This paper introduces a multi-feature fusion graph convolutional network (MFFGCN) for the accurate prediction of pedestrian trajectories. Unlike previous methods that primarily relied on distance-based interactions, MFFGCN considers a broader range of influencing factors, including movement direction, relative distance, and movement velocity. By integrating these factors, the proposed model is able to more comprehensively capture the complex social interactions among pedestrians. The model comprises several modules: a pedestrian movement direction interaction module, distance-based and velocity-based interaction graph modules, a graph convolutional network (GCN) for feature extraction, and a temporal convolutional network (TCN) for trajectory prediction. Experimental results on the ETH and UCY datasets demonstrate that MFFGCN outperforms state-of-the-art methods, achieving lower average displacement error (ADE) and final displacement error (FDE). The proposed model not only improves the accuracy of pedestrian trajectory prediction but also provides insights into the importance of considering multiple factors in modeling pedestrian interactions. Our source code is available at: https://github.com/XiaojingChen327/MFFGCN-model/tree/main.

In different contexts, one abstract concept (e.g., fruit) may be mapped into different concrete instance sets, which is called abstract concept instantiation. It has been widely applied in many applications, such as web search, intelligent recommendation, etc. However, in most abstract concept instantiation models have the following problems: (1) the neglect of incorrect label and label incompleteness in the category structure on which instance selection relies; (2) the subjective design of instance profile for calculating the relevance between instance and contextual constraint. The above problems lead to false prediction in terms of abstract concept instantiation. To tackle these problems, we proposed a novel model to instantiate the abstract concept. Firstly, to alleviate the incorrect label and remedy label incompleteness in the category structure, an improved random-walk algorithm is proposed, called InstanceRank, which not only utilize the category information, but it also exploits the association information to infer the right instances of an abstract concept. Secondly, for better measuring the relevance between instances and contextual constraint, we learn the proper instance profile from different granularity ones. They are designed based on the surrounding text of the instance. Finally, noise reduction and instance filtering are introduced to further enhance the model performance. Experiments on Chinese food abstract concept set show that the proposed model can effectively reduce false positive and false negative of instantiation results.