"Is not the truth the truth?": Analyzing the Impact of User Validations for Bus In/Out Detection in Smartphone-based Surveys

TL;DR

This study evaluates the robustness of smartphone-based BIBO passenger-state classification using BLE RSSI and GPS signals under realistic ground-truth label noise. It combines a semi-controlled field test with two autonomous buses, video-ground-truth, and a Wizard-of-Oz setup to model Poisson-like labeling errors, then benchmarks Random Forest and Multi-layer Perceptron classifiers under BLE and GPS features with rigorous OOS validation. The results show BLE-based BIBO is relatively tolerant to label noise and RF generally outperforms MLP, while GPS offers the best raw accuracy at higher energy cost; under label flipping, classifier bias can be substantial if high-quality ground-truth is unavailable. The work highlights the importance of accounting for labeling noise in smartphone-based travel surveys and introduces an imputation strategy for BLE gaps to support scalable, energy-efficient proximity sensing in urban environments.

Abstract

Passenger flow allows the study of users' behavior through the public network and assists in designing new facilities and services. This flow is observed through interactions between passengers and infrastructure. For this task, Bluetooth technology and smartphones represent the ideal solution. The latter component allows users' identification, authentication, and billing, while the former allows short-range implicit interactions, device-to-device. To assess the potential of such a use case, we need to verify how robust Bluetooth signal and related machine learning (ML) classifiers are against the noise of realistic contexts. Therefore, we model binary passenger states with respect to a public vehicle, where one can either be-in or be-out (BIBO). The BIBO label identifies a fundamental building block of continuously-valued passenger flow. This paper describes the Human-Computer interaction experimental setting in a semi-controlled environment, which involves: two autonomous vehicles operating on two routes, serving three bus stops and eighteen users, as well as a proprietary smartphone-Bluetooth sensing platform. The resulting dataset includes multiple sensors' measurements of the same event and two ground-truth levels, the first being validation by participants, the second by three video-cameras surveilling buses and track. We performed a Monte-Carlo simulation of labels-flip to emulate human errors in the labeling process, as is known to happen in smartphone surveys; next we used such flipped labels for supervised training of ML classifiers. The impact of errors on model performance bias can be large. Results show ML tolerance to label flips caused by human or machine errors up to 30%.

Figures (11)

• Figure 1: Experiment workflow.
• Figure 2: Sensing platform.
• Figure 3: Random Forest one flip experiment AUC Classification task using BLE RSSI signal only (p-values $<< 0.01$)
• Figure 4: Random Forest one flip experiment AUC Classification task using GPS signal only (p-values $<< 0.01$)
• Figure 5: Multi Layer Perceptron one flip experiment AUC Classification task using BLE RSSI signal only (p-values $<< 0.01$)