| This dissertation shows that mathematical techniques can be used by intelligent agents to determine the contents of text relevant to some subject domain, to derive partial semantics of that content, and to perform a concise summarization of it. The overall technique to implement these capabilities, developed by the author and named the MAYA Semantic Technique, is organized into four stages. The first determines the parts of speech of each word in each sentence of the page in order, to locate its subject. The second stage performs frequency analysis on each part of speech. From this, the third stage determines the key part of speech with the highest frequency. The fourth stage determines the verb and object of the sentence. The advantage of the MAYA Semantic Technique lies in its dependence on mathematical principles rather than rules of grammar, resulting in fast processing and accuracy when performed on unambiguous technical text. The MAYA Semantic Technique is to be part of a proposed architecture for an entire web-based intelligent tutoring system named CHARLIE by the author, portions of which have been implemented and will use the MAYA Semantic Technique. On a sample set of sentences, partial semantics derived using the MAYA Semantic Technique were approximately 80% accurate. |
