EC Yes GDL5 Yes NO GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany A parsing example SF Yes SF No EC Yes GDL5 Yes NO GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany A parsing example SF Yes SF No EC Yes GDL5 Yes NO GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany A parsing example SF Yes SF No EC Yes NO GDL5 Yes GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany A parsing example SF Yes SF No EC Yes NO GDL5 Yes GDL5 No SCL No SCL Yes EC No
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Applying a machine learning technique inside a P system Sample ! " # Decision 1 High High High Yes 2 High High High Yes 3 Low High Low Yes 4 Medium High High Not
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Applying a machine learning technique inside a P system Sample ! " # Decision 1 High High High Yes 2 High High High Yes 3 Low High Low Yes 4 Medium High High Not ( ( ( ( ! $%&' " $%&' # $%&' )*+%,%-. "/0 1 1 1 1 ! $%&' " $%&' # $%&' )*+%,%-. "/0 4 4 4 4 ! 2-3 " $%&' # 2-3 )*+%,%-. "/0 9 9 9 9 ! 5*6%78 " $%&' # 2-3 )*+%,%-. :;
19 th International Conference on Membrane Computing (CMC19) 4-7 September 2018 , Dresden, Germany Applying a machine learning technique inside a P system Sample ! " # Decision A generic algorithm to build decision trees from data 1 High High High Yes Input: A finite set of supervised tuples E 2 High High High Yes Output: A decision tree T Method: 3 Low High Low Yes 1) Create an arbitrary root 4 Medium High High Not 2) If all the tuples belong to class < = then return (root, < = ) else 1. Select an attribute ! with values > ( , > 1 , … , > 5 ( ( ( ( 5 / % ! $%&' " $%&' # $%&' )*+%,%-. "/0 2. Make a partition of E according to the attribute value / ( , / 1 , … , / 5 ∶ / = ⋃ %D( 3. Build decision trees for every subset: E ( , E 1 , … , E 5 1 1 1 1 ! $%&' " $%&' # $%&' )*+%,%-. "/0 endMethod 4 4 4 4 ! 2-3 " $%&' # 2-3 )*+%,%-. "/0 9 9 9 9 FGGH ! 5*6%78 " $%&' # 2-3 )*+%,%-. :; I = J K I = J M I = J L N L N M N K
Recommend
More recommend