Nowadays there is a wide range of applications for WebGIS which can add great value to modern economic, and building WebGIS system for specific scenarios is the common requirement of the industry. While currently separate WebGIS systems are deployed at different sites and operated by different owners, each of which has the whole set of functionalities of WebGIS, and thus introduce high cost of development and maintenance, which is a waste of resources as most of the functionalities are the same or similar. An edge computing based WebGIS architecture is proposed in the paper to meet the need of customization by applying the idea of SaaS. In this distributed architecture, the resource load is reasonably balanced between the server and the browser, which improves the overall performance of the system. Also it utilizes edge computing to reduce the pressure on the server by sharing map tiles among WebGIS clients. The proposed WebGIS system can not only be well customized and personalized as it is edge computing base, but also be well usable for large number of visits due to its distributed feature. The experiments show 5 concurrent requests per second, as well as response speed increases by more than 38.6% against traditional deployment. INDEX TERMS edge computing, customizable, WebGIS, high performance, map tile sharing, deviceenhanced MEC.

As one of the commonly used data mining algorithms, K-means has the advantage of fast clustering speed, but the disadvantage is that it is less effective for clustering non-spherical data. An improved K-means algorithm (IK-means) is proposed to enhance clustering efficiency for non-spherical data. The original dataset is clustered into a relatively larger number of high-density sub-clusters, and the final result is obtained by merging connected sub-clusters respectively. The connectivity among sub-clusters is evaluated by the sub-clusters density and the nearest distance class between sub-clusters. By testing on University of California, Irvine(UCI) datasets and several other artificial simulation datasets, the comparison of proposed IK-means algorithm against DBSCAN, KGFCM shows its clustering capability for data of arbitrary shape. The clustering Adjusted Rand Index (ARI) value for 72,000 sizes data is 24% higher than DBSCAN, and 95.2% higher than KGFCM. For larger datasets, the IK-means algorithm is faster than DBSCAN and KGFCM.

No abstract

No abstract