English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Logit models are usually applied when studying individual travel behavior,i.e., to predict travel mode choice and to gain behavioral insights on travelerpreferences. Recently, some studies have applied machine learning to modeltravel mode choice and reported higher out-of-sample predictive accuracy thantraditional logit models (e.g., multinomial logit). However, little researchfocuses on comparing the interpretability of machine learning with logitmodels. In other words, how to draw behavioral insights from thehigh-performance "black-box" machine-learning models remains largely unsolvedin the field of travel behavior modeling.  This paper aims at providing a comprehensive comparison between the twoapproaches by examining the key similarities and differences in modeldevelopment, evaluation, and behavioral interpretation between logit andmachine-learning models for travel mode choice modeling. To complement thetheoretical discussions, the paper also empirically evaluates the twoapproaches on the stated-preference survey data for a new type of transitsystem integrating high-frequency fixed-route services and ridesourcing. Theresults show that machine learning can produce significantly higher predictiveaccuracy than logit models. Moreover, machine learning and logit models largelyagree on many aspects of behavioral interpretations. In addition, machinelearning can automatically capture the nonlinear relationship between the inputfeatures and choice outcomes. The paper concludes that there is great potentialin merging ideas from machine learning and conventional statistical methods todevelop refined models for travel behavior research and suggests some newresearch directions.

Modeling Stated Preference for Mobility-on-Demand Transit: A Comparison of Machine Learning and Logit Models