Arti Thakur

About me:

I am a full-time research scholar at KIXLab(KAIST, School of computing), working with Prof Juho Kim. I, previously, worked as an android app developer with Samsung R&D Noida, India for more than 3 years.

I am interested in analyzing people's everyday behavior(e.g. engaging/avoiding social interaction, being active/inactive) revealed by technology. This involves extracting human behaviors such as their social engagement, activeness etc. in natural settings and predicting personal characteristics such as extraversion. In order to capture behaviors unobtrusively, I develop self-tracking technologies which are effective in capturing natural behaviors of people.

I am also interested in CMC (Computer-Mediated Communication e.g. expressing an opinion online, engaging in discussion). The work involves finding out the relationships between users behavior patterns and various factors such as platform affordances, content, community characteristics etc. Long term goal of this work is to enhance collaborative learning and engagement in these discussions environments.

Keywords: #HCI #Computer Mediated Communication #Physical Crowdsourcing #Human Centered Computing #Learning #Ubiquitous Computing #Mobile Computing#Large Scale Datamining #Collaborative learning

Resume: link to my CV

Research Projects:

Extracting workplace personalities with Online and Physical Data(KIXLab, current) : We extract workplace personality holistically by using users online and offline behavior data. We argue that social behavior extracted at workplace(online and offline), are related to extraversion. more..

Supporting learning and collaboration in online discussion(KIXLab, current) : Online discussions on various platforms such as new article comment section, on social media(e.g. facebook, twitter, reddit) are often very fragmented. We aim to build a platform which is designed to help users to learn and collaborate to generate richly labeled and well structured arguments while discussing online. more..

Crowdsourcing techniques to find Contextual Emotion Labels in videos(KIXLab) : Dialog videos contain rich contextual, emotional, and intentional cues of the characters and their surroundings. In this project we aim to find a crowdsourcing technique to leverage these rich emotion labels. more..

Emotion Analysis through Context Understanding: We developed a "scenario based" approach that focuses on created such situation with dominant emotion, using a set of context information from sensor data. Verification of each scenario is done when application detect the scenario and asks user to tag his/her current prevailing emotion.


2004-2007: B.Sc.physics from Delhi University(India).

2008-20011: MCA(Master of computer application) from BIT, Mesra(India)

Work Experience:

Oct, 2011- March, 2015 : Software Developer, Samsung R&D, Noida, India

Nov, 2015- march 2016 : NCLab, School of computing, KAIST

Aug, 2017- current : KIXLab, School of computing, KAIST


Hindi, English, Korean-Beginner level.


I enjoy reading and writing short stories, sketching, and trying different foods. Apart from these I love travelling. Some of the countries I have travelled so far are South Korea, Vietnam, USA and Maldives.