Modifying existing recipes to incorporate additional or replace existing ingredients

• US 9,870,550 B2
• Filed: 11/12/2015
• Issued: 01/16/2018
• Est. Priority Date: 11/12/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Byron, Donna K., DiMascio, Carmine M., Pinel, Florian, Winkler, Timothy P.
• Primary Examiner(s): GUERRA-ERAZO, EDGAR X
• Application Number:
  US14/938,907
• Publication Number:
  US 20170139902A1
• Time in Patent Office: 796 Days
• Field of Search: 704 1- 10
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders
• CPC Class Codes: G06Q 10/10 Office automation; Time management

Abstract

Mechanisms are provided for implementing a recipe modification system. The recipe modification system receives a request to modify an existing recipe from a requestor. The request identifies the existing recipe and an ingredient to be added to the existing recipe. The recipe modification system identifies a cluster of recipe elements associated with the ingredient to be added to the existing recipe and selects a representative member recipe element of the cluster. The recipe modification system modifies the existing recipe based on the selected representative member recipe element and generates a natural language text for the modified recipe based on the existing recipe and the selected representative member recipe element. The recipe modification system outputs the natural language text for the modified recipe to the requestor.

Claims

1. A method, in a data processing system comprising a processor and a memory comprising instructions which are executed by the processor to cause the processor to implement a recipe modification system, the method comprising:

receiving, by the recipe modification system, a request, from a requestor computing device, to modify an existing recipe in an electronic corpus of existing recipe data structures, wherein the request is an electronic input to the recipe modification system which identifies an existing recipe corresponding to an existing recipe data structure in the electronic corpus of recipes, and an ingredient to be added to existing ingredients already identified in content of the existing recipe data structure;

identifying, by the recipe modification system, a cluster of recipe elements, from a plurality of clusters of recipe elements, associated with the ingredient to be added to the existing recipe, wherein the plurality of clusters are identified based on a machine learning process executed on the electronic corpus of existing recipes;

selecting, by the recipe modification system, a representative member recipe element of the identified cluster;

modifying, by a cognitive computing engine of the recipe modification system, the existing recipe based on the selected representative member recipe element, wherein modifying the existing recipe comprises performing cognitive computing analysis of the existing recipe and the selected representative member recipe element to determine a timing or location in a sequence of preparation instructions already present in the existing recipe where new preparation instructions, corresponding to the selected representative member recipe element, are added to the sequence of preparation instructions to generate a modified recipe;

generating, by the cognitive computing engine of the recipe modification system, a natural language text for the modified recipe based on the modification of the existing recipe; and

outputting, by the recipe modification system, the natural language text for the modified recipe to the requestor computing device.

2. The method of claim 1, wherein the recipe elements comprise reusable branches of acyclic graph data structures corresponding to existing recipes in the electronic corpus of existing recipes.

3. The method of claim 2, wherein the recipe elements comprise reusable branches of acyclic graph data structures corresponding to existing recipes in the electronic corpus of existing recipes which have the same or a similar ingredient to the ingredient to be added to the existing recipe.

4. The method of claim 2, wherein the re sable branches comprise a sub-portion of a recipe, and wherein the reusable branches comprise recipe instructions for performing preparation of a portion of a recipe which are reusable in a plurality of recipes.

5. The method of claim 1, further comprising:

analyzing the electronic corpus of existing recipes to extract reusable branches of acyclic graph data structures corresponding to the existing recipes; and

clustering the extracted reusable branches into a plurality of clusters, wherein reusable branches in a same cluster have similar characteristics, and wherein identifying a cluster of recipe elements associated with the ingredient to be added to the existing recipe comprises selecting a cluster from the plurality of clusters that comprises at least one reusable branch having the ingredient to be added to the existing recipe.

6. The method of claim 5, wherein analyzing the electronic corpus of existing recipes comprises, for each existing recipe in the corpus:

performing natural language processing on the existing recipe to generate an acyclic graph for the existing recipe, wherein nodes represent at least one of ingredients of the existing recipe or actions to be performed with regard to ingredients in the existing recipe; and

identifying reusable branches of nodes and connections between nodes in the acyclic graph based on a predetermined relationship with a root node of the acyclic graph.

7. The method of claim 1, wherein identifying the cluster of recipe elements associated with the ingredient to be added to the existing recipe comprises:

identifying, in the plurality of clusters, more than one cluster of recipe elements associated with the ingredient to be added to the existing recipe;

filtering the more than one cluster of recipe elements based on recipe elements in the existing recipe to which the ingredient is to be added; and

selecting the cluster of recipe elements from remaining clusters of recipe elements after filtering.

8. The method of claim 7, wherein filtering the more than one cluster of recipe elements comprises filtering the more than one clusters to remove clusters of recipe elements that match or are incompatible with, the recipe elements in the existing recipe to which the ingredient is to be added.

9. The method of claim 7, wherein filtering the more than one cluster of recipe elements comprises filtering the more than one clusters to remove clusters of recipe elements that are not compatible with a dish type of the existing recipe to which the ingredient is to be added as determined based on one or more predetermined association rules.

