Myfitnesspal, Inc.

Under Armour, Inc.

Patrick Howell

Hesamoddin Salehian

Chul Lee

Joohyun Kim

Iman Barjasteh

Surender Reddy Yerva

Wenzhe Xue

Kyler Eastman

Layla Martin

Murtaza Ali

Sid Jatia

Madeline Taylor Rose

A method and system for providing more relevant search results and recommendation from a food database is disclosed. The method includes receiving a query, a first candidate food, and a second candidate food. The method includes generating vectors based on the query and food names of the first and second candidate foods using at least one embedding function of a machine learning model. The method includes determining nutrition content vectors from the nutritional data of the first and second candidate foods. The method includes generating a nutrition content vector based on the query using another embedding function of the machine learning model. The method includes determining which of the first and second candidate food is more relevant to the query based at least in part on the nutrition content vectors. The method includes providing search results or recommendation based on the determined relevance.

Deep multi-modal pairwise ranking model for crowdsourced food data

A method for operating a health tracking system, a health tracking system, and non-transitory computer-readable medium for operating a health tracking system are disclosed. The method comprises receiving a data record comprising at least a descriptive string and nutritional data regarding a consumable item to which the data record corresponds; determining a taste associated to the consumable item based on an evaluation of at least one of: (i) the descriptive string, and (ii) the nutritional data; and associating the determined taste with the data record in a database.

Method, system, and non-transitory computer-readable medium for associating taste with consumable records

A health tracking system and method of operation is disclosed herein. The method of operating the health tracking system comprises: receiving a first data record comprising at least a first descriptive string regarding a consumable item, the first descriptive string having at least one word thereof incorrectly spelled; generating a vector using the first descriptive string using a machine learning model; identifying a second descriptive string which corresponds to the consumable item and which has a correct spelling of the at least one incorrectly spelled word by applying the machine learning model to the generated vector; calculating a confidence factor regarding the identified second descriptive string using the machine learning model; and when it is determined that the confidence factor exceeds a predetermined threshold, (i) modifying the first data record by replacing the first descriptive string with the second descriptive string, and (ii) storing the modified first data record in the database.

Context and domain sensitive spelling correction in a database

System and method for associating user-entered consumable item description to an entry in a consumable item database. In one embodiment, formally structured restaurant menu item is matched to a large database of food items that has been constructed via crowd-sourcing. A novel, practical, and scalable machine learning solution architecture, consisting of two major steps is utilized. First a query generation approach is applied, based on a Markov Decision Process algorithm, to reduce the time complexity of searching for matching candidates. That is then followed by a re-ranking step, using deep learning techniques, to ensure matching quality goals are met.

System and method for associating user-entered text to database entries

A method of operating a health tracking system is disclosed. The method comprises: receiving a first data record comprising at least a first descriptive string regarding a consumable item, the first descriptive string having at least one word thereof incorrectly spelled; generating a vector using the first descriptive string using a machine learning model; identifying a second descriptive string which corresponds to the consumable item and which has a correct spelling of the at least one incorrectly spelled word by applying the machine learning model to the generated vector; calculating a confidence factor regarding the identified second descriptive string using the machine learning model; and when it is determined that the confidence factor exceeds a predetermined threshold, (i) modifying the first data record by replacing the first descriptive string with the second descriptive string, and (ii) storing the modified first data record in the database.

Apparatus and methods are given for providing enhanced product recommendations. In one embodiment, products are recommended to a consumer based on information manually entered by the consumer and/or derived from data relating to the consumer obtained from a health-monitoring platform. The recommended products are provided from a recommendation engine to a human operator or curator at an item selection entity, the curator selects a predetermined number of the recommended products for delivery to the consumer. Such delivery is established by the consumer to occur periodically (such as monthly) as part of a subscription service. Product recommendations are adjusted over time to relate to customer activity and interest data accumulated from the health-monitoring platform. In one embodiment, a recommendation engine is configured to specifically tailor its recommendation algorithms based on feedback information received from the curator and/or the consumer.

Methods and apparatus for enhanced product recommendations

A method and system for providing more relevant search results and recommendation from a food database is disclosed. The method includes receiving a query, a first candidate food, and a second candidate food. The method includes generating text feature vectors based on the query and food names of the first and second candidate foods using at least one first embedding function of a machine learning model. The method includes determining nutrition content vectors from the nutritional data of the first and second candidate foods. The method includes generating a nutrition content vector based on the query using a second embedding function of the machine learning model. The method includes determining which of the first and second candidate food is more relevant to the query based on the text feature vectors and the nutrition content vectors. The method includes providing search results or recommendation based on the determined relevance.

Growing community of inventors

San Francisco, CA, United States of America

Patrick Howell