Wearable sensing analysis : identifying alcohol drinking from daily physiological data

In practical psychology research, questionnaires and interviews with examinees are commonly used. Besides, to study about certain human behaviors such as our subject -- alcohol drinking craving, it usually lasts for several days which could be costing and time consuming, to both researchers and examinees. Previous work in our lab provided a smartphone-client and server based distributed system which does not need participants to go to the clinic doing any paperwork nor any invasive blood or medical-electronic based tests, so that saving time and money for better researching purposes. The system called mAAS and started to collect precious data from our recruits. Improvements that I made to the system are presented in this thesis. Also, a new data analysis pipeline is built and evaluated. Modern wearable technology is integrated without extra burdensome equipment in our system. Simple, they would be just smartphone and a vest suite with sensor in it, those are already our daily-carry personal stuffs. Smartphone application will be in charge of communicating with sensor suite and server wirelessly and safely. We then can get large numbers of detailed physiological information about participants in their real life. Data preprocessing is applied to get cleaned data. Features about ECG and RIP are extracted afterwards. As a result, involving machine learning methods, we could apply classification on these features and give out pretty good performance on understanding and analyzing tasks. In this thesis work, alcohol drinking activities are identified from daily body physiological data through the pipeline as presented.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Computer science (MU)

Grantor dc:publisher

University of Missouri--Columbia

Year dc:date.issued

2016

Author dc:creator

• Zhang, Chen

Advisor dc:contributor.advisor

• Shang, Yi, 1967-

Statement dc:rights

• OpenAccess.

Language dc:language.iso

eng, English

Handle dc:identifier.uri

https://hdl.handle.net/10355/57613

OAI identifier oai:identifier

oai:mospace.umsystem.edu:10355/57613