Semi-supervised Learning for Discrete Choice Models

TL;DR

This paper presents semi-supervised learning algorithms for discrete choice models, improving calibration when limited labeled data is available, with applications demonstrated in hotel booking and airline shopping cases.

Contribution

It adapts classic semi-supervised algorithms and introduces new methods that automatically determine the number of clusters using BIC for discrete choice modeling.

Findings

Improved prediction accuracy in case studies

Algorithms effectively determine optimal cluster numbers

Reduced computational effort in large-scale problems

Abstract

We introduce a semi-supervised discrete choice model to calibrate discrete choice models when relatively few requests have both choice sets and stated preferences but the majority only have the choice sets. Two classic semi-supervised learning algorithms, the expectation maximization algorithm and the cluster-and-label algorithm, have been adapted to our choice modeling problem setting. We also develop two new algorithms based on the cluster-and-label algorithm. The new algorithms use the Bayesian Information Criterion to evaluate a clustering setting to automatically adjust the number of clusters. Two computational studies including a hotel booking case and a large-scale airline itinerary shopping case are presented to evaluate the prediction accuracy and computational effort of the proposed algorithms. Algorithmic recommendations are rendered under various scenarios.