The Ternary Closure Of The Drug Domain From The Perspective Of Opportunity Trust Motivation Research On The Influence Mechanism Of Formation

Purpose:The theory of ternary closures is of great significance in social networks.As the smallest local structure and link generation mechanism in networks,ternary closures have the characteristics of structural balance and stability,and play an important role in the evolution of knowledge networks.Exploring their formation mechanisms is of great significance for mining tacit knowledge in networks.Currently,scholars have conducted a large amount of research on the formation mechanism of ternary closures in social networks,and believe that the formation of ternary closures in social networks can be attributed to three reasons-opportunity,trust,and motivation.However,there is little research on the formation mechanism of ternary closures in the field of undirected knowledge networks.Based on this,this study takes the pharmaceutical knowledge network as an example,introduces the perspective of the three mechanisms of opportunity,trust,and motivation that affect the formation of ternary closures in social directed networks,defines and measures the three mechanisms of opportunity,trust,and motivation in the pharmaceutical knowledge network in combination with social network indicators,and deeply explores the impact mechanism of the formation of ternary closures in the pharmaceutical field,Provide theoretical basis research for predicting potential tacit knowledge and network state changes in drug knowledge networks.Methods:This study obtained journal literature related to adverse drug reactions of metformin from 1991-2020 in the Pub Med database,and constructed a network of 10 slices from1991-2020 and a network of 10 time overlapping windows from 2000-2009.Based on Pearson correlation coefficients,the correlation between the values of three types of mechanisms,edge clustering coefficients,and the number of ternary closures is tested;Introduce more node attributes(proximity centrality of node pairs,intermediary centrality,and feature vector centrality)and network characteristics(average path length),and apply econometric methods to thoroughly examine the impact of node attributes and network characteristics on the three mechanisms.Results:Based on Pearson correlation coefficient testing,the correlation between the values of three types of mechanisms and the edge clustering coefficient and the number of ternary closures shows that in 10 time slice networks,the opportunity values of node pairs have a strong positive correlation with the edge clustering coefficient,with Pearson correlation coefficients greater than 0.5(p<0.001),respectively 0.9966,0.9941,0.9980,0.9965,0.9977,0.9983,0.9978,0.9981,0.9984,0.9978;The trust value of node pairs has a strong positive correlation with the number of ternary closures,with Pearson correlation coefficients greater than 0.5(p<0.001),respectively 0.9216,0.9619,0.9723,0.9686,0.9633,0.9737,0.9732,0.9664,0.9729,0.9741;The motivation value of node pairs has a strong positive correlation with the number of ternary closures,with Pearson correlation coefficients greater than 0.5(p<0.001),respectively 0.8456,0.8633,0.9195,0.9107,0.9047,0.9232,0.9305,0.9234,0.9336,and 0.9371.The regression analysis of node attributes,network characteristics,and inter node mechanisms using econometric methods shows that the proximity centrality of node pairs has a negative impact on opportunity and trust,and a positive impact on motivation mechanisms.The regression coefficients are-0.0504(p<0.001),-0.0384(p<0.001),and0.0277(p<0.001),respectively;The mediating centrality of node pairs has a positive impact on opportunity,trust,and motivation value mechanisms,with regression coefficients of 0.0219(p<0.001),0.4451(p<0.001),and 0.5465(p<0.001),respectively;The feature vector centrality of node pairs has a positive impact on opportunity,trust,and motivation mechanisms,with regression coefficients of 0.0984(p<0.001),0.2972(p<0.001),and 0.4878(p<0.001),respectively;The average path length of the network has a negative impact on the node pair opportunity mechanism,and a positive impact on the node pair trust and motivation mechanism.The regression coefficients are-0.2396(p<0.001),0.2253(p<0.001),and 0.2676(p<0.001),respectively.Conclusion:In this study,we re measure the impact mechanisms of opportunity,trust,and motivation on the formation of ternary closures in the pharmaceutical field,and explore the impact of node attributes and network characteristics on the three mechanisms in the pharmaceutical field.We further explore the formation mechanism of ternary closures in the pharmaceutical undirected weighted network.The study found that the index measures of the three mechanisms are all reasonable and can better reflect the number of node pairs with ternary closures;Node attributes and network characteristics have a certain promoting or inhibiting effect on the three mechanisms.These new feature discoveries can provide a further breakthrough direction for the study of the influence mechanism of ternary closures.