Using spatial and machine learning analyses to assess satisfaction with life in an urban historical downtown area

The research identifies the perceptions of socioeconomic factors which influence how satisfaction with life is seen in the historical downtown area in Quito (HDQ). Using data from a survey carried out in the HDQ, two machine learning (ML) models were applied to identify these perceptions: artificial neural networks and a decision tree. Additionally, spatial autocorrelation analysis was applied to identify hotspots relating to satisfaction with life. Both ML models had similar error values. Satisfaction with health, work, and socioeconomic status were found to impact satisfaction with life. There were marked hotspots related to satisfaction with life within the study area. Our findings provide key information for the urban planning of the HDQ and may be a useful reference for public policies which improve quality of life in historical urban areas.