DOC PREVIEW
UCF EEL 6788 - EmotionSense - A Mobile Phones based Adaptive Platform for Experimental Social Psychology Research

This preview shows page 1-2-3 out of 10 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

IntroductionSensing Human EmotionsSystem OverviewEmotionSense at a GlanceEmotionSense ManagerSpeaker and Emotion Recognition ComponentSensor MonitorsAction and Knowledge BaseInference EngineImplementationSpeaker Recognition SubsystemEmotion Recognition SubsystemAdaptation FrameworkEvaluationPerformance BenchmarksSpeaker RecognitionEmotion RecognitionDataset for Sensor BenchmarksSocial Psychology ExperimentOverview of the ExperimentResults and DiscussionRelated WorkConclusionsREFERENCESEmotionSense: A Mobile Phones based Adaptive Platformfor Experimental Social Psychology ResearchKiran K. RachuriComputer LaboratoryUniversity of [email protected] MusolesiSchool of Computer ScienceUniversity of St. [email protected] MascoloComputer LaboratoryUniversity of [email protected] J. RentfrowFaculty of Politics, Psychology, Sociologyand International StudiesUniversity of [email protected] LongworthDepartment of EngineeringUniversity of [email protected] AucinasComputer LaboratoryUniversity of [email protected]’s mobile phones represent a rich and powerful com-puting platform, given their sensing, processing and commu-nication capabilities. Phones are also part of the everydaylife of billions of people, and therefore represent an excep-tionally suitable tool for conducting social and psychologicalexperiments in an unobtrusive way.In this paper we illustrate EmotionSense, a mobile sens-ing platform for social psychology studies based on mobilephones. Key characteristics include the ability of sensingindividual emotions as well as activities, verbal and prox-imity interactions among members of social groups. More-over, the system is programmable by means of a declara-tive language that can be used to express adaptive rules toimprove power saving. We evaluate a system prototype onNokia Symbian phones by means of several small-scale ex-periments aimed at testing performance in terms of accuracyand power consumption. Finally, we present the results ofreal deployment where we study participants emotions andinteractions. We cross-validate our measurements with theresults obtained through questionnaires filled by the users,and the results presented in social psychological studies us-ing traditional methods. In particular, we show how speakersand participants’ emotions can be automatically detected bymeans of classifiers running locally on off-the-shelf mobilephones, and how speaking and interactions can be correlatedwith activity and location measures.ACM Classification KeywordsH.1.2 User/Machine Systems, J.4 Social and Behavioral Sci-ences, I.5 Pattern Recognition.General TermsAlgorithms, Design, Experimentation.Permission to make digital or hard copies of all or part of this work forpersonal or classroom use is granted without fee provided that copies arenot made or distributed for profit or commercial advantage and that copiesbear this notice and the full citation on the first page. To copy otherwise, orrepublish, to post on servers or to redistribute to lists, requires prior specificpermission and/or a fee.UbiComp ’10, Sep 26-Sep 29, 2010, Copenhagen, Denmark.Copyright 2010 ACM 978-1-60558-843-8/10/09...$10.00.Author KeywordsEmotion Recognition, Speaker Recognition, Social Psychol-ogy, Mobile Phones, Energy Efficiency.INTRODUCTIONMobile phones represent an ideal computing platform to mon-itor behavior and movement, since they are part of the every-day life of billions of people [1]. Recently, systems such asCenceme [22] and Betelgeuse [16] have shown the poten-tial of mobile phone sensing in providing information suchas user movement and activity for recreational and health-care applications. One possible use of these technologies isarguably the support to sociology experiments [20] whichinvolve studying people’s daily life and interactions. In thepast, this analysis has been performed with the help of cam-eras (in home/working environments or in laboratories), byusing voice recorders attached to people, and self reports us-ing daily diaries or PDAs [6]. However, these techniquesmay lead to biased results since people are aware of beingconstantly monitored. Instead, mobile phones offer an un-obtrusive means of obtaining information about the behaviorof individuals and their interactions.In this paper, we present EmotionSense, a framework forcollecting data in human interaction studies based on mo-bile phones. EmotionSense gathers participants’ emotions aswell as proximity and patterns of conversation by processingthe outputs from the sensors of off-the-shelf smartphones.This can be used to understand the correlation and the impactof interactions and activities on the emotions and behavior ofindividuals. In terms of system design, the key characteris-tics of this framework are programmability (social scientistscan describe the sensing tasks using a declarative language),and run-time adaptation (social scientists can write rules toactivate and deactivate sensors according to the user con-text). Although energy efficient sensing has previously beeninvestigated in works such as [30], this is the first paper topropose a fully context-aware programmable mobile sensingsystem for social psychology research. Social scientists canmodify the behavior of the system both in terms of sensingoperations based on the analysis of the available informationabout the user, and its environment. For example, they canwrite a rule to activate a voice sensor only if people are closeby to the user.More specifically, the key contributions of this work can besummarized as follows:• We design, implement, deploy, and evaluate a completesystem for experimental sociology and psychology that isable to provide information about social group dynamics,especially with respect to the influence of activity, groupinteractions, and time of day on the emotions of the indi-viduals, in an unobtrusive way.• We present the design of two novel subsystems for emo-tion detection and speaker recognition built on a mobilephone platform. These subsystems are based on Gaus-sian Mixture methods [27] for the detection of emotionsand speaker identities. EmotionSense automatically rec-ognizes speakers and emotions by means of classifiersrunning locally on off-the-shelf mobile phones.• We propose a programmable adaptive system with declar-ative rules. The rules are expressed using first order logicpredicates and are interpreted by


View Full Document

UCF EEL 6788 - EmotionSense - A Mobile Phones based Adaptive Platform for Experimental Social Psychology Research

Documents in this Course
Load more
Download EmotionSense - A Mobile Phones based Adaptive Platform for Experimental Social Psychology Research
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view EmotionSense - A Mobile Phones based Adaptive Platform for Experimental Social Psychology Research and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view EmotionSense - A Mobile Phones based Adaptive Platform for Experimental Social Psychology Research 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?