Skip to main content
To KTH's start page

Enriching Automated Travel Diaries Using Biometric Information

Time: Mon 2019-11-18 10.00

Location: E2, Lindstedtsvägen 3, Stockholm (English)

Subject area: Transport Systems

Doctoral student: Robin C. O. Palmberg , Systemanalys och ekonomi

Opponent: Biträdande Professor Clas Rydergren, Linköpings Universitet, Institutionen för teknik och naturvetenskap (ITN), Kommunikations- och transportsystem (KTS)

Supervisor: Docent Yusak O. Susilo, Systemanalys och ekonomi; Lektor Gyözö Gidofalvi, Geoinformatik

Export to calendar

Abstract

The methods for collecting travel data about travellers today incorporate either fully manual or semi-automatic elements, which makes the methods susceptible to errors. The travellers might respond subjectively rather than objectively or even wholly incorrect, albeit with or without purpose. For certain types of studies, these are still valid methods for collecting data. However, for specific target groups, it might be hard to respond using these methods, either because of physical or psychological limitations.

One of these target groups that is increasing rapidly is elderly in general, and dementia patients in particular, who suffer from fluctuating cognitive skills and memory. These conditions affect the recipient’s ability to answer truthfully and correctly. However, in the strive to form more accessible urban environments, the information regarding the need and behaviour of the said target group is crucial, meaning that new methods for collecting travel data need to be created.

The three papers included in this licentiate thesis present the development and trial of a new method for fully automated data collection using biometric data as a dimension. The method attempts to determine how the recipient is affected by the elements presented to them while they travel, such as the built environment, based on the variations in the biometric data dimension.

With the rapid advancements in information and communication technology, many new artefacts which open for new possible methods of data collection has been launched and are widely available. The methods and artefacts are not capable of meeting the requirements for the type of data collection method that would be needed to cater to the target group by themselves. However, by combing several types of currently available artefacts and methods, it is theoretically possible to cover the gaps of each artefact and method to create versatile methods for data collection (Paper I).

Such methods require tools for physical operationalisation. An exploratory development process has led to the creation of a software tool which could be used with several types of consumer hardware, which means that it would theoretically be possible to conduct extensive surveys fast with low costs where participants utilise their own hardware (Paper II).

In order to uncover the usefulness of the tool, an analysis was conducted on a limited dataset which had been collected as a result of a trial of the tool. In an attempt to prove the hypothesis “it is possible to understand how much the dimensions of data collected in specific locations affect the stress of travellers using heart rate as the dependent variable”, data-driven methods of data analysis were explored and utilised. Simple clustering methods, which disregarded any weighting on the dimensions, uncovered if there was any valuable information in the dataset at all. A model had to be created in order to understand better how the different dimensions of the collected data affected the participant (Paper III).

This set of papers should indicate whether this type of method is feasible to pursue with the current means of widely available technology and what sort of significance the collected data might hold when analysed with appropriate analysis methods.

urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-262880