Abstract: A method for tag-based search includes capturing an image, extracting a tag from the image, identifying a location associated with the captured image, and querying stored content for information that matches the location and the tag. Local storage is checked for the information first, and remote storage may be checked subsequently. Any located information may be used to augment the image. Information located in the remote storage may be saved in the local storage until it reaches a certain age, until it fails to be accessed for a threshold period of time, or until the location moves outside a threshold radius associated with a location of the information located in the remote storage.

Abstract: In an online concierge system, a shopper retrieves items specified in an order by a customer from a retail location. The online concierge system optimizes order fulfillment by selecting a retail location for an order that is most time-efficient and that is most likely to have each of the item in the order available. Hence, the online concierge system may select a less convenient retail location that is more likely to have each item being ordered available. To predict whether a retail location incompletely fulfill the order if selected to fulfill the order, the online concierge system trains a machine learning model based on prior orders fulfilled by the retail location, a shopper retrieving items in the order, items in the order, and other features.

Abstract: An online system receives a completed order of items fulfilled by a picker and a transaction log of the completed order from a retailer where the completed order was fulfilled. The online system compares the completed order of items to the transaction log of the completed order to identify one or more unmatched items. The online system generates, at an auditor client device, a user interface that shows the unmatched items and an option to mark each unmatched item as a fraud instance. The online system receives an indication of a fraud instance for one of the unmatched items from an auditor via the user interface. The online system adds a value of the fraud instance to a fraud total of the picker. If the online system determines that the fraud total for the picker exceeds a threshold, the fraud module deactivates the picker's account.

Abstract: An online concierge system allows users to order items within discrete time intervals later than a time when an order was received. Each order may require a different set of characteristics for fulfilment by shoppers. Because different shoppers may have different capabilities, it is most efficient to reserve shoppers with specialized characteristics for orders that require them. The online concierge system maintains a set of hierarchical structures for different characteristics of shoppers, with each level in a hierarchical structure having a value. To select a shopper to fulfill an order, the online concierge system scores identifies groups of shoppers having characteristics capable of fulfilling the order based on levels in the hierarchical structure for each characteristic of a group. A shopper from a group having a minimum score is selected to fulfill the order.

Abstract: An online system receives, from a customer mobile application (CMA) an order including a high-value item determines that the order includes the high-value item. The online system transmits an indication that the order includes the high-value item to a delivery mobile application (DMA). The DMA transmits a real-time location of a client device of a delivery agent to the online system. Responsive to determining that the delivery agent is at a delivery location, the online system transmits an indication to the DMA to display a user interface including an interactive element for requesting a signature from a customer. Responsive to receiving an indication of an interaction with the interactive element, the online system transmits an indication to the CMA to display a user interface with a signature element. The CMA transmits a signature received via the signature element to the online system, which stores the signature as verification information.

Abstract: An online concierge system maintains a graph of items available for purchase. The graph maintains edges between items, where an edge between an item and an additional item indicates that one or more customers have previously replaced the item with the additional item. The edge between the item and the additional item also identifies a number of times customers have replaced the item with the additional item. When a customer orders an item, the online concierge system traverses the graph of items to identify candidate replacement items for the ordered item and identifies one or more of the candidate replacement items to the customer. When identifying the candidate replacement items, the online concierge system accounts for distance between the ordered item and different candidate replacement items in the item graph.

Abstract: Multiple users of a household can each have a different profile associated with a common account for an online concierge system. This association of different profiles with the common accounts allows the concierge system to show a user of the household what other users of the household purchased and rank items for suggestion to the user based one purchases of other users, facilitating order building. The online concierge system also enables a user profile to designate a user profile associated with the common account for an order pickup or as a contact for a home delivery. Additionally, association of different user profiles with the common account may be used for account recovery of one of the user profiles. Further, different user profiles may have different permissions for creating an order with the online concierge system.

Abstract: An online shopping concierge system sorts a list of items to be picked in a warehouse by receiving data identifying a warehouse and items to be picked by a picker in the warehouse. The system retrieves a machine-learned model that predicts a next item of a picking sequence of items. The model was trained, using machine-learning, based on sets of data that each include a list of picked items, an identification of a warehouse from which the items were picked, and a sequence in which the items were picked. The system identifies an item to pick first and a plurality of remaining items. The system predicts, using the model, a next item to be picked based on the remaining items, the first item, and the warehouse. The system transmits data identifying the first item and the predicted next item to be picked to the picker in the warehouse.

