Results are based on an average of the 3-fold cross validation. The top performing ML model and their metrics are highlighted for the comparison. SVM denotes for Support Vector Machines, NB for Naive Bayes, k-NN for K-Nearest Neighbour, LR for Logistic Regression, RF for Random Forest.

Existing methods for calculating directional relations in polygons (i.e. the directional similarity model, the cone-based model, and the modified cone-based model) were compared to human perceptions of change through an online survey. The results from this survey provide the first empirical validation of computational approaches to calculating directional relations in polygonal spatial data. We have found that while the evaluated methods generally agreed with each other, they varied in their alignment with human perceptions of directional relations. Specifically, translation transformations of the target and reference polygons showed greatest discrepancy to human perceptions and across methods. The online survey was developed using Qualtrics Survey Software, and participants were recruited via online messaging on social media (i.e., Twitter) with hashtags related to geographic information science. In total sixty-one individuals responded to the survey. This survey consisted of nine questions. For the first question, participants indicated how many years they have worked with GIS and/or spatial data. For the remaining eight questions, participants ranked pictorial database scenes according to degrees of their match to query scenes. Each of these questions represented a test case that Goyal and Egenhofer (2001) used to empirically evaluate the directional similarity model; participants were randomly presented with four of these questions. The query scenes were created using ArcMap and contained a pair of reference and target polygons. The database scenes were generated by gradually changing the geometry of the target polygon within each query scene. The relations between the target and reference polygon varied by the type of movement, the scaling change of the polygon, and changes in rotation. The scenarios were varied in order to capture a representative range of variability in polygon movements and changes in real world data. The R statistical computing environment was used to determine the similarity value that corresponds with each database scene based on the directional similarity model, the cone-based model, and the modified cone-based model. Using the survey responses, the frequency of first, second, third, etc. ranks were calculated for each database scene. Weight variables were multiplied by the frequencies to create an overall rank based on participant responses. A rank of one was weighted as a five, a rank of two was weighted as a four, and so on. Spearman’s rank-order correlation was used to measure the strength and direction of association between the rank determined using the three models and the rank determined using participant responses.

As governing bodies continue to explore mileage fees as an alternative to the gas tax, of the uncertainty surrounding public support remains a critical barrier to policy uptake. This study examines the extent to which public perceptions of mileage fees are guided by misinformation or lack of information using a national, internet-based survey. We use hypothetical voting opportunities to gather respondent support for mileage fees, coupled with educational treatments that address mileage fee fairness, privacy, and costs. The findings indicate that respondents are largely misinformed or lack information about mileage fees and the gas tax. Pre-education, only 32% of respondents supported the policy, but post-education, 46% of respondents supported the policy. Through binomial, multinomial, and fixed effect modeling, we examined the factors associated with policy support, changes in policy support, and the educational treatments. Ultimately, our findings indicate that education can play a key role in increasing support for a mileage fee policy as an alternative to the gas tax. Methods An internet-based survey was used to assess nation-wide support for replacing state gas taxes with a mileage fee. Respondents were given three opportunities to vote for or against a mileage fee replacement, with educational treatments in between votes. The impact of education on respondent voting was evaluated using a variety of regression modelling methods. Respondents were recruited to the survey through Qualtrics. This company used quota-based sampling schemes to field the survey to every U.S. state. Since this research hypothesized that mileage fee opinions may be in part due to low information about mileage fees, we opted to omit respondents from states where widespread mileage fee education or mileage fee policies were implemented. As of July 2023, we identified California, Oregon, Utah and Hawaii as states where residents were likely meaningfully more educated about mileage fees and chose not to survey those populations. Three versions of the survey were released, each proposing mileage fees are collected using a different method. The three options proposed collecting mileage fees using (1) an annual odometer reading, (2) a plug-in device without GPS technology, and (3) a plug-in device with GPS technology. Besides differing in the method displayed for collecting mileage information, the surveys were identical.