Inductive reasoning with type-constrained encoding for emerging entities.

Knowledge graph reasoning, vital for addressing incompleteness and supporting applications, faces challenges with the continuous growth of graphs. To address this challenge, several inductive reasoning models for encoding emerging entities have been proposed. However, they do not consider the multi-batch emergence scenario, where new entities and new facts are usually added to knowledge graphs (KGs) in multiple batches in the order of their emergence. To simulate the continuous growth of knowledge graphs, a novel multi-batch emergence (MBE) scenario has recently been proposed. We propose a path-based inductive model to handle multi-batch entity growth, enhancing entity encoding with type information. Specifically, we observe a noteworthy pattern in which entity types at the head and tail of the same relation exhibit relative regularity. To utilize this regularity, we introduce a pair of learnable parameters for each relation, representing entity type features linked to the relation. The type features are dedicated to encoding and updating the features of entities. Meanwhile, our model incorporates a novel attention mechanism, combining statistical co-occurrence and semantic similarity of relations effectively for contextual information capture. After generating embeddings, we employ reinforcement learning for path reasoning. To reduce sparsity and expand the action space, our model generates soft candidate facts by grounding a set of soft path rules. Meanwhile, we incorporate the confidence scores of these facts in the action space to facilitate the agent to better distinguish between original facts and rule-generated soft facts. Performances on three multi-batch entity growth datasets demonstrate robust performance, consistently outperforming state-of-the-art models.

Clinical manifestations and EEG findings in children infected with COVID-19 and exhibiting neurological symptoms.

BACKGROUND: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) infection has many neurological manifestations, and its effects on the nervous system are increasingly recognized. There has been no systematic analysis of electroencephalography (EEG) characteristics in children exhibiting neurological symptoms of Coronavirus disease 2019 (COVID-19). The primary aim of this study was to describe the EEG characteristics caused by COVID-19 infection in children who were showing neurological symptoms and to assess the relationship between COVID-19-related EEG changes and clinical features in these children. METHOD: This study included 125 pediatric patients infected with SARS-CoV2 and showing neurological symptoms, and their continuous EEG was recorded. In addition, the demographic and clinical characteristics of these patients were analyzed and the correlation between the two was investigated. RESULTS: Abnormal EEG findings were detected in 31.20% (N = 39) of the patients. Abnormal discharges (43.59%) were the most common EEG abnormalities, followed by background abnormalities (41.03%). The proportion of patients diagnosed with febrile seizure was higher in the normal EEG group than in the abnormal EEG group (P = 0.002), while the opposite was true for epilepsy and encephalitis/encephalopathy (P = 0.016 and P = 0.003, respectively). The independent associated factors of abnormal EEG were age and total length of stay (P < 0.001 and P = 0.003, respectively). Non-specific EEG abnormalities were found in COVID-19-related encephalitis/encephalopathy. CONCLUSION: Our study corroborated that a small group of pediatric patients infected by COVID-19 and showing neurological symptoms may exhibit abnormal EEG. This study could help improve the understanding of clinical and EEG characteristics in children with COVID-19 and inform triage policies in other hospitals during the COVID-19 pandemic.