Offline location search using Reverse K-Mean Clustering & GSM communication

Abstract

Converting Navigation to Offline thereby designing it more user-friendly with wider accessibility to a wider range of people is the main motivation behind the work. This paper proposes a Reverse K-Mean Clustering approach that utilizes the concept of k-mean clustering of finding centroids and works inversely as the traditional k-mean to search for the locations in the geographical dataset (map). A 2-d false grid with respect to the Jaipur City map has been designed in which the centroid and the clusters of the corresponding locations would be highlighted. The main idea is to input the centroid (k-centroids for k-clusters) from the ground reference positions through the mouse click on the geographical map. If the desired location is found then the image of the particular location which is loaded in the system gets displayed. A GSM feature has also been integrated which uses MSComm Ports, AT commands and Bluetooth technology to directly call the desired location's number that is displayed below the location's image. This application serves the user as an offline navigation system and allows the user to easily search for the locations of Jaipur City such as petrol pumps, hospitals, police-stations, fire stations, railway stations, bus stations. It neither requires any prior internet connection for finding out places nor any GPS technology or other such connectives [1][2]