The accuracy of information network partition is not high and the characteristics of metapath cannot represent the attributes of network nodes in the existing academic citation recommendation algorithms.In order to so...The accuracy of information network partition is not high and the characteristics of metapath cannot represent the attributes of network nodes in the existing academic citation recommendation algorithms.In order to solve the problems,a similarity measurement algorithm,community merging and time effect PathSim(CMTE-PathSim),based on community merging and time effect is proposed.On the premise of dividing heterogeneous information network(HIN)effectively,the algorithm considers the influence of node information on the characteristics of metapath.The results of Top-k query verify the effectiveness of CMTE-PathSim on real datasets and improve the quality of citation recommendation.展开更多
Choosing an effective classification and recognition method in a large protein database plays a crucial role in the classification of enzymes.In previous studies on enzyme classification,only node characteristic infor...Choosing an effective classification and recognition method in a large protein database plays a crucial role in the classification of enzymes.In previous studies on enzyme classification,only node characteristic information of amino acid were generally considered in the process of model training.The characteristics of amino acid nodes and topological structure in enzyme protein structure are proposed in this paper.The model was trained by graph neural network.By comparing with K nearest neighbor,support vector machine,random forest and multi-layer perceptron,it is shown that the graph neural network method has great advantages.The accuracy obtained by graph neural network is obviously higher than others.展开更多
基金Supported by the Scientific Research Fund of Liaoning Provincial Education Department(L2019048)Talent Scientific Research Rund of LSHU(2016XJJ-033)of China.
文摘The accuracy of information network partition is not high and the characteristics of metapath cannot represent the attributes of network nodes in the existing academic citation recommendation algorithms.In order to solve the problems,a similarity measurement algorithm,community merging and time effect PathSim(CMTE-PathSim),based on community merging and time effect is proposed.On the premise of dividing heterogeneous information network(HIN)effectively,the algorithm considers the influence of node information on the characteristics of metapath.The results of Top-k query verify the effectiveness of CMTE-PathSim on real datasets and improve the quality of citation recommendation.
基金Supported by the Scientific Research Fund of Liaoning Provincial Education Department(L2019048)Talent Scientific Research Fund of LSHU(2016XJJ-033)of China。
文摘Choosing an effective classification and recognition method in a large protein database plays a crucial role in the classification of enzymes.In previous studies on enzyme classification,only node characteristic information of amino acid were generally considered in the process of model training.The characteristics of amino acid nodes and topological structure in enzyme protein structure are proposed in this paper.The model was trained by graph neural network.By comparing with K nearest neighbor,support vector machine,random forest and multi-layer perceptron,it is shown that the graph neural network method has great advantages.The accuracy obtained by graph neural network is obviously higher than others.
