摘要：China has a huge volume of historical resources on its contemporary history. Lots of valuable knowledge are hidden in those resources and cannot be utilized easily. It is an urgent problem to mine the implicit semantic knowledge scattered in a large number of historical resources and to reorganize the historical knowledge and facts in a fined-grained manner, so that can help user to explore the historical knowledge for research and education. This paper proposes a method, which is called “Mining down, Organizing up”, to semantically represent and organize historical knowledge on contemporary China hidden in historical encyclopedia text. Based on the proposed historical ontology of contemporary Chinese, this method extracts knowledge objects and facts from the unstructured historical text items by utilizing text mining technologies, represents the historical knowledge in semantically enriched way, and interlinks the related historical knowledge objects and facts to form a historical knowledge network of the contemporary China. By mining the historical facts and the historical knowledge network, the authors get more valuable patterns from the historical knowledge which could be used to form the new organization scheme to reorganize the historical knowledge in a more vivid way. Based on this method, the authors developed a system which can represent and organize historical knowledge of contemporary China in a fined-grained manner, support user to explore historical knowledge by providing functions such as semantic retrieval, historical objects and facts clustering, visualization navigation, association analysis, and chronicle facts reconstruction etc.
摘要：Novelty seeking (NS) is a personality trait reflecting excitement in response to novel stimuli. High NS is usually a predictor of risky behaviour such as drug abuse. However, the relationships between NS and risk-related cognitive processes, including individual risk preference and the brain activation associated with risk prediction, remain elusive. In this fMRI study, participants completed the Tridimensional Personality Questionnaire to measure NS and performed a probabilistic decision making task. Using a mathematical model, we estimated individual risk preference. Brain regions associated with risk prediction were determined via fMRI. The NS score showed a positive correlation with risk preference and a negative correlation with the activation elicited by risk prediction in the right posterior insula (r-PI), left anterior insula (l-AI), right striatum (r-striatum) and supplementary motor area (SMA). Within these brain regions, only the activation associated with risk prediction in the r-PI showed a correlation with NS after controlling for the effect of risk preference. Resting-state functional connectivity between the r-PI and r-striatum/l-AI was negatively correlated with NS. Our results suggest that high NS may be associated with less aversion to risk and that the r-PI plays an important role in relating risk prediction to NS.
摘要：Nicotine addiction is associated with risky behaviors and abnormalities in local brain areas related to risky decision-making such as the dorsal anterior cingulate cortex (dACC), anterior insula (AI), and thalamus. Although these brain abnormalities are anatomically separated, they may in fact belong to one neural network. However, it is unclear whether circuit-level abnormalities lead to risky decision-making in smokers. In the current study, we used task-based functional magnetic resonance imaging (fMRI) and examined resting-state functional connectivity (RSFC) to study how connectivity between the dACC, insula, and thalamus influence risky decision-making in nicotine addicts. We found that an increase in risky decision-making was associated with stronger nicotine dependence and stronger RSFC of the dACC-rAI (right AI), the dACC-thalamus, the dACC-lAI (left AI), and the rAI-lAI, but that risky decision-making was not associated with risk level-related activation. Furthermore, the severity of nicotine dependence positively correlated with RSFC of the dACC-thalamus but was not associated with risk level-related activation. Importantly, the dACC-thalamus coupling fully mediated the effect of nicotine-dependent severity on risky decision-making. These results suggest that circuit-level connectivity may be a critical neural link between risky decision-making and severity of nicotine dependence in smokers.