Wenjun Ma

Dempster’s combination rule in Dempster–Shafer theory is a powerful and effective tool for multi-sensor data fusion. However, counterintuitive results are possible under the condition of a high conflict between pieces of evidence. This study demonstrates that existing conflict measurements cannot prevent such results and, thus, proposes a quantitative conflict measurement based on the concept of essential conflict. This work analyzes two characteristics of the essential conflict, namely belief absolutization and uncorrectable assertions. In addition, considering the desirable properties of the measurement, this study demonstrates that the measurement of essential conflict can reveal the essence of counterintuitive results in Dempster’s combination process. Finally, properties and examples are used to validate the proposed measurement.

The success of current entity alignment (EA) tasks largely depends on the supervision information provided by labeled data. Considering the cost of labeled data, most supervised methods are challenging to apply in practical scenarios. Therefore, an increasing number of works based on contrastive learning, active learning, or other deep learning techniques have been developed, to solve the performance bottleneck caused by the lack of labeled data. However, existing unsupervised EA methods still face certain limitations; either their modeling complexity is high or they fail to balance the effectiveness and practicality of alignment. To overcome these issues, we propose a learnable convolutional attention network for unsupervised entity alignment, named LCA-UEA. Specifically, LCA-UEA performs convolution operations before the attention mechanism, ensuring the acquisition of structural information and avoiding the superposition of redundant information. Then, to efficiently filter out invalid neighborhood information of aligned entities, LCA-UEA designs a relation structure reconstruction method based on potential matching relations, thereby enhancing the usability and scalability of the EA method. Notably, a similarity function based on consistency is proposed to better measure the similarity of candidate entity pairs. Finally, we conducted extensive experiments on three datasets of different sizes and types (cross-lingual and monolingual) to verify the superiority of LCA-UEA. Experimental results demonstrate that LCA-UEA significantly improved alignment accuracy, outperforming 25 supervised or unsupervised methods, and improving by 6.4% in Hits@1 over the best baseline in the best case.