[LSI 운영위원회 연사초빙] KAIST 전산학과 / 오혜연 교수님

서윤영, 2012-06-22 12:46:33

조회 수
추천 수



KAIST 전산학과 오혜연 교수님







sentiment & emotion analysis


I will present our research results in two related topics.


First, I will present a computational framework for understanding the

social aspects of emotions in Twitter conversations. We explore

in-depth questions of the emotional patterns in conversational

interactions. We look for meaningful patterns of emotional exchanges

in a conversation, and those patterns may depend on the topics and

words of the conversation. We also hypothesize that conversational

partners can influence each others' emotions and topics. Further, we

discover interesting patterns in the overall emotions, affected by the

lexical usages of the interlocutors. To find these patterns, we

develop a novel computational framework, based on LDA, to discover the

emotions from an unannotated corpus of Twitter conversations, and we

evaluate the model with a human-annotated corpus. We find that

conversational partners usually express the same emotion, but when

they do not, one of the conversational partners tends to respond with

a positive emotion to a negative emotion rather than vice versa. We

also show that tweets containing sympathy, apology, and complaint are

significant emotion influencers. Finally, we discover lexical

patterns, such as the usage of profanity, that influence the overall

emotion of a conversation.


Second, I will present research findings on the relationship between

tie strength and self-disclosure. In social psychology, it is

generally accepted that one discloses more of his/her personal

information to someone in a more intimate and trusting relationship,

often called a "strong tie" in the social network literature. We

question and study how tie strength affects self-disclosures in the

context of Twitter conversations. To estimate the tie strength between

Twitter conversational partners, we propose several metrics based on

tweeting behavior, common friendships, and network topology. To

measure the degree of self-disclosure, we look at topics, emotions,

sentiments, lexical patterns, and personally identifiable information

(PII) and personally embarrassing information (PEI). Our results

illustrate that in general, there is a significant trend that validate

the findings in social psychology; in strong tie relationships,

Twitter users show significantly more frequent behaviors of

self-disclosure. However, there are interesting exceptions to this

general trend; in very strong and very weak tie relationships measured

by network centralities, some of the self-disclosure behaviors deviate

far from the general trend. We analyze and discuss these results in

detail and propose directions for further studies of tie strength and

self-disclosure in Twitter and other SNS users.



0 댓글

Board Menu


Page 1 / 1
번호 제목 글쓴이 날짜 조회 수

[LSI 운영위원회 연사초빙] KAIST 건설및환경공학과 / 손훈 교수님

  • file
서윤영 2012-10-05 4255

[LSI 운영위원회 연사초빙] KAIST 화학과 / 이해신 교수님

  • file
서윤영 2012-10-05 3737

[LSI 운영위원회 연사초빙] KAIST 전기및전자공학과 / 김종환 교수님

  • file
서윤영 2012-10-05 32360

[LSI 운영위원회 연사초빙] KAIST 전산학과 / 오혜연 교수님

  • file
서윤영 2012-06-22 27245

[LSI 운영위원회 연사초빙] Parsons대학 / Erin Cho 교수님

  • file
서윤영 2012-06-18 10685

Board Links

Page Navigation

  • 1