A method and system in which the system receives input data for a first unit to determine a unit feature signal and a unit demand signal including predictive demand features. The system then determines, for the first unit at each of a plurality of offer values, a set of associated estimated probabilities that the offer value will be accepted within a short-term time period; determines an expected time-to-sell for the first unit at each of the plurality of offer values; determines an optimal offer value for the first unit; and displays the optimal offer value for the first unit on a webpage.

An online concierge system trains a machine learning conversion model that predicts a probability of receiving an order from a user when the user accesses the online concierge system. The conversion model predicts the probability of receiving the order based on a set of input features that include price and availability information. For each access to the online concierge system, the online concierge system applies the conversion model to a current price and availability and to an optimal price availability. The online concierge system generates a metric as the difference between the two predicted probabilities of receiving an order.

In an approach to jointly learning uncertainty-aware trend-informed neural network for a demand forecasting model, a machine learning model is trained to capture uncertainty in input forecasts. The uncertainty in a latent space is represented using an auto-encoder based neural architecture. The uncertainty-aware latent space is modeled and optimized to generate an embedding space. A time-series regressor model is learned from the embedding space. A machine learning model is trained for trend-aware demand forecasting based on said time-series regressor model.

A management system operates in conjunction with entities to provide component recommendations for objects. The management system trains a machine learning model used to generate the component recommendations. The machine learning model is trained based on historical component entries describing components previously provided and identifiers of the components. The management system generates training data by classifying the historical component entries into predetermined component classifications. After the machine learning model is trained, the management system generates a customized recommendation of components for an object based on likelihoods of selection of the predetermined component classifications.

A system and method for predicting behavior and/or outcomes related to a consumer's experience with an organization are implemented. Household data for households that are associated with a customer service interaction as of a certain date is collected, the household data having been created over a first pre-determined period of time preceding the certain date. The household data is analyzed to identify positive household data sets and negative household data sets. The positive household data sets relate to customer service interactions which preceded a high level customer service interaction within a subsequent period of time and the negative household data sets relate to customer service transactions which did not precede a high level customer service interaction with the subsequent period of time. The positive household data sets and the negative household data sets are processed in the aggregate, using a trained support vector machine model, to determine cumulative differences between data contained within the positive household data sets and the negative household data sets. Each day, daily household data is collected. The daily household data describes individual customer service transactions occurring during a previous calendar day. The daily household data is processed using the model to determine whether each individual customer service transaction occurring during the previous calendar day is more similar to the positive household data sets or to the negative household data sets. The individual customer service transactions that are more similar to the positive household data sets are flagged for proactive intervention.

A system and method for predicting behavior and/or outcomes related to a consumer's experience with an organization are implemented. Household data for households that are associated with a customer service interaction as of a certain date is collected, the household data having been created over a first pre-determined period of time preceding the certain date. The household data is analyzed to identify positive household data sets and negative household data sets. The positive household data sets relate to customer service interactions which preceded a high level customer service interaction within a subsequent period of time and the negative household data sets relate to customer service transactions which did not precede a high level customer service interaction with the subsequent period of time. The positive household data sets and the negative household data sets are processed in the aggregate, using a trained support vector machine model, to determine cumulative differences between data contained within the positive household data sets and the negative household data sets. Each day, daily household data is collected. The daily household data describes individual customer service transactions occurring during a previous calendar day. The daily household data is processed using the model to determine whether each individual customer service transaction occurring during the previous calendar day is more similar to the positive household data sets or to the negative household data sets. The individual customer service transactions that are more similar to the positive household data sets are flagged for proactive intervention.

Disclosed are one or more embodiments for a unique and personalized experience for a user interacting with an electronic commerce site by identifying user-preferred item attributes using supervised machine learning and presenting items to the user in an arrangement that is based on the identified item attributes. A shopping mission is determined according to user interactions with an electronic commerce site. The shopping mission is applied to an attribute prediction model that is trained to detect user-preferred item attributes for items included the item category and estimate a likelihood that an item containing a particular attribute will be purchased or interacted with during the interactions with the electronic commerce site.

Disclosed are one or more embodiments for a unique and personalized experience for a user interacting with an electronic commerce site by identifying user-preferred item attributes using supervised machine learning and presenting items to the user in an arrangement that is based on the identified item attributes. A shopping mission is determined according to user interactions with an electronic commerce site. The shopping mission is applied to an attribute prediction model that is trained to detect user-preferred item attributes for items included the item category and estimate a likelihood that an item containing a particular attribute will be purchased or interacted with during the interactions with the electronic commerce site.

Techniques and solutions are described for facilitating the use of machine learning techniques. In some cases, filters can be defined for multiple segments of a training data set. Model segments corresponding to respective segments can be trained using an appropriate subset of the training data set. When a request for a machine learning result is made, filter criteria for the request can be determined and an appropriate model segment can be selected and used for processing the request. One or more hyperparameter values can be defined for a machine learning scenario. When a machine learning scenario is selected for execution, the one or more hyperparameter values for the machine learning scenario can be used to configure a machine learning algorithm used by the machine learning scenario.

Event-based replenishment simulation for an enterprise supply chain is described. On a per item, per node, per epoch basis, the simulation may generate a stream of action events based on forecasted demand, supply chain logic, and policy inputs that are applied to a current-run state of the supply chain in order to yield a stream of observation events. Requested metrics may be received, and the observation events may then be transformed to predict values for the metrics as output of the simulation. The simulation may be repeated for a given epoch using discrete demand values from a demand distribution, for a plurality of epochs, and/or across a plurality of items at a plurality of nodes. Resultantly, the simulation output can be used for predicting a future run-state of the supply chain across items and nodes.