Chatbots are natural web interface to information portals
In this paper we describe a way to access information using chatbot, without the need for sophisticated natural language processing or logical inference. FAQs are Frequently-Asked Questions documents, designed to capture the logical ontology of a given domain. Any Natural Language interface to an FAQ is constrained to reply with the given Answers, so there is no need for NL generation to recreate well-formed answers, or for deep analysis or logical inference to map user input questions onto this logical ontology; simple (but large) set of pattern-template matching rules will suffice. In this paper and as an evidence for this argument, the FAQ in the School of Computing (SoC) at the University of Leeds as well as Perl, Linux, and Python were used to retrain the ALICE chat-bot system, producing FAQchat. The replies from FAQchat looks like results generated by WWW search engines such as AskJeeves or Google. User trials with AskJeeves, Google and FAQchat demonstrate that FAQchat is a viable alternative, and in fact many users prefer it to Google as tool to access FAQ databases. The restricted domain of an FAQ is special case of Question Answering which does not require the sophisticated analysis techniques.