LotRec: A Recommender for Urban Vacant Lot Conversion
This addresses resource allocation for city planners dealing with vacant lots, but it is incremental as it applies existing recommender methods to a new domain-specific dataset.
The paper tackles the problem of recommending whether and what to convert urban vacant lots into, using determinants of conversion, and achieves mean F-measures of 90% for conversion prediction within a city, 91% for conversion type prediction within a city, and 85% for cross-city conversion prediction.
Vacant lots are neglected properties in a city that lead to environmental hazards and poor standard of living for the community. Thus, reclaiming vacant lots and putting them to productive use is an important consideration for many cities. Given a large number of vacant lots and resource constraints for conversion, two key questions for a city are (1) whether to convert a vacant lot or not; and (2) what to convert a vacant lot as. We seek to provide computational support to answer these questions. To this end, we identify the determinants of a vacant lot conversion and build a recommender based on those determinants. We evaluate our models on real-world vacant lot datasets from the US cities of Philadelphia,PA and Baltimore, MD. Our results indicate that our recommender yields mean F-measures of (1) 90% in predicting whether a vacant lot should be converted or not within a single city, (2) 91% in predicting what a vacant lot should be converted to, within a single city and, (3) 85% in predicting whether a vacant lot should be converted or not across two cities.