20. The method of claim 2, wherein the reusable branches are portions of the acyclic graph that involve a number of ingredients that is less than a total number of ingredients in the corresponding recipe, and whose nodes, other than a result node in the portion of the acyclic graph, are not referenced by other nodes in the acyclic graph.

10. A computer program product comprising a non-transitory computer readable medium having a computer readable program stored therein, wherein the computer readable program, when executed on a computing device, causes the computing device to implement a recipe modification system that operates to:

receive a request, from a requestor computing device, to modify an existing recipe in an electronic corpus of existing recipes, wherein the request is an electronic input to the recipe modification system which identifies an existing recipe corresponding to an existing recipe data structure in the electronic corpus of recipes, and an ingredient to be added to existing ingredients already identified in content of the existing recipe data structure;

identify a cluster of recipe elements, from a plurality of clusters of recipe elements associated with the ingredient to be added to the existing recipe, wherein the plurality of clusters are identified based on a machine learning process executed on the electronic corpus of existing recipes;

select a representative member recipe element of the identified cluster;

modify the existing recipe based on the selected representative member recipe element, wherein modifying the existing recipe comprises performing cognitive computing analysis of the existing recipe and the selected representative member recipe element to determine a timing or location in a sequence of preparation instructions already present in the existing recipe where new preparation instructions, corresponding to the selected representative member recipe element, are added to the sequence of preparation instructions to generate a modified recipe;

generate a natural language text for the modified recipe based on the modification of the existing recipe; and

output the natural language text for the modified recipe to the requestor computing device.

11. The computer program product of claim 10, wherein the recipe elements comprise reusable branches of acyclic graph data structures corresponding to existing recipes in the electronic corpus of existing recipes.

12. The computer program product of claim 11, wherein the recipe elements comprise reusable branches of acyclic graph data structures corresponding to existing recipes in the electronic corpus of existing recipes which have the same or a similar ingredient to the ingredient to be added to the existing recipe.

13. The computer program product of claim 11, wherein the reusable branches comprise a sub-portion of a recipe, and wherein the reusable branches comprise recipe instructions for performing preparation of a portion of a recipe which are reusable in a plurality of recipes.

14. The computer program product of claim 10, wherein the recipe modification system further operates to:

analyze the electronic corpus of existing recipes to extract reusable branches of acyclic graph data structures corresponding to the existing recipes; and

cluster the extracted reusable branches into a plurality of clusters, wherein reusable branches in a same cluster have similar characteristics, and wherein identifying a cluster of recipe elements associated with the ingredient to be added to the existing recipe comprises selecting a cluster from the plurality of clusters that comprises at least one reusable branch having the ingredient to be added to the existing recipe.

15. The computer program product of claim 14, wherein analyzing the electronic corpus of existing recipes comprises, for each existing recipe in the corpus:

performing natural language processing on the existing recipe to generate an acyclic graph for the existing recipe, wherein nodes represent at least one of ingredients of the existing recipe or actions to be performed with regard to ingredients in the existing recipe; and

identifying reusable branches of nodes and connections between nodes in the acyclic graph based on a predetermined relationship with a root node of the acyclic graph.

16. The computer program product of claim 10, wherein identifying the cluster of recipe elements associated with the ingredient to be added to the existing recipe comprises:

Identifying, in the plurality of clusters, more than one cluster of recipe elements associated with the ingredient to be added to the existing recipe;

filtering the more than one cluster of recipe elements based on recipe elements in the existing recipe to which the ingredient is to be added; and

selecting the cluster of recipe elements from remaining clusters of recipe elements after filtering.

17. The computer program product of claim 16, wherein filtering the more than one cluster of recipe elements comprises filtering the more than one clusters to remove clusters of recipe elements that match or are incompatible with, the recipe elements in the existing recipe to which the ingredient is to be added.

18. The computer program product of claim 16, wherein filtering the more than one cluster of recipe elements comprises filtering the more than one clusters to remove clusters of recipe elements that are not compatible with a dish type of the existing recipe to which the ingredient is to be added as determined based on one or more predetermined association rules.

19. An apparatus comprising:

a processor; and

a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, cause the processor to implement a recipe modification system that operates to;

receive a request, from a requestor computing device, to modify an existing recipe in an electronic corpus of existing recipes, wherein the request is an electronic input to the recipe modification system which identifies an existing recipe corresponding to an existing recipe data structure in the electronic corpus of recipes, and an ingredient to be added to existing ingredients already identified in content of the existing recipe data structure;

identify a cluster of recipe elements, from a plurality of clusters of recipe elements, associated with the ingredient to be added to the existing recipe, wherein the plurality of clusters are identified based on a machine learning process executed on the electronic corpus of existing recipes;

select a representative member recipe element of the identified cluster;

modify the existing recipe based on the selected representative member recipe element, wherein modifying the existing recipe comprises performing cognitive computing analysis of the existing recipe and the selected representative member recipe element to determine a timing or location in a sequence of preparation instructions already present in the existing recipe where new preparation instructions, corresponding to the selected representative member recipe element, are added to the sequence of preparation instructions to generate a modified recipe;

generate a natural language text for the modified recipe based on the modification of the existing recipe; and

output the natural language text for the modified recipe to the requestor computing device.