Identifying regional characteristics of transportation research with Transport Research International Documentation (TRID) data.

• We study how transportation research topics correlate with geographic regions using bibliographic data. • The collected dataset covers 174,848 research articles in 1091 journals and hundreds of conferences. • A geoRegion-topic model is proposed to derive the research topic distribution for each geographic region. • Transportation research exhibits region-specific characteristics at the country/region and U.S. state levels. • The relation between research topic similarity and geographic proximity depends on the geographic level. Since the planning and operations of transportation systems depend on factors ranging from economic to political, which tend to vary significantly over geographic regions (geoRegions), the following research question arises: do transportation topics in various geographic regions differ significantly? A review of existing transportation bibliographic studies shows that the information on the geographic location of study areas is seldom available and thus rarely analyzed. This study thus presents a geoRegion-topic model that extends a widely used topic model, latent Dirichlet allocation, to explore the region-specific characteristics of transportation research based on large-scale bibliographic data collected from Transport Research International Documentation. Top geographic regions that are frequently used as study areas in transportation research are identified at both the country/region and U.S. state levels. The results confirm that transportation research exhibits clear region-specific characteristics, namely each geographic region is associated with a unique transportation topic distribution. After exploring the relation between topic similarity and geographic proximity, we find in general countries/regions that are geographically proximate have quite similar transportation topic distributions, while this correlation cannot be observed at the U.S. state level. Other relevant findings are also reported, along with potential bias in the analysis. [ABSTRACT FROM AUTHOR]