Concept-cognitive learning is a concept learning method that simulates human cognitive process based on formal concept analysis. Most of the current concept-cognitive learning methods only consider conceptual similarity and ignore the influence of prior decision information, resulting in the loss of practical details. To solve this problem, a concept-cognitive learning model based on decision significance is put forward for concept classification in a dynamic environment by extracting prior decision information to describe the significance of decision making. The neighborhood granule is constructed by cosine similarity, and the progressive process of concept cognition is discussed. For the dynamic environment, the decision significance and confidence degree are proposed to design the computational method of concept classification with the consideration of the validity of the a priori decision information. The effectiveness and superiority of the proposed method are verified by simulation experiments.

Automatic citation intent classification is one of hot issues in the field of bibliometrics.The existing citation intention classification models engender the limitations in extracting textual features and fusing citation contextual features and citation external features. Therefore, a citation intent classification method based on MPNet pretraining and multi-head attention feature fusion is proposed. The position compensation structure is introduced to improve the masked language model and permuted Language model.The syntactic word-frequency features and structure features of citations are combined. A feature extraction method is proposed for citation intent classification task. The multi-head attention mechanism is introduced for feature fusion to improve the classification accuracy. The experimental results on ACL-ARC datasets demonstrate that the proposed method achieves better performance in citation intent classification task with robustness on the unbalanced data.