[Purpose/significance]In the crisis situation, users are often troubled by information overload. In order to alleviate its adverse effects, this study proposes an information barrier construction method based on information relevance, quality, and influence. The topic model is used to classify the information, and the topics and information unrelated to the crisis events are filtered.[Method/process]Based on the ELM and crisis communication matrix as well as the existing information quality evaluation indicators, an evaluation system for the quality of microblog information in the crisis situation is constructed. According to the external and internal characteristics of microblog under the crisis, we construct secondary indicators from the perspective of source credibility and information content quality, and use the entropy method to determine the weight of each indicator. The number of reposts, likes, and trust indicators are used to measure the influence of microblogging. An information barrier based on information influence is constructed.[Result/conclusion]Taking the Omicron COVID-19 outbreak as an example, this study conducted topic modeling, quality, and impact assessment on microblogging, and filtered irrelevant, low-quality, and low influence information. The information barrier construction method proposed in this study is helpful to mitigate the adverse effects of information overload in crisis situations.

[Purpose/significance]This study aims to solve the problem of traditional short text matching models being difficult to apply to long text matching tasks such as legal case retrieval. [Method/process]For the task of legal case matching, this paper proposes a Legal Text Matching model based on RoFormer （LTMR）. In the coding layer, the legal case is encoded through the RoFormer model and the legal feature extractor. In the reasoning layer, the context and interactive information of long text are further extracted by using interactive attention and self-attention mechanisms. We conducted the empirical research by applying the proposed model to the CAIL2019-SCM dataset. [Result/conclusion]Compared to the baseline methods, the LTMR model achieved the best results. The research sheds light on promoting the application of legal case matching.