This paper proposes a method that combines multiple models and high-confidence dictionary for event nugget detection. This method introduces dictionary features into maximum entropy model and conditional random fields model respectively, then combines the results of two models. In addition, the lexical length and the length of the dependency path between the trigger and negation or speculation in event realis recognition are considered to improve the accuracy of event realis detection. Compared to the method based on maximum entropy model, the experiment results show that proposed method can get 6.43% gain of F1 in event nugget recognition and 1.69% gain of F1 in event realis recognition.
To explore the moderating role of interpretative bias in the relation of attentional bias and social anxiety. In study 1, a positive attentional training program, using a modified dot-probe task, was used to modify the attentional bias in a nonclinical sample of students. After two days training, results revealed no different change on self-reported anxiety. The participants showed preference for positive information post-training, while avoidance pre-training in the 500 ms condition. Based on the founding of study 1, data collected from college students were used to investigate the relationship among attentional bias, interpretative bias and social anxiety by regression analysis in study 2. There was a significant interaction of interpretative bias by attentional bias scores, which meant the existence of moderating effect. Attentional bias can predict social anxiety under high interpretative bias condition, but not in individuals with low interpretative bias. The results provide a new perspective of interpretative bias to view the influences of attentional bias on social anxiety.