The Application of Neuro-Fuzzy Decision Tree in Optimal Selection of Technological Innovation Projects
Hongxia Jin, Jianna Zhao, Xiaoxuan Chen · 2007
When the mathematical model of projects selection is established in the technological innovation, the conventional methods have deficiencies in dealing with the fuzzy uncertainty. To improve the mathematical model, in this paper, Neuro-fuzzy decision tree(Neuro-FDT) is introduced to the research on the innovation projects selection . Fuzzy decision trees are powerful, top-down, hierarchical search methodology to extract human interpretable classification rules. However, they are very poor in classification accuracy. Neural networks-fuzzy decision tree improves FDT's classification accuracy and extracts more accuracy human interpretable classification rules. The fuzzy rules enable a decision-maker to decide the optimal projects selection of technological innovation. Comparing with the conventional methods, the mathematical model of neuro-fuzzy decision tree can be easily established by fully utilizing the information of projects. The result of the positive research indicated that this mathematical model is very valid for innovation projects selection and it will have a good application prospect in this area.