Abstract: A computer-implemented method including automatically generating predictions of a respective number of items that a user is likely to reorder in each of groups of the items that a user has ordered historically. The method also can include ranking the groups based on the predictions of the respective number of the items the user is likely to reorder in each of the groups. The method additionally can include transmitting for display to the user a user interface including the groups of the items. Other embodiments are described.

Abstract: A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform operations: training, using labeled training data and a list of substitutes for an item, a machine learning algorithm; determining, using the machine learning algorithm, as trained, a respective similarity score for each substitute of the list of substitutes; ranking each substitute of the list of substitutes based on its respective similarity score; and re-training the machine learning algorithm based on at least the labeled training data and a highest ranked substitute of the list of substitutes. Other embodiments are disclosed herein.

Abstract: A system including one or more processors and one or more non-transitory computer-readable media storing computing instruction that, when executed on the one or more processors, cause the one or more processors to perform operations: generating, using a training procedure, labels based at least in part on price band activity data from a time period; training, using the training procedure, an affinity prediction model of a machine learning architecture; analyzing, using the affinity prediction model of the machine learning architecture, as trained, the price band activity data indicating interactions of a user with items; and generating, using the labels and the affinity prediction model of the machine learning architecture, as trained, one or more price affinity predictions for one or more items for the user. Other embodiments are disclosed herein.

Abstract: A system can include one or more processors and one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform: selectively aggregating in-session user activity of a user with historical activity data of the user into one or more respective groups based on interactions of the user with a GUI over a period of time; predicting, using a set of predictive algorithms, one or more intents of the user based on the one or more respective groups; and facilitating a display of an altered GUI on an electronic device of the user based on the one or more intents of the user, as predicted. Other embodiments are disclosed herein.

Abstract: This application relates to systems and methods for providing search results based on a primary intent. In some examples, a disclosed system includes a memory resource storing instructions; and one or more processors coupled to the memory resource. The one or more processors are configured to execute the instructions to: receive, from a user, a search query including a plurality of words, identify a plurality of intention terms from the words of the search query, compute, for each of the plurality of intention terms, a compatibility score between the intention term and a query context associated with the intention term, determine, from the plurality of intention terms, a primary intention term having a maximum compatibility score among the plurality of intention terms, and generate, based on the primary intention term, a set of intent-based search results in response to the search query, the set of intent-based search results identifying a set of items associated with the primary intention term.

Abstract: A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions configured to run on the one or more processors and perform generating personalized product-type metrics for the user based at least in part on a user embedding for the user and product-type embedding Gaussian mixture distributions; determining top product types based at least in part on personalized product-type complementarity metrics generated using the personalized product-type metrics and cosine similarity measurements; generating a set of first items associated with the top product-types; ranking each respective item in the set of first items generated using an item-level embedding Gaussian distribution for the anchor item and a respective item-level embedding Gaussian distribution for the each respective item; and selecting a set of top items as the personalized complementary item recommendations based on the ranking. Other embodiments are disclosed.

Abstract: A method including training a recurrent neural network model to create a trained model based at least in part on: (a) first images associated with first items on a website, (b) first search terms used by users of the website to search for the first items on the website, and (c) personal features of the users. The method also can include receiving an input image that was uploaded by a current user. The method additionally can include obtaining a user encoded representation vector for the current user based on a set of personal features of the current user. The method further can include generating an image encoded representation vector for the input image. The method additionally can include deriving search terms that are personalized to the current user for the one or more items depicted in the input image, using the trained model and based on the user encoded representation vector for the current user and the image encoded representation vector for the input image. Other embodiments are disclosed.

Abstract: A basket-aware recommendation system can include a computing device configured to receive an indication of a navigation from a user device. The indication includes a user identifier and a set of basket items. The computing device is configured to generate a set of eligible items based on the set of basket items. The computing device is configured to generate a feature representation based on the corresponding item category count within the set of eligible items and select items of the set of eligible items within the corresponding item category. The computing device is configured to compute a number of category display items based on the potential click value and the normalized relevance value. The computing device is configured to transmit the number of category display items for each category of the set of item categories to the user device for display on a user interface.

Abstract: A system comprising one or more processors and one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform operations comprising: evaluating, using a price band determination model, degrees of expensiveness of items relative to each other in item type categories; generating, using the price band determination model, price bands associated with item type categories; assigning each of the items to a respective one of the price bands associated with a respective one of the item type categories; and presenting, to one or more end-user applications, at least one other item corresponding to at least one of the price bands associated with at least one of the item type categories. Other embodiments are disclosed.

Abstract: This application relates to apparatus and methods for automatically determining item relevancy based on textual information. In some examples, a computing device receives a search query, and a plurality of items corresponding to the search query. The computing device may identify one or more features of the search query. The computing device may generate relevancy values for each of the items based on the features of the search query, and features of each of the plurality of items. For example, the computing device may generate, for each of the items, a plurality of relevance values, each relevance value generated based on a feature of the search query and corresponding features of the item. The computing device may transmit the generated relevancy values for the plurality of items. In some examples, the computing device may rank the plurality of items based on the generated relevancy values.

Abstract: This application relates to apparatus and methods for automatically determining and providing digital recommendations including discounted items with associated savings. In some examples, a computing device obtains savings data including savings items. In response to obtaining the savings data, the computing device obtains a plurality of clusters. Each cluster of the plurality of clusters includes a set of product types. The computing device obtains user data associated with a user, and based on the savings data and the plurality of clusters, determines a set of initial items for recommendation. The set of initial items includes a subset of the savings items. The computing device also determines final recommendation items for the user based on the set of initial items and the user data.

Abstract: Systems and methods including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform functions comprising: generating one or more feature vectors for a user, the one or more feature vectors at least comprising transaction-based features and slot-based features; generating, using a machine learning architecture, a repurchase prediction for the user based, at least in part, on the one or more feature vectors; generating, using the machine learning architecture, a time slot prediction for the user based, at least in part, on the one or more feature vectors, the time slot prediction predicting a time slot desired by the user for an upcoming transaction; and executing a reservation function that facilitates reserving of the time slot for the user. Other embodiments are disclosed herein.

Abstract: In some examples, a system may be configured to obtain a set of features of a set of users including one or more features of transaction of the set of users and one or more features of engagement data of the set of users. Additionally, the system may be configured to implement a first set of operations that generate output data including a plurality of churn scores, based on the set of features. In some examples, each churn score of the plurality of churn scores being associated with a particular user of the set of users and characterize a likelihood of a churn event of the corresponding user.

Abstract: A system including one or more processors and one or more non-transitory media storing computer instructions configured to run on the one or more processors and perform: identifying a first sub-population of case individuals from a gross population of the case individuals; presenting first test content to a first test sub-population of the case individuals, the first test content is selected according to a first statistical model; measuring a first test sub-population average feedback metric based on first test content feedback provided from the first test sub-population of the case individuals; determining that the first test sub-population average feedback metric exceeds a first control population average feedback metric of a first control population of the case individuals; and determining that a probability value for a difference between the first test sub-population average feedback metric and the first control population average feedback metric is less than a predetermined significance level value.

Abstract: A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, perform certain acts. The acts can include obtaining reorder likelihood scores for items that a user has ordered historically. The acts also can include grouping the items into groups using a taxonomy. The acts additionally can include adjusting the groups based on a respective number of items in each of the groups and a respective group-specific threshold for each of the groups. The acts further can include ranking the items within the groups based on the reorder likelihood scores. Other embodiments are described.

Abstract: Systems and methods for content selection and presentation are disclosed. Training data including indicating one or more interactions with one or more content elements and associated with one of a plurality of individual contexts is received. A selection model is trained by applying a reinforcement learning mechanism and an individual explore-exploit mechanism. A context for a user is selected by applying the selection model, which is configured to determine an expected future reward value of at least one of the plurality of individual contexts, determine an expected future reward value of a global context based on a past click-through rate, a reward value, and a future click-through rate, and, select the global context or one of the one or more individual contexts based on a comparison of the expected future reward values.

Abstract: Systems and methods for customizing an interface are disclosed. Interaction data is received for at least one interaction with a network environment and an intent of the at least one interaction is identified based on the interaction data. A layout is selected for a user interface based on the identified intent. The layout includes at least one zone.

Abstract: A system comprising one or more processors and one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform: determining a list of possible substitutes for an item when the item is out of stock; training a machine learning algorithm, using labeled training data as input and a list of possible substitutes for the item as output; determining, using the machine learning algorithm, as trained, a respective similarity score for each substitute of the list of possible substitutes; determining a respective historical substitution score for each possible substitute; determining a respective final score for each possible substitute comprises using at least one or more rectifiers having ReLU non-linearity; ranking each possible substitute; storing a selection of the highest ranked possible substitute as additional training data with the labeled training data; and re-training the machine learning algorithm

Abstract: Systems and methods including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform acts of: determining price bands for an item type category; associating items included in the item type category with the price bands; analyzing, using an affinity prediction model of the machine learning architecture, price band activity data indicating interactions of a user with respective items included in each of the price bands for the item type category; and generating one or more price affinity predictions for the user based, at least in part, on the price band activity data, wherein the one or more price affinity predictions predict a preference of the user for respective items associated with one or more of the price bands. Other embodiments are disclosed herein.

Abstract: A method being implemented via execution of computing instructions configured to run at one or more processors and stored at one or more non-transitory computer-readable media. The method can include training two sets of item embeddings for items in an item catalog and a set of user embeddings for users, using a triple embeddings model, with triplets. The triplets each include a respective first user of the users, a respective first item from the item catalog, and a respective second item from the item catalog, in which the respective first user selected the respective first item and the respective second item in a respective same basket. The method also can include randomly sampling an anchor item from a category of items selected by a user. The method additionally can include generating a list of complementary items using a query vector associated with the user and the anchor item.