The present invention provides a recommender system for recommending a food recipe to a user. The recommender system includes an automatic data collection module. Such a module comprises a recipe crawler engine configured to obtain recipe data available from online resources; a word vector engine configured to transform the recipe data into word vector data; and a feature extraction engine that analyses the word vector data to identify unique characteristics or attributes from the recipe data. The recommender system is further configured to annotate the identified attributes into a recipe database and to execute a matchmaking algorithm that correlates the attributes with a user profile. The recommender system is further configured to generate a user recommendation based on the result from the matchmaking algorithm. The featured automatic data collection module is enabled to classify a great number of recipes in an automated manner. This feature allows prescinding from a manual classification process, advantageously enables scalability, optimizes the recommender system functionality, improves user experience, and saves time and costs during development and operation of an online service platform.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Mohr, Christian, "Recommender System for Recipes (Netflix Personalization)_ID-05777", Technical Disclosure Commons, (March 28, 2022)