Abstract: Methods and systems for generating a recommendation to alter the menu of a restaurant based on the transaction data associated with multiple merchants are described herein. In an example, a method generates one or more first representations of one or more items in the menu and comparing the one or more first representations with one or more candidate tags corresponding to generic names associated with menu items. Based on the similarity between the one or more first representations and the one or more candidate tags, the method identifies a portion of the one or more candidate tags. The method further determines that a first characteristic associated with the merchant corresponds to a second characteristic associated with multiple merchants and selects the multiple merchants based on the second characteristic. Based on transaction data associated with the multiple merchants, the method further generates a recommendation to alter the menu.

Abstract: Generating a vector space model to analyze menu items for restaurants in order to determine optimized geographic locations for merchants to position restaurants is described herein. In one example, the techniques include converting the menu items into vectors using the vector space model, and using the vectors, grouping or assigning the menu items to cuisine types. Once assigned to cuisine types, one or more servers of a food-ordering service may analyze transaction histories for customers in a particular geographic area to determine how “adventurous” the customers are, or how diverse the customer's orders are amongst cuisine types. If the customers in a particular geographic area are adventurous or diverse consumers, the food-ordering service may recommend that a merchant establish a dine-in restaurant in the geographic area, offer food delivery to the geographic area, and/or include the geographic area in a schedule for a mobile restaurant.

Abstract: In one embodiment, a method includes, by one or more computing devices associated with a payment service, receiving, by the payment service and from a first client device associated with a sender, a request for a digital gift to be added to a payment account of a recipient, identifying, one or more recommended digital gifts for the recipient based on transaction history of the recipient on the payment service, where the one or more digital gifts are identified using a machine-learning model trained using a transaction history of one or more users of the payment service, providing, in response to receiving a selection of one of the recommended digital gifts from the first client device, and sending, by the payment service to a second client device associated with the recipient, a notification that the sender has provided the recipient with the identified digital gift.

Abstract: Generating a vector space model to analyze menu items for restaurants in order to determine optimized geographic locations for merchants to position restaurants is described herein. In one example, the techniques include converting the menu items into vectors using the vector space model, and using the vectors, grouping or assigning the menu items to cuisine types. Once assigned to cuisine types, one or more servers of a food-ordering service may analyze transaction histories for customers in a particular geographic area to determine how “adventurous” the customers are, or how diverse the customer's orders are amongst cuisine types. If the customers in a particular geographic area are adventurous or diverse consumers, the food-ordering service may recommend that a merchant establish a dine-in restaurant in the geographic area, offer food delivery to the geographic area, and/or include the geographic area in a schedule for a mobile restaurant.