Constructing the Three Graphs for the Large-Scale Heterogeneous Information System

The paper proposes a bunch of social capital oriented perspectives to resolve the issue of accessing the large-scale heterogeneous information over the Internet. It believes the most critical task for that is to construct the three social graphs that connect the human capital components, i.e., the data, the channel and the human. As the underlying infrastructure, the graphs evolve upon various behaviors of information accessing. Moreover, it also sustains those behaviors in terms of exhibiting high quality information with high performance. As the dominance of the system, the three social graphs play the role of a broker between the physical capital and the human capital. However, it supposes that neither the graphs can be created without sufficient users' participations nor their growing procedures can be imitated by any algorithms. To initiate the system and foster its growth, it puts forward the three approaches, i.e., the rough, the strict and the compromised, to construct the seed graphs. In them, the compromised one is feasible because of its low cost and strengthening information accessing. It also describes the practical implementation of the algorithms and the relevant evaluation solutions.