This conference presentation discusses an ensemble approach for meaningful location identification from geotagged social media content through a developed R package. Extracting such ‘meaningful’ locations from geotagged data traces is a cutting-edge research area. However, this field is currently using mostly bespoke, custom approaches that are difficult to verify and/or reproduce. To bring clarity and consistency to this field, we present an easy-to-use software package (written within the R ecosystem, which is one of the industry-standard software/languages for this type of analysis) that allows users to either use several built-in algorithms or easily define their own algorithms. This also enables users to choose an ensemble approach (i.e. using multiple algorithms) to increase the reliability of the extracted locations.Extracting such locations is an essential step in the analysis of geotagged data traces and is most often used to extract home and office locations based on the spatio-temporal behavior of users. Although we use social media data in this presentation, the same methods can be applied to other data sources such as cell phone data or mobile app data that have similar characteristics. Once home locations are extracted, these datasets can then be used to analyze and predict urban mobility patterns, for different groups of people and for different neighborhoods.