C4.5 System

A C4.5 System is a decision tree learning tool that implements the C4.5 algorithm.

• Context:
  • It was developed by Ross Quinlan's research group.
  • It can process both Discrete and Continuous Learning Tasks.
  • It was a common Benchmark System for Supervised Learning Tasks through the 1990s.
  • It implements the C4.5 Algorithm.
  • The source for the eight release is at http://www.rulequest.com/Personal/c4.5r8.tar.gz
  • It is written in Sun C, so if compiling with gcc 4 or hivher, it may require the inclusion of the code #include <stdlib.h> in many of the .cfiles and compilation with the flags: CFLAGS = -fno-builtin-exit -fno-builtin-strcat -fno-builtin-strncat -fno-builtin-strcpy -fno-builtin-strlen -fno-builtin-calloc ...
• Example(s):
  • ../Src/c4.5 -f golf
• Counter-Example(s):
  • C5 System.
  • ID3 System.
  • rpart System.
  • FOIL System.
• See: J. Ross Quinlan.

---

---

References

• http://www.rulequest.com/Personal/c4.5r8.tar.gz

1996

• (Quinlan, 1996) ⇒ J. Ross Quinlan. (1996). “Improved Use of Continuous Attributes in C4.5.” In: Journal of Artificial Intelligence Research, 4.

1993

• (Quinlan, 1993a) ⇒ J. Ross Quinlan. (1993). “C4.5: Programs for machine learning." Morgan Kaufmann. ISBN:1558602380

---