Skip to content

Helping Computers to Understand Human Language

Dr. Yang Liu’s graduate students help computers to understand human language.  Pictured left to right are: Mohammad Hadi Bokaei, Yandi Xia, Yang Liu, Youngchan Kim, Chen Li, and Rui Xia.

The research in Dr. Yang Liu’s Speech and Language Processing Lab reflects the myriad of problems humans encounter when understanding language.  Written communication tends to be more formal than verbal communication.  Verbal communication includes more slang and sentence fragments.  Text communication includes standard and nonstandard abbreviations along with misspelled words and auto-correct spellings that may change the intended word.

If you have ever been frustrated by talking with a computerized agent, the breakdown in communication may be the result of an error in one or more of the many steps from recognizing your speech, to understanding it, and then responding to it. Hopefully, the research being done in Liu’s research lab will help with this process. If you have ever been bothered by the increasing amount of information, from everyday meetings to the news from many different sources, to the large volume of social media data, the research being done in Liu’s research lab will help with better information processing.  Some current research interests of Dr. Liu’s graduate students are summarized in the following table.


Yandi Xia Working on extracting unstructured event information from text and converting it into structured one to make huge information data machine manageable.
Chen Li Working on summarizing news articles. This can help reader focus on particular and important information given a topic. Busy executives can use it to read and summarize multiple newspapers, with a focus on their particular business interest(s).  Chen Li is also working on informal text normalization — converting non-standard English word like ‘tmr’ and ‘thx’ to their correct form (‘tomorrow’, ‘thanks’). The prevalence of informal words in social media text such as twitter and comments in facebook makes it very hard for computers to understand the text. This work is going to benefit many language technologies when dealing with informal text.
Youngchan Kim Working on classifying users in social media into different attributes (such as gender, age, interest). Rapid growth in social media allowed companies to seek more opportunities in advertising, personalization, and recommendation. Identifying user attributes automatically and accurately can improve the user’s experience with social media.
Rui Xia Working on recognition a person’s emotion from the person’s speech.
Mohammad Hadi Bokaei, visiting student from Sharif University in Iran Working on automatically summarizing multiparty meetings.

Yang Liu joined the UT Dallas faculty in 2005.  Since that time, she has supported many graduate students with their research.  Her interest is in human language processing include:

  • Speech recognition and understanding, spoken language processing
  •  Natural language processing (summarization, sentiment analysis, information extraction)
  •  Speech/language disorder, clinical applications
  •  Emotion recognition
  •  Social media analysis
  •  Machine learning in speech/language processing

New graduate students, interested in speech and language processing research should email their CV to Dr. Liu via email (yang.liu@utdallas.edu).

The  Department of Computer Science at UT Dallas  is one of the largest CS departments in the United States with more than 750 undergraduate, 500 master, and 125 PhD students.  They are committed to exceptional teaching and research in a culture that is as daring as it is supportive.

2nd Annual Teacher Inspiration Award Ceremony
"Protecting Your Privacy in the Era of Online and Mobile Social Networks," Dr. Murat Kantarcioglu