Automatic diet tracking system and method

• US 9,449,530 B1
• Filed: 02/11/2016
• Issued: 09/20/2016
• Est. Priority Date: 02/13/2015
• Status: Active Grant

• Current Assignee: Genesant Technologies, Inc.
• Original Assignee: Genesant Technologies, Inc.
• Inventors: Mrowka, James J., Robey, Paola N., Christ, Athanasios G., Robey, Daniel S.
• Primary Examiner(s): Spooner, Lamont
• Application Number:
  US15/041,780
• Time in Patent Office: 222 Days
• Field of Search: 704/1, 704/9, 704/10
• US Class Current: 1/1
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G09B 19/0092 Nutrition
• CPC Class Codes: G16H 20/60 relating to nutrition control, e.g. diets

Abstract

An automatic diet tracking system and methods comprising: i) voice-transcribed or typed text natural language processing and automatic tracking to record food, food quantity, and nutrition data, ii) multi-food administration to record multiple foods and related data in a single user voice-transcribed or typed text submission, and iii) location-based diet recommendations system that provides customized food recommendations to users based on user preferences and user physical location. Further, such automatic diet tracking system and location-based diet recommendations system are usable through computers, tablets, mobile phones, smart watches, wearables and other similar devices.

Claims

1. A computer-implemented method of automatic diet tracking, comprising:

receiving text from a user that describes a food that is to be tracked;

parsing the received text into text segments;

identifying automatically in each parsed text segment a food quantity value and a food quantity unit for said food that is to be tracked, said identifying comprising searching said parsed text segment for a quantity value followed directly by a quantity unit, and assigning said quantity value and said quantity unit to be said food quantity value and said food quantity unit for the food to be tracked, and upon not finding a quantity value followed directly by a quantity unit, selecting as said food quantity value and said food quantity unit a most frequently occurring quantity value and quantity unit for said food to be tracked;

cleaning the parsed text segments to identify and remove words, connected spaces, and punctuation that are not used to identify food to produce parsed cleaned text;

processing the parsed cleaned text segments using a text match algorithm to find said food that is to be tracked in each parsed cleaned text segment comprising ranking each food text match found using a ranking process, and selecting the food with a predetermined rank to be the food that is to be tracked; and

reporting diet tracking information for said food to be tracked.

2. The method of claim 1, wherein said parsing further comprises removing all quantity values and quantity units from said parsed text segments, and wherein said ranking said found food text matches comprises searching a past history of said user for previous matches, and selecting from said past history a most recent food tracked as said food to be tracked.

3. The method of claim 2, wherein upon a food match not being found in said searching, then searching a user population history for previous matches, and selecting a food that has been most often tracked as said food to be tracked.

4. The method of claim 3, wherein upon a food match not being found in said user population history, processing said parsed text using a search string modifying algorithm to create a modified text string;

performing a fuzzy search on a food database for food results relevant to said modified text string;

scoring said food results to create a food text match score; and

selecting as said food to be tracked a food having the top score.

5. The method of claim 4, wherein said scoring comprises determining from a user data history lookup for said user whether a food of said food results is designated as a favorite food of said user, and, if so, scoring that food to have the highest score, otherwise, upon no favorite food designation being found, determining from an entire user population history a total count for a number of times each food has been tracked, and scoring the food with the highest count as the food with the highest score.

6. The method of claim 4, wherein said search string modifying algorithm creates said modified search string by performing one or more of:

(i) removing, or replacing with spaces, connecting dashes between words, if any, (ii) removing an apostrophe followed by a letter “

s” and

removing an apostrophe only, if any, and (iii) inserting a space between connected words and numbers that have no space, if any.

7. The method of claim 1, wherein said searching for a food quantity value and a food quantity unit comprises performing a fuzzy logic search of said parsed text for a written number or a numeric value, and upon finding said written number or numeric value, performing a further fuzzy logic search of said parsed text for a match of a quantity unit term directly following said written number or numeric quantity and, if found, assigning said written number or numeric quantity to said food quantity value and assigning said quantity unit term to said food quantity unit.

8. The method of claim 7, wherein upon not finding said written number or numeric quantity, performing a further fuzzy logic search of said parsed text for a match of a quantity unit term, and, if found, assigning said quantity unit term to said food quantity unit, and, if not found, searching a user data history for said user for a most recently tracked food quantity unit for said food to be tracked, and setting the food quantity unit for said food to be tracked to be the most recently tracked food quantity unit.

