About Me

My name is Wenbo Wang, a final year Computer Science Ph.D student at Kno.e.sis Center, Wright State University. My broad research interests include: Text Mining, Natural Language Processing and Social Computing. My current focus is on Emotion Identification, Social Media Analytics and Sentiment Analysis. I am working with Professors Amit Sheth and T.K. Prasad.

In my spare time, I like thinking about how to do things differently with the help of technology. I enjoy brainstorming crazy ideas with friends and I am constantly thrilled by these amazing ideas!


Work

  • 2008 Sep. - present, Research Assistant, Kno.e.sis Center, Wright State University
  • 2013 Summer, Microblog Search, Intern, Bing Social (Microsoft)
  • 2011 Spring - 2012 Spring, Teaching Assistant, Wright State University
  • 2010 Summer - 2011 Spring, Data Integration, Intern, Yahoo!
  • 2005 Sep. - 2008 Apr., Research Assistant, Beijing University of Posts and Telecommunications

Education


Projects

Emotion Identification: identifying people's emotions (e.g., joy, sadness, anger, love, etc.) has implications in many fileds: personal retrospection, suicide prevention, workplace productivity, etc. I focused on identifying people's emotions from their texts at the sentence level.

  • Keywords: Support Vector Machine, Multinomial Naive Bayes, Logistic regression, χ2 Feature Selection
  • Which features are effective for identifying emotions? (SocialCom2012, BII2012)
  • How to improve the performance by combining the rule-based approach with the Machine Learning-based approach? (BII2012)
  • How to harness Twitter 'big data' to tackle this problem? (SocialCom2012)
  • How to identify emotions from text in different domains, e.g., blogs, news, tweets, etc. (in progress)

Social Media Analytics: large amount of social media data provides a great opportunity to analyze people's activities and opinions in real time. Specifically:

  • Keywords: NLTK, Twitter Streaming data, Cohen’s Kappa, LIWC, Gender identification, Sentiment analysis
  • We examine the characteristics of cursing activities on Twitter, involving the analysis of about 51 million tweets and about 14 million users. We explore a set of questions that have been recognized as crucial for understanding cursing in offline communications by prior studies, including the ubiquity, utility, and contextual dependencies of cursing. (CSCW2014) (Media coverage: Fast Company, TIME)
  • Study the spectrum of Twitter users who participate in the on-line discussion of elections, and examine the predictive power of different user groups. We characterize users across both user participation dimensions such as engagement degree, tweet mode, and content type, and demographic dimensions such as political preference, etc. (SocInfo2012)

Topic Specific Sentiment Analysis: we automatically extract sentiment expressions for a given target (e.g., movies, person) from a corpus of unlabeled tweets. Specifically (ICWSM2012):

  • Keywords: SentiWordnet, MPQA, Urban Dictionary, Stanford parser, Nonlinear Optimization (L-BFGS-B)
  • Recognize a diverse and richer set of sentiment-bearing expressions in tweets, including formal and slang words/phrases.
  • Assess the target-dependent polarity of each sentiment expression. The polarity of sentiment expression is determined by the nature of its target.
  • Provide a novel formulation of assigning polarity to a sentiment expression as a constrained optimization problem over the tweet corpus.
  • Example words and phrases: "want my money back", "must see", "luv".

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. My work:

  • Keywords: Storm, Logistic Regression, Naive Bayes, MySQL, PHP, Servlet.
  • Integrate my emotion identification work into the system.
  • System performance optimization.


Publications (Google Scholar)


Skills

Programming (Java, Python, R, Perl, Pig, SQL, SPARQL), Database Management System (MySQL, PostgreSQL, Virtuoso), Machine Learning (Weka, scikit-learn, LIBSVM, LIBLINEAR), Big Data (MapReduce, Hadoop, Storm), Information Retrieval (Lucene), Nature Language Processing (NLTK, Stanford CoreNLP), Code Management (SVN, Git, Ant) and open to learning new skills.


Professional Service

  • Program Commitee: HT 2014 LBR+DC, HT 2014, WI 2014, WI 2013
  • External Reviewer: ICWSM 2014, ICWSM 2013, ISMIS 2014, NLDB2014, IC3 2014, RAMSS (Workshop) 2013, WWW 2013, NLDB 2013, WWW 2012, ICWSM 2012, LSM 2012 (Workshop), ER 2012, ESWC 2011, AAAI 2010, ICWSM 2010

Patents

  • Lu Chen, Wenbo Wang, Amit Sheth. "Topic-specific Sentiment Extraction", US Patent pending by Wright State University
  • Wenbo Wang, Lei Duan. "Temporal User Engagement Features", US Patent pending by Microsoft

Resume

Download


Contact

  • Email: [firstname][lastname]@gmail.com OR [firstname]@knoesis.org
  • LinkedIn: link