Ingredient Association via Position Probability in Recipe Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing recipe search systems require users to manually register attribute values for ingredients, which is time-consuming and effort-intensive, and do not efficiently extract related ingredient names.
Innovation Solution
An information processing apparatus that calculates the probability of ingredient names being main ingredients based on their ordinal position in recipe lists and associates related ingredient names, reducing the need for manual attribute registration and improving search efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a dictionary-based method is used to extract related ingredient names, then the accuracy of ingredient association is improved, but the time and effort required to prepare the dictionary and register attribute values increases
Solution Approach 1:
The system automatically extracts related ingredient names by analyzing the ordinal positions of ingredients in recipe lists without requiring manual dictionary preparation. The calculation means computes probability values based on ingredient positions, and the association means automatically associates ingredients with similar probability patterns, enabling the system to self-generate the association data that would otherwise require manual effort
Solution Approach 2:
The manual mechanical process of dictionary preparation and attribute value registration is replaced by an automated computational system. The calculation means uses algorithmic processing to analyze ingredient positions and compute probability values, substituting the manual mechanical work with automated information processing
2Manufacturing precision
If manual attribute value registration is required for each ingredient, then the precision of ingredient classification is improved, but the complexity of the system increases
Solution Approach 1:
The system extracts the essential feature for ingredient classification directly from the recipe structure itself - the ordinal position of ingredients in the ingredient list. By taking out and utilizing this existing structural information, the system avoids the need for complex manual attribute registration while maintaining classification precision
Solution Approach 2:
The ordinal position of ingredients in the ingredient list serves multiple functions: it indicates ingredient importance, determines probability calculation, and enables automatic association. This multi-functional use of a single piece of information simplifies the system by eliminating the need for separate attribute registration mechanisms
Data Source
AI summary
Related ingredient names are extracted from ingredient names included in recipes while reducing the time and effort of a person. An information processing apparatus acquires one or more ingredient names included in each of one or more recipes included in one division of a plurality of divisions of recipes. Next, on the basis of an ordinal number of presentation of each of the acquired ingredient names in an ingredient list in the recipe, the information processing apparatus calculates a value indicating a probability that the ingredient name is a name of a main ingredient of the recipe belonging to the one division for each of the ingredient names. Then, the information processing apparatus associates a plurality of the ingredient names, among the acquired ingredient names, whose calculated values satisfy a predetermined condition with each other as related terms and stores the plurality of the ingredient names in a storage means.