The Exponential Dynamic Analysis of Network Attention Based on Big Data

Based on Baidu Index and Internet Big Data, this paper analyzes the overall relationship of information flow spatial network by using social network method, and finds that index-time has dual structural characteristics and keeps changing trend. The research shows that the scale, correlation degree and control efficiency level of information flow show obvious structural characteristics of time dissimilation. Secondly, we study the behavior characteristics of index, dynamically analyze the temporal and spatial changes of big data of network attention and index dynamics, observe the inflow of index data in July and August, and find that the maximum value of a single index reaches 250,457 times. The inflow was the highest in the third quarter, with the maximum index reaching 659,329 times, showing the peak state of the whole period. Through information flow, the correlation between data is revealed, and the time distribution characteristics of monthly peak and quarterly peak are analyzed. Finally, it is concluded that the attention of information flow network has a clear direction in the two dimensions of "index-time", showing its correlation.