Analysis Of Realistic Social Relation Network And Relation Search Technology

In recent years, with the rapid development of Internet Technology, relationships of people are mapping to the network and form social networks through interactions. It makes the real society and the virtual society continues to cross, and provides people with a new way of communication.In reality, everyone lives in a relation world. The various stages of life have different relationships, such as blood relationship like parents, relatives etc, geopolitical relationship among neighborhoods, villagers etc, occupational relationship on study and work, marriage relationship, and congenial relationship based on personal preferences. These relationships support the personal emotion world and affect our work and life. Relationship can bring huge commercial value. However, a person’s energy is limited, and someone can’t fully remember all relationships. It needs to establish a personal network system of social relations in order to manage individual relationship well. These relationships should be studied because it may play a role at a critical moment. At present, the research of social networks is mainly aim at virtual relationships. Therefore, the analysis and search technique of the realistic social network has important research significance and practical value.In this thesis, the social relations are taken as the key study object,and the domestic theory and methods of social network are analyzed. The main contributions are as follows:1. The relationship types of the realistic social relations are studied. It has a difference between virtual and the realistic social relations. According to the reality society, relationships are divided into five categories.2. Social network model is designed. The realistic social relation network is described as an weighted directed graph. Individuals are represented as nodes in the network and the directed edges are used to represent the relationship between the individuals. According to the characteristics of the relationship, the relationship between individuals is directed and asymmetric.3. The influence factors and quantification pattern of closeness degree are analyzed. In the network, the degree of intimacy between nodes is expressed as closeness degree. Through the research of the relationship, the influence factors of the closeness degree are divided into six aspects, such as trust, interaction frequency, duration of relationship, degree of reciprocity, character, profession. Influence factors are quantified with fuzzy analysis method.4. The calculation model of closeness degree is designed. The closeness can be used for relationship search. The fuzzy theory is used to establish the calculation model of nodes closeness degree. Closeness degree and cumulative closeness degrees are calculated.5. Search strategies are designed in the thesis. The high degree search strategy which is based on closeness degree and the multipath search strategy are offered. Experiments show that average path length of high degree search strategy based on closeness degree is lower and success rate is higher compared with Random-Walk-Search and maximum degree search. Various paths can be conducted by the multipath search strategy.