Senior Design Projects

Speech and Language Processing by Computer

PROJECT DESCRIPTION

We will be developing an artificial agent capable of sspeech and language processing. By speech and language processing, we mean those computational techniques that process written (and spoken) human language to communicate with a computer. For example recall the following famous piece from the screen play of 2001: A space Odyssey:

Dave Bowman: 'Open the pod bay doors, Hal.'

Hal: 'I'm sorry Dave, I'm afraid I can't do that'

HAL 9000 computer in the film 2001: A space odyssey is an agent capable of advanced language processing behavior including understanding and generating English.

Each team is to develop a basic question-answering (Q/A) system. The Q/A system should be able to retrieve information from an unstructured or semi-structured source, such collection of web pages or a text database. Input and output of this system should be well-formed sentences of natural language. For example, question: 'How many people live in Tucson?', Answer, 'About half a million'. An inference engine should be developed that can generate valid inferences from a set of sentences. For example, if the input contains 'John went to school before his Mom, Mary, came home' and 'Tom, Mary's husband, arrived home shortly after Mary', then the inference engine should be able to generate the inferences 'John went to school before Tom came home', 'Tom is John's father', 'Tom is not Mary's brother', etc.

We intend to have two teams, each developing one such artificial agent. At the end of the year we will hold a Q/A competition to determine which agent performs better.

TEAMSIZE:

2 teams, each with 3-4 students. Computer engineering students, or EE students with digital system option are preferred. .

DISCIPLINES NECESSARY:

Computer Engineering, or EE digital system option.

SKILLS NECESSARY:

Good knowledge and skill in computer software, including:

Knowledge of Data Structures, i.e. CSC 342

Knowledge of Object Oriented programming, i.e. ECE373 or CSC335

Some knowledge of software design is a plus.

KEY CONTACT:

Dr. Michael Marefat, ECE, marefat@ece.arizona.edu,

Mr. Scott Farrar, computational linguistics,farrar@u.arizona.edu