When the traditional decision tree algorithm is applied to the ordinal classification task, there are two problems:the traditional decision tree algorithm does not introduce the order relation, so it cannot learn and extract the order structure of the dataset; in real life, there is a lot of fuzzy but not exact knowledge, however the traditional decision tree algorithm cannot deal with the data with fuzzy attribute value. To solve these problems, an ordinal decision tree algorithm based on fuzzy advantage complementary mutual information was proposed. Firstly, the dominant set was used to represent the order relations in the data, and the fuzzy set was introduced to calculate the dominant set for forming a fuzzy dominant set. The fuzzy dominant set was able to not only reflect the order information in the data, but also obtain the inaccurate knowledge automatically. Then, the complementary mutual information was generalized on the basis of fuzzy dominant set, and the fuzzy advantage complementary mutual information was proposed. Finally, the fuzzy advantage complementary mutual information was used as a heuristic method, and an decision tree algorithm based on fuzzy advantage complementary mutual information was designed. Experimental results on 5 synthetic datasets and 9 real datasets show that, the proposed algorithm has less classification errors compared with the classical decision tree algorithm on the ordinal classification tasks.

Monte Carlo Tree Search (MCTS) shows excellent performance in chess game problem. Most existing studies only consider the success and failure feedbacks and assum that the results follow the Bernoulli distribution. However, this setting ignores the usual result of draw, causing inaccurate assessment of the disk status and missing of optimal action. In order to solve this problem, Ternary Multi-Arm Bandit (TMAB) model was constructed and Best Arm identification of TMAB (TBBA) algorithm was proposed. Then, TBBA algorithm was applied to Ternary Minimax Sampling Tree (TMST). Finally, TBBA_tree algorithm based on the simple iteration of TBBA and Best Action identification of TMST (TTBA) algorithm based on transforming the tree structure into TMAB were proposed. In the experiments, two arm spaces with different precision were established, and several comparative TMABs and TMSTs were constructed based on the two arm spaces. Experimental results show that compared to the accuracy of uniform sampling algorithm, the accuracy of TBBA algorithm keeps rising steadily and can reach 100% partially, and the accuracy of TBBA algorithm is basically more than 80% with good generalization and stability and without outliers or fluctuation ranges.

Active contour model is widely used in image segmentation and object contour extraction, and the edge-based Geodesic Active Contour (GAC) model is widely used in the object extraction with obvious edges. But the process of GAC evolution costs many iterations and long time. In order to solve the problems, the GAC model was improved with Bessel filter theory. Firstly, the image was smoothed by Bessel filter to reduce the noise. Secondly, a new edge stop term was constructed based on the edge detection function of Bessel filter and incorporated into the GAC model. Finally, the Reaction Diffussion (RD) term was added to the constructed model for avoiding re-initialization of the level set. The experimental results show that, compared with several edge-based models, the proposed model improves the time efficiency and ensures the accuracy of segmentation results. The proposed model is more suitable for practical applications.

The objects with inhomogeneous intensity or multi-gray intensity by using active contour, a supervised active contour algorithm named KSOAC was proposed based on Kernel Self-Organizing Map (KSOM). Firstly, prior examples extracted from foreground and background were input into KSOM for training respectively, and two topographic maps of input patterns were obtained to characterize their distribution and get the synaptic weight vector. Secondly, the average training error of unit pixel of two maps were computed and added to energy function to modify the contour evolution; meanwhile, the controlling parameter of energy item was obtained by the area ratio of foreground and background. Finally, supervised active contour energy function and iterative equation integrated with synaptic weight vectors were deduced, and simulation experiments were conducted on multiple images using Matlab 7.11.0. Experimental results and simulation data show that the map obtained by KSOM is closer to prior example distribution in comparison with Self-Organizing Map (SOM) active contour (SOAC), and the fitting error is smaller. The Precision, Recall and F-measure metrics of KSOAC are higher than 0.9, and the segmentation results are closer to the target; while the time consumption of KSOAC is similar to that of SOAC. Theoretical analysis and simulation results show that KSOAC can improve segmentation effectiveness and reduce target leak in segmenting images with inhomogeneous intensity and objects characterized by many different intensities, especially in segmenting unknown probability distribution images.