From contingency tables to various forms of knowledge in databases
Advances in knowledge discovery and data mining
Selection of relevant features and examples in machine learning
Artificial Intelligence - Special issue on relevance
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Data Mining: Concepts, Models, Methods and Algorithms
Data Mining: Concepts, Models, Methods and Algorithms
| Hi-index | 0.00 |
The University of Alabama has developed a software system called the Critical Analysis Reporting Environment (CARE). CARE was designed to provide information for the analysis of automobile crash data. One of the most important applications of CARE is in enabling the decision maker to determine what causes crashes. In this paper, a modified genetic algorithm is used to identify the potential problem areas which are the combination of causal attributes. To find the combination of attributes that causes the accident is a NP-hard combinatorial problem when a big number of variables is involved. However, with an appropriate fitness function, the given combinatorial problem can be solved by a genetic algorithm where only a fraction of possible combinations are used to guide the search to produce the best results. Genetic algorithms that are loosely based on biological evolution provide an effective search method when the problem solving procedure can not be well defined. The crash records of the year 2000 in Alabama's Walker County will be used to demonstrate and to evaluate the proposed algorithm.