Description Logic-Based Approximate Joke Comparison

Computational recognition of humor in texts is an aspect of natural language processing. How to algorithmically recognize humorous texts is not clear. Humor recognition in semantically untagged text may be easier with a humor ontology. Our speculation is that once such an ontology is created, an unmarked text may be analyzed using concepts in this ontology to determine whether the text is humorous. Such an ontology can be created from a set of central jokes, taken from the Internet or from books. Each joke of the central set can be described using description logic. Jokes often appear to be similar to each other. The similarity value of two jokes can be measured using one of the humor theories. New jokes can be recognized by comparison to the central set, if it contains jokes that are similar to the new ones