Study of Spatial Autocorrelation Changes Based on Newly Movement Data for the Purpose of Discovering Patterns of Human Inter-city Movement

The discovery of patterns of human movement in inner-city environments is one of the most important parameters in studies such as urban planning and geospatial studies. One of the sources that are widely used today to explore patterns of human movement is movement-based social media data. These media provide a huge amount of data in two dimensions of time and space. The purpose of this study is to explore and survey the hidden patterns of human inter-urban movement based on movement data derived from human daily activities in the process of sharing information on location-based social media and taking into account the semantic dimension of the data. In this study, movement data from the foursquare social media is used to provide a semantic dimension to the data. In order to discover the hidden patterns of human inter-urban movement, the capability and efficiency of spatial-temporal autocorrelation analysis have been evaluated. In this research, using statistical analysis and considering the time dimension in the first stage, a significant process of changes in the spatial-temporal autocorrelation of the studied data is revealed with respect to the urban subdivision based on Thiessen polygonization method. Secondly, the problem of the trend of spatial-temporal autocorrelation changes and the relationship between information sharing, location and urban area at different times of day, in order to extract precise intra-urban movement patterns using semantic clustering of location-based data has been examined in the most prominent patterns of urban movement in different time periods. The results of this study demonstrate the high capability of spatial-temporal autocorrelation analyzes based on the semantic dimension of movement data derived from foursquare social media in discovering hidden patterns of human movement at the urban level.