Extending the web mining in both scientific and general fields, our aim is to outline the topics appearing in the join domains of the two fields of Health and urban contexts, that are not or low covered by stakeholders (actors) who could be national, regional, or local governments, international or national organization or association (like NGOs), health institutions or private companies.

A large sample was collected from the internet allowing to process and analyze more than 123,000 pages, by Topic modelling analysis (machine learning) allowing to build graphs of the most discussed themes by the scientific literature and by the different public or private actors (nearly 5'500 actors). These graphs have been built by continent and for general approaches over 4 time periods from 2000 to 2021.