I am working in Prof. Zhang's team on the tecWAVE project. tecWAVE is a tool to help children develop their vocabularies. by providing context-specific definitions to words when the word is clicked. The context will be based on the word's place in the text and the reading level of the user. Here's a little more information about the project,
Features of tecWAVE include:
- The WaveMachine will automatically annotate words that a student is unfamiliar with based upon grade/reading level using artificial intelligence, especially natural language processing techniques that:
1. Automatically disambiguate the annotation based on the context
2. Automatically generate pictures, pronunciation keys, and dictionary entries for these words
3. Automatically provide an annotation in a child's native language when appropriate
- The WaveMachine personalizes the annotations individually, figuring out which words should and should not be annotated, based on the child's background along with explicit and implicit feedback.
- The application automatically tracks the history of students' learning by their use of clicks through activities which assists in refining the system so that it works more effectively in the classroom.
- The application also supports social annotation through Web 2.0 applications by seeking feedback from teachers and parents to furthur improve the quality of annotations. It also generates detailed reports for teachers.
I will be working with graduate student Jiazhong Nie, who is responsible for the text definition module of the project. This module provides definitions in English, Chinese and Spanish of English words. For words with multiple senses, definitions are chosen according to the context in which they occur. This word sense disambiguation process is performed using a supervised machine learning approach. We are trying to enrich this process by trying out different techniques and then testing the data.