Advances and Trends in Genetic Programming: Volume 1: Classification Techniques and Life Cycles

Advances and Trends in Genetic Programming provides the reader with complete coverage of the most current developments in Genetic Programming for Artificial Intelligence. Volume 1 - Classification Techniques and Life Cycles provides a thorough look at Classification as a systematic way of predicting class membership for a set of examples or instances using the properties of those examples. Classification arises in a wide variety of real life situations, such as detecting faces from large database, finding vehicles, matching fingerprints, and diagnosing medical conditions. A classification algorithm requires huge amount of accuracy and reliability that is very difficult for human programmers. Therefore, there is a need to develop an automated computer-based classification system that can classify the required objects. Presents the latest advances in Genetic Programming for Artificial IntelligenceDiscusses automated computer-based classification algorithms and systems, including comparison of different types of machine learning, and two-class versus multi-class classificationIncludes discussion of tree-based Genetic Programming, the Intron problem, Dynamic Fitness Evaluation, Crossover and Mutation Operators, and presentation of an integrated model-based Genetic Programming Algorithm for multi-class classification INDICE: Section 1: Overview on Machine Learning 1. Introduction on Machine Learning, Genetic programming life cycles, and classification in multi class problems 2. Inter-comparison of different types of machine learning algorithm for classification 3. Two class versus multi-class classification for numeric data 4. Types of genetic programming and their applications Section 2: Tree-Based Genetic Programming 5. Tree-based Genetic programming for Classification 6. Diversity in initial population of Genetic programming 7. Intron in Genetic programming 8. The problem of Bloat in Genetic Programming: Effects of bloat on the Classifier evolvement Section 3: Crossover and Mutation Operators in Genetic Programming 9. Dynamic Fitness Evaluation: It's effects on training paradigm 10. Crossover and Mutation Operators: How they Work in Parallel to Improve the Genetic Programming Life Cycle 11. An Integrated model-based Genetic Programming Algorithm for the Multi-class Classification

• ISBN: 978-0-12-818020-4
• Editorial: Academic Press
• Encuadernacion: Rústica
• Páginas: 220
• Fecha Publicación: 01/12/2020
• Nº Volúmenes: 1
• Idioma: Inglés