Future, Gaming - Sunday, August 17, 2008 12:16 - 0 Comments

Mashup: AIML + Inform 7 + tag clouds + social network services

Let’s talk about chatbots and light artificial intelligence.  First, some definitions:

1. AIML:  Artificial Intelligence Markup Language, is an XML dialect for creating natural language software agents. The XML dialect called AIML was developed by Richard Wallace and a worldwide free software community between the years of 1995 and 2002. {ed: Those familiar with ALICE bots will know that AIML is used for making the bots work.}

2. Inform 7 - a programming language and design system for interactive fiction originally created in 1993 by Graham Nelson. … In 2006, Nelson released Inform 7 (briefly known as Natural Inform), a completely new language based on principles of natural language and a new set of tools based around a book-publishing metaphor.

3. Tag Clouds - A tag cloud or word cloud (or weighted list in visual design) is a visual depiction of user-generated tags, or simply the word content of a site, used typically to describe the content of web sites. Tags are usually single words and are typically listed alphabetically, and the importance of a tag is shown with font size or color.  Thus both finding a tag by alphabet and by popularity is possible. The tags are usually hyperlinks that lead to a collection of items that are associated with a tag.

4. Social Network Services - A social network service focuses on building online communities of people who share interests and activities, or who are interested in exploring the interests and activities of others. Most social network services are web based and provide a variety of ways for users to interact, such as e-mail and instant messaging services.

–

I’ve spent a lot of time recently pondering the creation of my own AI characters for use in my game/fiction/universe project. The idea of mashing up the four topics above are purely theoretical.

Here is a possible scenario: Provide a substantial series of questions and answers to a chat bot, followed by  a cloud of keywords (tag cloud) on a given topic to a bot. The bot would have a learning algorithm to ask ‘What is _____ ?’. The answer would be provided in the natural language syntax of using the Inform 7 input editor. “The Blogosphere is a thing.” This would create AIML for the bot to add to its memory .

Social network services are only relevant here because of the consistent patterns and relationships on public display. “Joe is a person. Jane is a person. Joe is connected to Jane (and Mary and Bob and Sue). Jane is married to Bob” and so on. This could apply to the parsing of XFN (XHTML Friends Network, a microformat that provides a granular level or relationships “Bob is a colleague of Jane”.

Social network services did not have the critical mass during the development of AIML as they do today. This would result in a bot having the memory and broad context based on the volumes of data fed to it.

While this is conceptually possible, it may not be the most elegant solution. Just considering the mashup of AIML + Inform + tag clouds + and social network relationships, may put our brains and thought process in a different mind space- resulting in smart assistants or richer game characters– for those of us outside the realm of chatbots or interactive fiction.

Additional reading: Gamasutra: Beyond AIML: Chatbots 102

Friendfeed discussion here