Making Legacy Open Data Machine Readable by Crowdsourcing

An approach is described for converting legacy statistical data in image format into a machine-readable and reusable format by using crowdsourcing. Requesting crowd workers not only to extract tables from graph images but also to reconstruct them in spreadsheets can produces structures including attribute names and values as properties of the reconstructed graph objects. A quality control mechanism was developed that improves the accuracy of extracted tables by aggregating tables created by different workers for the same chart image and by utilizing the data structures obtained from the reproduced chart objects. Experimental results using the White Paper on Tourism published by the Japan Tourism Agency demonstrated that the proposed approach is effective.