A Comparison of KNN Algorithm and MNL Model for Mode Choice Modelling

Abstract

Mode choice modelling helps to identify potential users of traffic and plays an important role in policy and decision-making by the government. With the advancement of artificial intelligence and machine learning techniques, several studies were carried out to analyse the performance of mode choice models in which the backpropagation algorithm was used. However, for faster convergence of parameters, it would be interesting to explore other efficient algorithms of machine learning as the conjugated gradient search in spite of the backpropagation algorithm. The present study adds to the literature about the performance of the K-Nearest Neighbour (KNN) algorithm in mode choice modelling and compared the KNN model with the traditional MNL model. It was unveiled that the variables, which are found significant and important in both models are the same. It is also found that the KNN model is outperforming MNL with a prediction accuracy of 73.84%.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 9 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.