Social media texts spontaneously produced and uploaded by the public contain a wealth of disaster information. As a supplementary data source for remote sensing, they have played an important role in disaster reduction and emergency response in recent years. However, social media also has certain flaws, such as insufficient location information, etc. This affects the efficiency of combining these data with remote sensing data. For flood disasters in particular, extensively flooded areas limit the distribution of social media data, which makes it difficult for these data to function as they should. In this paper, we propose a disaster reduction framework to solve these problems. We first used an approach that was based on search engine and lexical rules to automatically extract disaster-related location information from social media texts. Then, we combined the extracted information with the upload location of social media itself to construct location-pointing relationships. These relationships were used to build a new social network, which can be used in combination with remote sensing images for disaster analysis. The analysis integrated the advantages of social media and remote sensing. It can not only provide macro disaster information in the study area but can also assist in evaluating the disaster situation in different flooded areas from the perspective of public observation. In addition, the timeliness of social media data also improved the continuity and situational awareness of flood monitoring. A case study of the flood disaster in the Yangtze River Basin in China in 2020 was used to verify the effectiveness of the method described in this paper.

Emergency situation awareness during sudden natural disasters presents significant challenges. Traditional methods, characterized by low spatial and temporal resolution as well as coarse granularity, often fail to comprehensively capture disaster situations. However, social media platforms, as a vital source of social sensing, offer significant potential to supplement disaster situational awareness. This paper proposes an innovative framework for disaster situation awareness based on multimodal data from social media to identify social media content related to typhoon disasters. Integrating text and image data from social media facilitates near real-time monitoring of disasters from the public perspective. In this study, Typhoon Haikui (Strong Typhoon No. 11 of 2023) was chosen as a case study to validate the effectiveness of the proposed method. We employed the ERNIE natural language processing model to complement the Deeplab v3+ deep learning image semantic segmentation model for extracting disaster damage information from social media. A spatial visualization analysis of the disaster-affected areas was performed by categorizing the damage types. Additionally, the Geodetector was used to investigate spatial heterogeneity and its underlying factors. This approach allowed us to analyze the spatiotemporal patterns of disaster evolution, enabling rapid disaster damage assessment and facilitating emergency response efforts. The results show that the proposed method significantly enhances situational awareness by effectively identifying different types of damage information from social sensing data.

Logs record valuable data from different software and systems. Execution logs are widely available and are helpful in monitoring, examination, and system understanding of complex applications. However, log files usually contain too many lines of data for a human to deal with, therefore it is important to develop methods to process logs by computers. Logs are usually unstructured, which is not conducive to automatic analysis. How to categorize logs and turn into structured data automatically is of great practical significance. In this paper, LTmatch algorithm is proposed, which implements a log pattern extracting algorithm based on a weighted word matching rate. Compared with our preview work, this algorithm not only classifies the logs according to the longest common subsequence(LCS) but also gets and updates the log template in real-time. Besides, the pattern warehouse of the algorithm uses a fixed deep tree to store the log patterns, which optimizes the matching efficiency of log pattern extraction. To verify the advantages of the algorithm, we applied the proposed algorithm to the open-source data set with different kinds of labeled log data. A variety of state-of-the-art log pattern extraction algorithms are used for comparison. The result shows our method is improved by 2.67% in average accuracy when compared with the best result in all the other methods.

In disaster emergency response, spatial location information embedded within social media texts holds substantial value for the rapid localization of affected areas and the implementation of precise rescue operations. Existing research predominantly employs natural language processing and deep learning technologies for geographic information extraction; however, two critical limitations persist: first, insufficient integration of textual semantic features for disaster relevance determination, resulting in inadequate correlation between extracted results and actual disaster locations; second, absence of mechanisms for identifying affected sites in multi-location contexts, thereby compromising decision support efficacy. Addressing these challenges, this study proposes a hierarchical disaster location information extraction framework that integrates semantic understanding. The framework operates through a three-tier hierarchy: data-level adversarial augmentation, semantic-level dynamic parsing, and parameter-level scale optimization. It achieves three core functionalities: (1) precise determination of disaster relevance for geographic location information; (2) identification of affected areas in multi-location contexts; (3) establishment of a logarithmic scaling relationship between LLM parameter scale and optimal prompt sample size.

Image inpainting is a crucial task in computer vision that aims to restore missing and occluded parts of damaged images. Deep-learning-based image inpainting methods have gained popularity in recent research. One such method is the deep image prior, which is unsupervised and does not require a large number of training samples. However, the deep image prior method often encounters overfitting problems, resulting in blurred image edges. In contrast, the second-order total generalized variation can effectively protect the image edge information. In this paper, we propose a novel image restoration model that combines the strengths of both the deep image prior and the second-order total generalized variation. Our model aims to better preserve the edges of the image structure. To effectively solve the optimization problem, we employ the augmented Lagrangian method and the alternating direction method of the multiplier. Numerical experiments show that the proposed method can repair images more effectively, retain more image details, and achieve higher performance than some recent methods in terms of peak signal-to-noise ratio and structural similarity.

Disaster-related social media posts often report casualties, rescue needs, infrastructure damage, shelter demand, and local situation changes earlier than formal channels, yet their brevity and noise make operational classification difficult. This study examines whether a practical and reproducible classification pipeline can support fine-grained, emergency-oriented semantic recognition under a deliberately conservative evaluation setting. We convert 14,392 English tweets from CrisisSense-LLM into six actionable semantic categories, partition the data by proxy event groups, and compare a TF-IDF logistic-regression baseline, supervised BERT-base fine-tuning, and zero-shot natural language inference. The evaluation is further extended to mapped HumAID data, a manually reviewed 177-post Chinese boundary-test set, a Chinese-to-English translation bridge, and a fixed-budget selective adjudication simulation. BERT-base achieves the best held-out main-test performance (Macro-F1 = 0.8824), outperforming TF-IDF (0.6133) and zero-shot inference (0.3581), and reaches 0.8132 Macro-F1 on HumAID without retraining. Direct English-to-Chinese transfer is ineffective, whereas multilingual BERT and translation bridging improve Chinese Macro-F1 to 0.2684 and 0.3603, respectively. With 600 reviewed posts, selective adjudication reaches 0.7792 Macro-F1 on the main test set and 0.7153 on HumAID. These findings indicate that the central contribution is not a new model architecture, but an empirically validated workflow that combines supervised fine-tuning, leakage-aware evaluation, external validation, cross-lingual stress testing, and information-driven human review. The novelty therefore lies in the reproducible integration of data mapping, group-aware evaluation, external and cross-lingual stress testing, and selective human review into a single emergency-oriented assessment workflow.