Automated set up of web-based natural language interface
View Patent ↗A procedure automates the process of setting up an instance of a conversational natural language interface for a Web site. By automating the process of setting up a new Web site, the process enables a new interface to be created by anyone. Subsequent manual tuning of the interface is possible and much easier to do than creating an interface from scratch. In order to set up an instance of a natural language conversational interface, it is necessary to define a hierarchy of topics into which individual documents or Web pages can be classified, provide a keyword index for those documents for an associated search engine, and for each node in the hierarchy, specify a mechanism for associating an input natural language (NL) query to the node.
1. An automated method for setting up a natural language interface in a Web site, comprising the steps of:
defining a hierarchy of topics into which individual documents or Web pages can be classified;
generating a keyword index for those documents; and
for each topic in the hierarchy, associating a set of n-grams to a topic in the topic hierarchy, which set of n-grams is distinctive to that topic and wherein the n-grams maybe sparse or non-sparse n-grams
wherein the step of generating a keyword index comprises the step of extracting sparse n-grams of keywords for each group of pages in the topic hierarchy.
2. An automated method for setting up a natural language interface in a Web site, comprising the steps of:
automatically inducing a topic hierarchy by examining a structure of the Web site;
creating n-grams from pages in the Web site that are associated with a topic in the topic hierarchy wherein the n-grams may be sparse in-grams or non-sparse n-grams; and
creating rules from the n-grams, wherein each topic has associated rules that are used to decide if a new input document or query references the topic,
wherein the step of creating rules is performed automatically and further comprising the optional step of manually editing the rules.
3. An automated method for setting up a natural language interface in a Web site, comprising the steps of:
defining a hierarchy of topics into which individual documents or Web pages can be classified;
generating a keyword index for those documents;
for each topic in the hierarchy, associating a set of n-grams to a topic in the topic hierarchy, which set of n-grams is distinctive to that topic and wherein the n-grams maybe sparse or non-sparse n-grams; and
converting the set of n-grams to classification rules.