IFCEM based recognition method for target with interval-overlapped hybrid attributes

When the attributes of unknown targets are not just numerical attributes, but hybrid containing linguistic existing recognition methods effective. In addition, it is more difficult to identify densely distributed in feature space, especially when there interval overlap between attribute measurements different target classes. To address these problems, a novel method based on intuitionistic fuzzy comprehensive evaluation model (IFCEM) proposed. For database divided into individual classes and overlapping classes, for continuous interval-valued term set (CIVLTS) used describe characteristic. A cloud model-based an area-based proposed obtain decision information query respectively. An improved inverse weighted kernel c-means (IWK-FCM) algorithm solution weight vector. The possibility matrix applied determine identity category target. Finally, case study composed parameter sensitivity analysis, accuracy analysis. comparison with other methods, taken verify superiority method.