Viant Technology LLC

Varoujan Bedirian

Adrian Witas

Kyle Kincaid

A method and system provide the ability to forecast events that match a criterion. Input is received and includes an event set and a criterion. The event set includes events and each of the events is a record with features. Each feature is an instance of a feature type (FT) and feature value (FV) combination. The criterion is a logical statement that specifies a criterion FT, a criterion FV, and an operator. The event set is partitioned and a partition BitSet is built for each partition. Each partition length of each partition BitSet is based on a number of the events. Based on the partition BitSets, a partition FT-FV count is created. For each partition, a partition FT criterion BitSet is created and a partition Criterion BitSet is computed based thereon. The forecast is then computed based on the partition Criterion BitSet.

Method and system for forecasting information based on events and criterion

A method and system provide the ability to determine matching subjects for features provided in a stimulus. Subject information is pre-processed by receiving input (a feature type definition and a subject set) and building BitSets across all subjects in the subject set. The stimulus is received and includes stimulus features. For each feature type in the stimulus features, all feature type BitSets from the subject set that match are assembled. An evaluation BitSet is built by combining the assembled feature type BitSets. Matching subjects that match the stimulus are determined based on the evaluation BitSet.

Growing community of inventors

Los Angeles, CA, United States of America

Varoujan Bedirian