Research On Centrality Word Cloud Layout Method For LBS Interest Points

With the development of mobile Internet and location-based technology,Location Based Service(LBS)has become more and more closely connected with people’s life and travel.People usually use LBS-enabled mobile applications to explore the surrounding points of interest(POI),such as attractions,hotels and supermarkets.Due to the large number of these points of interest,rich attributes and complicated semantics,targeted visualization methods are needed to express the spatial distribution,thematic semantics and coupling with the user’s location to better assist the user’s spatial perception of the surrounding environment.To address this problem,this paper applies the ideas of "geographic word cloud" and "deformation map" to the visualization of LBS points of interest,and proposes a visualization method for multidimensional information expression of points of interest--centrality word cloud.The centrality word cloud uses a center-force oriented algorithm to dynamically layout the textual symbols that characterize the points of interest.The method first sets the search conditions for the points of interest in the specified area,then extracts the names of the points of interest as the shape contours of the text symbols involved in the layout based on the search results,and maps the size and color of the symbols to the thematic attributes of the points of interest such as categories,ratings,reviews,etc.;then selects a layout center point(a geographic feature point characterizing the area,or the current location of the user in the LBS),and places all the points of interest in the same location as the center point.All the interest points are placed at the position overlapping with the center point;under the initial layout,the interest points overlap with each other,and a force-oriented model is constructed using the center-forceoriented algorithm to simulate the moving trajectory of the interest point symbols,and the text symbols move one by one under the action of attractive and repulsive forces in the model,and stop moving when they reach the suitable placement position;during the moving process,to prevent the overlapping conflicts between different interest points,this study adopts the idea of quadtree to prevent the conflict between different interest points.In order to prevent the overlapping conflicts between different points of interest during the movement,this study adopts the idea of quadtree to segment the shape contour of the text for optimizing the collision detection between symbols,and finally obtains the adjusted visual layout effect.In order to verify the readability and accuracy of the layout results,three indexes such as compactness,orientation retention and relevance are calculated and volunteers are invited to evaluate the generated layout in five aspects: aesthetics,compactness,recognizability,satisfaction and task completion,and the results show that this method can show the multidimensional thematic characteristics of interest points more intuitively;express the distribution of interest points more completely;accurately retain the center point and the results show that this method can present the multidimensional thematic characteristics of interest points in a more intuitive way;express the distribution of interest points in a more complete way;and accurately retain the real relative position relationship between the center point and interest points.It presents the user with a compact and informative visual layout of interest points,and highlights the information of interest to the user,helping him/her to quickly browse the key information in the graph.More importantly,the centrality word cloud proposed in this paper can show the correlation between points of interest and user’s location(e.g.public transportation travel distance,driving time,accessibility,etc.),highlighting the intrinsic connection between points of interest and "central points".This paper couples the thematic information of the point of interest(identification information,location information,evaluation information and other attributes)in the same layout space,uses text as the main symbolic marker,and integrates various visual variables to express the multidimensional thematic information of the point of interest,thus realizing a pan-map representation for the purpose of improving map cognitive efficiency.