About Me

My name is Lu Chen and I am a PhD student at the Ohio Center of Excellence in Knowledge-enabled Computing (Kno.e.sis Center) at Wright State University. My advisor is Dr. Amit P. Sheth.

I am broadly interested in Text Mining, Natural Language Processing and Social Media Analytics, with a particular focus on Sentiment Analysis and Extraction of Subjective Information. My research applies multidisciplinary methods to understand people, e.g., who they are, what they think, how they feel, through analyzing what they say and how they behave in social media space.

Recent Activities

My paper "Clustering for Simultaneous Extraction of Aspects and Features from Reviews" has been accepted to appear at NAACL HLT 2016.

I was awarded an "Outstanding Reviewer Award" for WSDM 2016. There were 21 reviewers (among all the 376 reviewers) nominated by the Senior PC members for this award, and I am so honored to be one of them :D

I received the "Outstanding Student Award" for the 2014-2015 academic year from the Department of Computer Science and Engineering.

We organized a workshop at ICWSM 2015 -- Religion on Social Media.


  • Intern, Big Knowledge, Samsung Research America (SRA), Silicon Valley, CA, USA. June 2014 - Aug. 2014
  • Intern, Social Computing, Qatar Computing Research Institute (QCRI), Doha, Qatar. Feb. 2014 - May 2014
  • Intern, Big Knowledge, Samsung Research America (SRA), Silicon Valley, CA, USA. May 2013 - Aug. 2013
  • Instructor, College of Computer and Information Science, Southwest University (SWU), Chongqing, China. Sep. 2006 - Aug. 2010


Subjectivity and Sentiment Analysis: I work on understanding and modeling different types of subjectivity, and explore techniques for capturing, analyzing and utilizing subjective information (e.g., sentiment, opinion, emotion) in text.

  • Target-specific Sentiment Analysis: Extract sentiment expressions (both formal and slang words/phrases) for a given target from unlabeled corpora. Develop an optimization-based approach to assess the target-dependent polarity of sentiment expressions. ( ICWSM 2012 )
  • Aspect and Feature Extraction for Opinion Mining: Develop a clustering algorithm and a domain-specific similarity measure to simultaneously identify product features and group them into aspect categories from online reviews. ( NAACL 2016 )
  • Emotion Classification: Use Twitter hashtags to create a large training dataset with self-labeled emotions (e.g., joy, sadness, anger, love, etc.). Explore effective features for emotion classification. ( SocialCom 2012, BII 2012 )
  • Improving Data Quality for Emotion Classification: Improve annotation quality by iteratively and interactively correcting mislabeled instances using active learning. Explore feature weighting techniques and a non-linear distribution spreading algorithm to recognize annotation errors. ( ACL 2014 )
  • Domain Adaptation for Emotion Identification: Leverage self-labeled Twitter data to identify emotions from text in different domains, e.g., blogs, news, tweets, etc. (in submission)
  • Extracting Users’ Medication/Treatment/Drug Experiences from Web Data: Extract users' subjective experience of using medication and treatment (e.g., for epilepsy or asthma) from web forum and social media posts. Extract drug users’ sentiment about non-medical use of pharmaceutical opioids (e.g. OxyContin, Loperamide, etc.) from web forum posts. ( JBI2013, JDAD2012 )

Social Media Analytics: I work on collecting data from Social Media and analyzing the data to gain insights into a wide range of topics about people and the society.

  • Religiosity and Well-being: Study the U.S. religious landscape on Twitter based on the analysis of over 250k U.S. Twitter users who self-declared their religions/belief. Examine the effects of religion on happiness expressed in tweets. Explore the topic preference and word usage of each group of users and how these affect their happiness. ( SocInfo 2014. Media Coverage: Washington Times, Washington Post, Huffington Post, MIT Technology Review )
  • Cursing on Social Media: Examine the characteristics of cursing activity on Twitter, involving the analysis of about 51 million tweets and about 14 million users. Explore the ubiquity, utility, and contextual dependencies of cursing behaviors. ( CSCW 2014, media coverage: Time, Fast Company, Gizmodo )
  • Election Prediction: Study the spectrum of Twitter users who participate in the online discussion of elections, and examine the predictive power of different user groups. Characterize users across both user participation dimensions such as engagement degree, tweet mode, and content type, and demographic dimensions such as political preference. ( SocInfo 2012 )
  • Context-Aware Harassment Detection: Develop context-aware techniques to glean information about the people involved and their relationships, and to determine and evaluate potential harassment and harassers. ( Wiki Page )

Twitris: 360-degree Social Media Analysis (demo, media coverage: Mashable, Semanticweb): A Semantic Social Web application with real-time monitoring and multi-faceted analysis of social signals to provide insights and a framework for situational awareness, in-depth event analysis and coordination, emergency response aid, reputation management etc. My work includes:

  • Popular Topics: Design and implement the algorithm for extracting popular topics from tweets.
  • Target-specific sentiment analysis: Identify the sentiment-rich topics/entities with respect to a specific event, and extract the sentiments associated with these topics/entities.

Publications (Google Scholar)


  • "Outstanding Student Award" for the 2014-2015 academic year, Department of Computer Science and Engineering, Wright State University
  • "Outstanding Reviewer Award" for WSDM 2016. There were 21 reviewers (among all the 376 reviewers) nominated by the Senior PC members for this award.

Professional Service

  • Workshop co-organizer: Religion on Social Media @ ICWSM 2015
  • Program Commitee: WWW 2016, ICWSM 2016, WSDM 2016, WebSci 2015, ICWSM 2015, WebSci 2014, SocInfo 2014
  • External Reviewer: CSCW 2016, ESWC 2016, ICWSM 2014, WWW 2013, AAAI 2013, ICWSM 2013, LSM at NAACL 2013, ICWSM 2012, LSM at NAACL 2012, ICWSM 2011, LSM at ACL 2011.