Anomaly detection for open data quality improvement

It is noted that an increasing number of Open Data (OD) projects are making governmental and corporate data available to public with free access and reuse. One of the barriers of getting benefits from OD is the quality of published data. This problem and its causes are analyzed, metrics and strategies of improvement of the quality of OD are considered, the general strategy using anomaly detection techniques and its' implementation for cases of time and categorical contexts are proposed.