AI Solutions for Always-Optimized Operations
Patent: US20190138622A1
A classifier is computed as follows. For a first set of values of primary field(s)… A classifier is computed according to a selected subset of the classification features.
Patent: US20190205470A1
Generates hypotheses by creating queries from dataset attributes, searching an unstructured data corpus, and defining new attributes based on the search results.
Patent: US20210056437A1
Matches subpopulations of users and entities by analyzing latent factors from a recommender process, computing semantic models, and identifying statistically significant pairs of user-entity features that drive correlations.
Patent: US9324041B2
Generates a stream of combination functions by combining building‑block functions, applies each to data units, and identifies those correlated with a target variable.
Patent: US20170017900A1
Generates classification features from raw datasets containing arbitrary object types, selects pivotal features based on correlation with target variables, and documents them for machine learning use.
Patent: US10977581B2
Automatically links values from a labeled primary dataset to secondary, unclassified datasets to generate binary classification features, and builds a classifier based on a selected subset of these features.
