Abstract: An online concierge system generates suggested instructions for presentation to a user. The online concierge system access instruction examples corresponding to a target item category and generates candidate instruction representations based on instruction messages within each instruction example. The online concierge system generates preliminary scores for the candidate instruction representations that are directly related to an intra-category frequency of use of the instruction tokens of the candidate instruction representation within the target item category. The online system normalizes these preliminary scores for the candidate instruction representations based on the inter-category frequency of use of the instruction tokens in all item categories to generate final scores for the candidate instruction representations. The online concierge system selects a set of instruction representations based on these final scores and generates suggested instructions based on the set of instruction representations.

Abstract: An online concierge system displays an ordering interface to users that displays items offered by various warehouses. The online concierge system includes machine learning availability model that estimates an item's availability and visually distinguishes items offered by a warehouse having less than a threshold availability from other items. Because information from a warehouse that an item that was out of stock is now in stock is often delayed, the online concierge system transmits a request to a shopper fulfilling an order to check for an item's availability at a warehouse. For example, the online concierge system allows users to include a request for an indication of an item's availability when placing an order. When the online concierge system receives a threshold number of requests for the item, the online concierge system prompts a shopper fulfilling an order including items near the item for the item's availability.

Abstract: An online concierge system allows users to add items to an order and provide instructions to a shopper specifying attributes for selecting an item in the order from a warehouse. To simplify entry of the instructions, the online concierge system converts previously received instructions comprising unstructured text into structured suggestions by extracting words from the previously received instructions. The suggestions are associated with items or generic item descriptions. When a user includes an item in an order, the online concierge system displays one or more suggestions associated with the item as selectable options in an interface to simplify specification of instructions for the item.