Open data constantly produced by cities are thought to belong
to the taxpayers as well as to the city itself, therefore there
is a need for easy accessibility for all. This trend is
supported by the state government as well as the European
Union. The issue is,  the data that are supposed to be
published are often not integrated enough. Different
departments and workers are in charge of different datasets.
Employees who work with raw data can lack needed technical
expertise and have little time for this, on top of their normal
workload. That can give rise to mistakes and later deliverance
of open data to people. The goal of this diploma thesis was to
automate the process of open data publishing.
Automation was performed by developing scripts that ensure the
whole process of data publishing. Different algorithms were
used for different datasets. Raw data were processed to erase
all sensitive information that would prevent lawful publishing.
Moreover, the form of data was adjusted to fit national
recommendations. The developed scripts start up automatically
in defined intervals and provide input into portal CKAN.
In collaboration with city Žďár nad Sázavou we were able to
automate the publishing of three selected city datasets,
including publishing all accessible data from the past. In
conclusion, we compared the estimated overall costs of manual
data publishing to automatic publishing by developing a script
and presented future visions.