Abstract:
There is a voluminous data generated by people every day on geoportals which is being used tremendously. These geoportals are the life line to many people to get quick information about the location including address, geographical location, category and reviews of the place as well. The geo tagged data available on these portals is useful for better planning to improve in various sectors such as business, public services, event planning and disaster management. However, there is no way to verify the information available on these portals as these are not always contributed by domain experts. This data is collected through crowd sourcing therefore it is an important to make sure the accuracy of this crowd sourced data available on the geoportals. All existing verification methods are based on the crowd sourced information which again can’t claim as the reliability of this data. To address this challenge, it becomes vital to explore the insight of geotagged data available on geoportal to understand the features affecting the accuracy most. There must be some ground truth data which can be taken as the reference data while exploring the crowd sourced data. In this paper, the exploratory statistical analysis is provided for crowd sourced geotagged data including the data set preparation using ground truth data as a reference data. In-depth analysis on prepared dataset will lay the foundation for developing automated accuracy measuring models which makes this vast data more useful for planning various citizen centric services specifically in developing nations.
Keywords: