Abstract: Methods and systems for processing an order are disclosed. A unified shopping cart receives a first request from a first tenant interface and second request from a second tenant interface, different from the first tenant interface. Each request includes first request data including a customer identifier, a selection of one or more items, and a desired delivery mode. The first request and the second request are associated with the same customer, and are processed using a unified payment transaction.

Abstract: Methods and systems are described for optimizing delivery modes used to transport items to a customer. A delivery optimization system includes an online ordering system, a delivery mode optimizer, an order allocator, and a ship label and manifest generator. The delivery mode optimizer selects an optimal shipping mode based on information received from carriers. The lowest cost option to deliver the goods to the destination by a promised delivery date is selected at the time the order is received. A shipping node is selected that can supply the items needed to fulfill the order within the desired timeframe. The delivery mode is reevaluated after the order is prepared for shipment to determine if available shipping options have changed.

Abstract: A multi-modal machine learning model is disclosed that may be implemented in a recommender system. The model may generate a multi-modal embedding based on a user query, one or more user-selected items, and a conversation history. The one or more user-selected items may include text data and visual data. In some embodiments, the recommender system may use the multi-modal embeddings to recommend one or more items of a plurality of items. In some embodiments, the multi-modal model may be integrated with other systems.

Abstract: In some aspects, a system for enhancing performance of directed information delivery is provided. In one example, an advertising recommendation server receives an advertisement request including an advertisement for display on at least one of a plurality of webpages and a requested performance of the advertisement. The server extracts context data from the advertisement, and provides the extracted context data to a data model useable to generate predictions of performance of the advertisement when displayed on each of a plurality of webpages, trained based on performance data associated with previous displays of one or more previous advertisements on one or more of the plurality of webpages and context data from the one or more previous advertisements. A webpage is identified that meets expected performance, and a recommendation is provided to a decision platform.

Abstract: A system for managing and transferring a virtual shopping cart. The system includes a product identification system that is programmed to identify products from a product identification tag, a cart identification system that is configured to generate a cart identifier that uniquely identifies the virtual shopping cart, and a cart management engine. The cart management engine is configured to receive products to add to the virtual shopping cart from the product identification system, associate the products with the cart identifier, and transfer access to the virtual shopping cart from a first device to a second device.

Abstract: Techniques are provided for assessing and determining risks that are posed by various users to a website. The determined risk for each user can be used to modify the user experience of the website in a manner that is commensurate with the risk (or non-risk) posed by the user. Assessment of risks posed a website guest can be performed by collecting guest-related data from a plurality of different service engines, aggregating contextual information from the guest-related data, and calculating a risk score based on the contextual information. The risk score can represent the internal reputation of the guest using the guest device. The risk score can be transmitted to a remote computing device and used to modify content of the website according to the risk score. Further, a recommended action can be determined based on the risk score, which can be taken by a backend service for the website guest.

Abstract: Generally, the present disclosure relates to methods and systems for algorithmically generating apparel recommendations. In some example aspects, human-identified complementarity of a subset of products can be used to train a neural network, which is in turn used to generate a compatibility score for items. Based on such values, compatible items can be identified and recommended to a user.

Abstract: A website vulnerability test is performed by automatically checking that a website has not been compromised by malicious third party scripts. A system can test a dynamic behavior of a website that indicates a functional user flow through the website. A set of rules are applied against a log of dynamic behavior of the website, as well as static code of the website, to identify potential compromise by malicious scripts. Some rules can be configured for detecting modification of a third party script, or modified behavior of a third party script, in an attempt to detect security monitoring activity against the script and hide its presence from the security monitoring activity.

Abstract: In some implementations, a method performed by data processing apparatuses includes receiving, from a requestor device, an item location request for an item. Map information is received for a mapped space in which the item is located. Item location coordinates are received corresponding to locations within the mapped space at which the item has been previously selected or scanned. The item location coordinates are mapped with respect to the mapped space, and a plurality of clusters are determined for the item location coordinates. For each cluster, a representative location of the cluster is determined. One of the clusters is selected, based at least in part on its representative location, and an estimated location of the item is provided to the requestor device, based at least in part on the representative location of the selected cluster.

Abstract: A swimwear garment configured to be worn by a wearer having buttocks, the swimwear garment comprising an outer rear layer and a rear compression layer. The outer rear layer defines rear portions of two leg openings and extends upwardly toward a waist of the swimwear garment in a manner configured to at least partially cover the buttocks of the wearer. The rear compression layer is coupled to the outer rear layer along an inside surface of the outer rear layer and defines a bottom periphery and opposing side edges. The rear compression layer extends upwardly from the bottom periphery to at least to the waist of the swimwear garment. Each of the opposing side edges of the rear compression layer are sewn to the outer rear layer. The bottom periphery of compression fabric layer is entirely positioned above the at least rear portions of the leg openings.

