Semantic Hackers need NLP (Natural Language Processing), PROLOG and Assembly Language…
Semantic Hacker is a cool blog (and site) about the Semantic Web. I’ve placed them in my Web 3.0 blogroll today. They are offering an API (Application Programming Interface in geek-lingo) that can be integrated in Semantic Web applications using “Concept-Spaces”, based on weighted keywords.
However, an attempt I made to run their on-line test proved immediately and conclusively that their innovative API badly needs (Prolog-based?) Natural Language methods (NLP) of parsing and understanding human text. Without such methods, I’m afraid that the Semantic Hacker’s API… sucks! (Although this failure is occasional; most of the time the API test being quite successful). You can check this out by looking at the text I submitted to their on-line test, together with their (rather off-topic) results (copied and pasted from their test-form page):
Input Text Cut and paste your own text into this area.
Suppose there is an Internet Community with just one moderator; and that every person in the community keeps himself moderated: some by moderating themselves, some by obeying the (one and only) moderator. It seems reasonable to imagine that the moderator obeys the following rule: He moderates all and only those members of the community who do not moderate themselves. Under this scenario, we can ask the following question: Does the moderator moderate himself?
Simplified Semantic Signature®
…/EverQuest_Games/EverQuest/Server_Specific 38…/Space_Combat/Star_Wars_Games/Star_Wars_Galaxies/Clans_and_Guilds 26Society/Philosophy/Chats_and_Forums 24Home/Family/Pregnancy/Chats_and_Forums24Arts/Literature/Authors/Blake,_William 23
Related Wikipedia Articles
1.) Moderate2.) Quarterdeck









