| Researches on social networks and preference models have already penetrated many aspects of our lives.Although much of the existing researches have revealed some of their characteristics,there are still many deficiencies regarding their internal laws and mechanisms.In response to these deficiencies,a small-scale social network model that integrates various social network characteristics,coupled with a modeling method for social network group preferences based on generalized triadic closure were put forward in this thesis.The following several tasks were accomplished in this thesis:(1)To delve into the deficiencies of traditional social network models,and a small-scale social network model that integrates multiple social network characteristics was proposed by means of combining the complexity,community and similarity of social networks.(2)On the basis of small-scale social network models,a simulation of the time series of social network node participation in activities was achieved.In accordance with the similarity of time series,a new social network was built in the form of a triadic closure to conduct ample analysis.Finally,an optimized community partition algorithm was used to achieve the partition of preference communities,thus completing the entire process of preference modeling.According to the experiment findings,the small-scale social network model proposed in this thesis was more closely aligned with real social networks in terms of various complex network statistical characteristics such as degree distribution,average distance,and clustering coefficient;Meanwhile,the preference modeling method presented in this thesis has exhibited a certain degree of accuracy and stability. |