9. The method of claim 8, wherein upon not finding a most recently tracked food quantity unit in said user data history, searching an entire user population history for a most frequently tracked food quantity unit for said food to be tracked, and setting said food quantity unit for the food to be tracked to be said most frequently tracked food quantity unit.

10. The method of claim 1 further comprising providing diet recommendations based upon food availability within a predetermined distance of a location of said user and target nutrient values of said user.

11. Non-transitory computer readable media embodying executable instructions for controlling a computer for an automatic diet tracking method, comprising:

receiving text from a user that describes a food that is to be tracked;

parsing the received text into text segments;

identifying automatically in each parsed text segment a food quantity value and a food quantity unit for said food that is to be tracked, said identifying comprising searching said parsed text segment for a quantity value followed directly by a quantity unit, and assigning said quantity value and said quantity unit to be said food quantity value and said food quantity unit for the food to be tracked, and upon not finding a quantity value followed directly by a quantity unit, selecting as said food quantity value and said food quantity unit a most frequently occurring quantity value and quantity unit for said food to be tracked;

cleaning the parsed text segments to identify and remove words, connected spaces, and punctuation that are not used to identify food and produce parsed cleaned text;

processing the parsed cleaned text segments using a text match algorithm to find said food that is to be tracked in each parsed cleaned text segment comprising ranking each food text match found using a ranking process, and selecting the food with a predetermined rank to be the food that is to be tracked; and

reporting diet tracking information for said food to be tracked.

12. The non-transitory computer readable media of claim 11, wherein said parsing further comprises removing all quantity values and quantity units from said parsed text segments, and wherein said ranking said found food text matches comprises searching a past history of said user for previous matches, and selecting from said past history a most recent food tracked as said food to be tracked.

13. The non-transitory computer readable media of claim 12, wherein upon a food match not being found in said searching, then searching a user population history for previous matches, and selecting a food that has been most often tracked as said food to be tracked.

14. The non-transitory computer readable media of claim 13, wherein upon a food match not being found in said user population history, processing said parsed cleaned text using a search string modifying algorithm to create a modified text string;

performing a fuzzy search on a food database for food results relevant to said modified text string;

scoring said food results to create a food text match score; and

selecting as said food to be tracked a food having the top score.

15. The non-transitory computer readable media of claim 14, wherein said scoring comprises determining from a user data history lookup for said user whether a food of said food results is designated as a favorite food of said user, and, if so, scoring that food to have the highest score, otherwise, upon no favorite food designation being found, determining from an entire user population history a total count for a number of times each food has been tracked, and scoring the food with the highest count as the food with the highest score.

16. The non-transitory computer readable media of claim 14, wherein said search string modifying algorithm creates said modified search string by performing one or more of:

(i) removing, or replacing with spaces, connecting dashes between words, if any, (ii) removing an apostrophe followed by a letter “

s” and

removing an apostrophe only, if any, and (iii) inserting a space between connected words and numbers that have no space, if any.

17. The non-transitory computer readable media of claim 11, wherein said searching for a food quantity value and a food quantity unit comprises performing a fuzzy logic search of said parsed text for a written number or a numeric value, and upon finding said written number or numeric value, performing a further fuzzy logic search of said parsed text for a match of a quantity unit term directly following said written number or numeric quantity and, if found, assigning said written number or numeric quantity to said food quantity value and assigning said quantity unit term to said food quantity unit.

18. The non-transitory computer readable media of claim 17, wherein upon not finding said written number or numeric quantity, performing a further fuzzy logic search of said parsed text for a match of a quantity unit term, and, if found, assigning said quantity unit term to said food quantity unit, and, if not found, searching a user data history for said user for a most recently tracked food quantity unit for said food to be tracked, and setting the food quantity unit for said food to be tracked to be the most recently tracked food quantity unit.

19. The non-transitory computer readable media of claim 18, wherein upon not finding a most recently tracked food quantity unit in said user data history, searching an entire user population history for a most frequently tracked food quantity unit for said food to be tracked, and setting said food quantity unit for the food to be tracked to be said most frequently tracked food quantity unit.

20. The non-transitory computer readable media of claim 11 further comprising providing diet recommendations based food availability within a predetermined distance of a location of said user and target nutrient values of said user.