# US 2006/0154218 A1 · Interactive learning system based on template-template structure

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2006/0154218 A1
- Document ID: US20060154218A1
- Status: Published application
- Kind: A1
- Application number: 10/550,090
- Filed: 2004-03-22
- Published: 2006-07-13
- Earliest claimed: 2003-03-22
- Assignees: Sunflare Co., Ltd.
- Inventors: Naoyuki Tokuda; Liang Chen
- Web page: https://patentgrep.com/patent/US20060154218A1

## CPC

- G09B 19/06: Foreign languages (with audible presentation of material to be studied G09B5/04)
- G09B: EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS (devices for psychotechnics or for testing reaction times A61B5/16; games, sports, amusements A63; projectors, projector screens G03B)

## Abstract

A learning system uses the concept of template automaton. “Various expected examples of learners” consisting of “correct” answers and “incorrect” answers are collected and a representative NLP technique such as HCS (heaviest common character string) or LCS (longest common character string) algorithm is used as an effective error diagnosis engine in the language learning system. These examples embedded in the template are used for diagnostic analysis of the answers of the learners. This diagnosis is performed by selecting a path of the highest similarity with the input sentence of the learner among a plenty of candidate paths. Thus, it is possible to automatize and simplify the time-requiring authoring task used in the language-oriented intelligent learning system.

## Claims

### Claim 1

A dialogue learning system having a template-template structure based on extraction rules and exploiting the expanding power of buggy rules, wherein some of nodes in the template-template structure are marked with extracting rule-associated symbols enabling the extraction of many templates or a so-called large template.

### Claim 2 (depends on 1)

A dialogue learning system according to claim 1, which uses a template-template scheme wherein extraction rule-associated special symbols are assigned to some of nodes in a expanded template thereby allowing one or more of the templates to be extracted.

### Claim 3 (depends on 1)

A dialogue learning system according to claim 1, which is based on a extracting rule used for marking nodes in a single template-template for extracting a plurality of templates from a single template-template.

### Claim 4 (depends on 1)

A dialogue learning system according to claim 1, which is based on a buggy rule used for expanding a slim template-template into a larger template-template or thicker template-template.

### Claim 5 (depends on 1)

A dialogue learning system according to claim 1, which is based on a extracting rule wherein the extraction rule is always associated with a set of symbols, {s1, s1, - - - , sn}, and each of the symbols is assigned to one or more nodes in the template, and one or more values is assigned to the associated symbols.

### Claim 6 (depends on 1)

A dialogue learning system according to claim 1, which uses an HCS matching algorithm to match an inputted sentence with a template-template embedded with an extracting rule.