Abstract: For each order in a supply chain, order details are accessed. A disposition of each load of each order, the disposition comprising a plurality of expected milestones and a plurality of actual milestones. Each load is identified as either completed or projected based on the disposition of the load. An inventory readiness metric for each load is determined based on the disposition of the load. An inventory readiness root cause for each load is determined based on the disposition of the load. Load data for each load is stored in a supply chain data store. The load data comprising the disposition of the load, the inventory readiness metric for the load, and the inventory readiness root cause for the load. Queries about the supply chain are responded to using the load data.

Abstract: Methods and systems for simulating fulfillment of digital orders within a retail supply chain are disclosed. One method includes receiving a selection of a first operational parameter of a supply chain model. The supply chain simulation model is a transaction-level model representative of a digital order fulfillment process within a retail supply chain network. The selection of the first operational parameter includes a default value for the first operational parameter and an experimental value for the first operational parameter that is different from the default value. Simulations of a set of predicted digital orders within the retail supply chain network, using the supply chain simulation model as modified in accordance with the first operational parameter, are performed. Scenario evaluations including predicted metrics associated with each of a cost, a capacity, and a guest experience for the digital order fulfillment process may be output and displayed on a user interface.

Abstract: Methods and systems for managing transportation vessels within an enterprise system are disclosed. In example aspects, based on current inventory levels and transportation vessel constraints, an optimized inventory order may be placed to ensure efficient ordering of an item from a vendor. In some instances, a forward ordering adjustment may be made to efficiently use transportation vessel capacity.

Abstract: Disclosed herein are systems and methods for determining whether an unknown product matches a scanned barcode during a checkout process. An edge computing device or other computer system can receive, from an overhead camera at a checkout lane, image data of an unknown product that is placed on a flatbed scanning area, identify candidate product identifications for the unknown product based on applying a classification model and/or product identification models to the image data, and determine based on the candidate product identifications, whether the unknown product matches a product associated with a barcode that is scanned at a POS terminal in the checkout lane. The classification model can be used to determine n-dimensional space feature values for the unknown product and determine which product the unknown product likely matches. The product identification models can be used to determine whether the unknown product is one of the products that are modeled.

Abstract: In general, methods and system for modeling the incremental value of a response to different types of treatment are disclosed. Some examples include modeling discount sensitivity for specific guests. One aspect is a method for modeling incremental sales for a retail enterprise which includes generating a proxy campaign. In some embodiments, the proxy campaign is used to train an uplift model to predict an uplift score for each guest in response to a proposed campaign. In some embodiments, the uplift score is used to select guests for the proposed campaign.

Abstract: For a measurement period, signal strengths, device identifiers and access point identifiers are received from a plurality of access points. A record for the measurement period is stored in random access memory with the record being associated with a single device identifier and containing access point identifiers associated with signal strengths received for the measurement period. The location of a device is determined for the measurement period by retrieving spatial coordinates of a corresponding access point for each access point identifier in the record for the measurement period and setting the location of the device for the measurement period to the average of the retrieved spatial coordinates.

Abstract: The present disclosure is directed to a system and method for quickly and efficiently resolving customer service issues associated with online shopping. When a customer requiring assistance contacts a customer service agent, the agent may use the customer's identification information to search for the customer's account. After locating the customer's active shopping cart or the customer account, the agent may open a session to view and edit the customer's shopping cart information, order information or account information. The agent may use the one or more administrative options available to the agent and unavailable to the customer to make changes to the customer account as needed. After making changes to the customer account, the agent may direct the customer to refresh their customer account in order to get updated with the changes made by the agent, thus resolving the customer's issues quickly and efficiently.

Abstract: Physical shopping carts can have product detection systems and associate physical shopping carts with mobile computing devices (e.g., smartphones, tablet computing devices, smart watches, wearable computing devices). For example, physical shopping carts can be equipped with one or more product detection systems (e.g., scanners, sensors, cameras) that can electronically tally products that are placed in physical shopping carts. Mobile computing devices can be associated with and mounted on the physical shopping carts to provide a variety of enhanced shopping cart features not possible with conventional physical shopping carts, such as electronically tracking the contents of a shopping cart, checking-out from the mobile computing device (instead of at conventional check-out areas, such as point of sale terminals), and others.