User state sensitive system and method for nutrient analysis using natural language interface

• US 6,556,963 B1
• Filed: 09/24/1997
• Issued: 04/29/2003
• Est. Priority Date: 09/24/1997
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Tetzlaff, Linda S.
• Primary Examiner(s): Edouard, Patrick N.
• Application Number:
  US08/936,796
• Time in Patent Office: 2,043 Days
• Field of Search: 704/1, 704/7, 704/9, 704/10, 704/275, 128/291, 128/630, 128/921, 434/238, 434/127, 600/300, 235/375
• US Class Current: 704/9
• CPC Class Codes: G06F 40/237 Lexical tools
• CPC Class Codes: G16H 20/60 relating to nutrition control, e.g. diets
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

An easy to use system and method personalizes dynamic nutritional requirements of a user. A natural language user interface is used to input information to a nutrient analysis system that integrates nutritional guidelines from diverse sources. A user of the system describes elements of a meal that she has either eaten or is considering eating, and the system responds with dynamic, personalized, state-sensitive feedback about the nutrient components of her choices in relation to her personal nutritional objectives.

Claims

1. A nutrient analysis system comprising:

a natural language food processor for receiving a natural language input from a user, wherein the natural language input describes food and is input other than by a displayed menu selection;

an input parser that uses a language model to parse the natural language input into food components;

a speech recognition system coupled to the natural language processor to record meals and other food intake events;

a first database accessed by the natural language processor, said database being a model of the speech used to specify food selection;

a second database storing a food model;

a fuzzy logic search engine that searches the food model in the second database using a canonical description of a food which consists of a quantity, a size modifier, a unit, a food modifier and a food, to generate one or more hits, wherein said food model is updated with input from a user to accommodate new foods, new food names and nutrient requirements;

a feedback generator that produces one or more running totals for one or more nutrients in the hits; and

a third database accessed by the feedback generator, said database describing a state of the user as it relates to the user'"'"'s nutrient intake, the state of the user including static attributes, current state and a state history, wherein said third database stores a set of rules corresponding to a predetermined diet plan for a user, and wherein the feedback generator compares the food described in said natural language input with said set of rules.

2. The nutrient analysis system recited in claim 1 wherein the input parser utilizes a canonical set of descriptive rules conforming to rules of natural language.

3. The nutrient analysis system recited in claim 1 further comprising a database of rules accessed by the search engine, said database storing a natural language description of eaten food.

4. The nutrient analysis system recited in claim 3 wherein said database of rules decodes the USDA nutrient database names.

5. The nutrient analysis system recited in claim 4 wherein a multiplicity of nutrient target values are extracted from a multiplicity of sources in a predetermined order.

6. The nutrient analysis system recited in claim 4 wherein a multiplicity of sources are presented to said user and said user selects a source from among said multiplicity of sources.

7. The nutrient analysis system recited in claim 1, wherein said third database stores information indicative of at least one of a state of the user as it relates to the user'"'"'s nutrient intake, the state of the user including static attributes, current state and a state history.

8. The nutrient analysis system recited in claim 1 wherein said natural language input specifies a unit of size of said food.

9. The nutrient analysis system recited in claim 8, wherein said unit of size is a non-numeric unit of size.

10. The nutrient analysis system recited in claim 1, wherein said feedback generator provides an indication of whether the food satisfies said diet plan.