Abstract: An online concierge system receives an order from a customer. The online concierge system transmits a notification to the customer's client device indicating that the order is ready for pick up and receives location data from the customer's client device as the customer travels to a pickup location. In response to the online concierge system receiving a first indication that the customer has entered an outer geofence, the online concierge system transmits a second notification to a runner's client device that the customer is in transit. In response to the online concierge system receiving a second indication that the customer has entered an inner geofence, the online concierge system starts a timer. When the online system receives a confirmation that the order has been picked up by the customer, it stops the timer and computes a wait time for pick up of the order.

Abstract: When analyzing operation of an online system, such as an online concierge system, that receive queries from users and return results in response to those queries, queries received by the online system may include different numbers of terms. The variations in numbers of terms included in different queries makes it difficult to efficiently group queries for analysis, resulting in inefficient analysis of individual queries when analyzing different queries based on their structure. To more efficiently analyze different queries, the online system groups queries based on a number of terms in the query. In some embodiments, queries are grouped based on a logarithmic scale of the number of terms in the query, while in other embodiments queries are grouped based on a square root of the number of terms in the query.

Abstract: A customer places an order of items to be purchased with an online concierge system. The online concierge system provides the order to a picker who shops for the items at a retailer and delivers them. The online concierge system requests an image of a receipt of the order from the picker. The online concierge system performs image processing on the image of the receipt and uses machine learning and optical character recognition to determine the actual amounts purchased of items. The online concierge system charges the customer based on the actual amounts purchased of each item.

Abstract: An online system receives a recipe from a customer mobile device. The online system performs natural language processing on the recipe to determine parsed ingredients. For each of one or more of the determined parsed ingredients, the online system maps the parsed ingredient to a generic item. The online system queries a product database with the mapped generic item to obtain one or more products associated with the mapped generic item. The online system applies a machine-learned conversion model to each of the one or more products to determine a conversion likelihood for the product. The conversion model may be trained based on historical data describing previous conversions made by customers presented with an opportunity to add products to an order. The online system selects a product from the one or more products based on the determined conversion likelihoods and adds the selected product to an order.

Abstract: An online shopping concierge service allows shoppers to purchase items on behalf of customers and checkout using a mobile application, circumventing traditional point-of-sale check-out systems. A customer places an order using a mobile application or website associated with the online shopping concierge service. The online shopping concierge service charges a payment instrument of the customer in the value of the selected items. The system transmits the order to a shopper, who receives an order for fulfillment on a mobile device. The shopper collects and scans items using a mobile application. The mobile application transmits an identification of the items for purchase and their total cost to the online shopping concierge service, which transmits payment to the retailer. Alternatively, the mobile application encodes an identification of the items for purchase into an encoded image, which is scanned by a cashier, allowing the shopper to complete an accelerated check-out.

Abstract: A method and computer program product for managing information about transactions occurring in a physical establishment includes gathering information about multiple transactions at a physical establishment, and storing the information as metadata. A byte stream corresponding to the metadata is generated. Several elements of the metadata are associated with a particular session based on predefined criteria.

Abstract: A method for tag-based search includes capturing an image, extracting a tag from the image, identifying a location associated with the captured image, and querying stored content for information that matches the location and the tag. Local storage is checked for the information first, and remote storage may be checked subsequently. Any located information may be used to augment the image. Information located in the remote storage may be saved in the local storage until it reaches a certain age, until it fails to be accessed for a threshold period of time, or until the location moves outside a threshold radius associated with a location of the information located in the remote storage.

Abstract: An online shopping concierge system identifies a set of delivery orders and a set of delivery agents associated with a location. The system allocates the orders among the agents, each agent being allocated at least one order. The system obtains agent progress data describing travel progress of the agents to the location, and order preparation progress data describing progress of preparing the orders for delivery. The system periodically updates the allocation of the orders among the agents based on the agent progress data and the order preparation progress data. This involves re-allocating at least one order to a different delivery agent. When a first agent arrives at the location, the system assigns to the first agent the orders allocated to the first agent. The system then removes the first agent from the set of available delivery agents, and removes the assigned delivery orders from the set of delivery orders.

Abstract: An online concierge system receives an order from a customer with a list of items and allocates the order to a retailer at which the items can be picked. The retailer is associated with a payment processing system having a corresponding tokenization process. The online concierge system uses the tokenization process for the retailer to generate a payment token for the customer based on stored payment data and transmits the payment token to the payment processing system. The online concierge system determines a first payment amount from the customer to the retailer based on items picked at the retailer and a second payment amount from the retailer to the online concierge system. The payment processing system charges the customer a sum of the first payment amount and the second payment amount using the payment token. The online concierge system receives the second payment amount from the payment processing system.