Utilizing one hash permutation and populated-value-slot-based densification for generating audience segment trait recommendations

US11109085B2 · kind B2 · utility

Assignee

• Adobe Inc. · US

Inventors

• Anup Rao · San Jose, US
• Yasin Abbasi Yadkori · San Francisco, US
• Tung Mai · San Jose, US
• Ryan A. Rossi · Mountain View, US
• Ritwik Sinha · Bengaluru, IN
• Matvey Kapilevich · Ardsley-on-Hudson, US
• Alexandru Ionut Hodorogea · Bucharest, RO

Key dates

• Filing date: Mar 28, 2019
• Grant date: Aug 31, 2021
• Expiry date: Jul 22, 2039

Classification

• Technology area (CPC H): Electricity
• CPC primary: H04N21/4826
• WIPO field: Audio-visual technology
• WIPO sector: Electrical engineering

Abstract

The present disclosure relates to training a recommendation model to generate trait recommendations using one permutation hashing and populated-value-slot-based densification. In particular, the disclosed systems can train the recommendation model by computing sketch vectors corresponding to traits using one permutation hashing. The disclosed systems can then fill in unpopulated value slots of the sketch vectors using populated-value-slot-based densification. The disclosed systems can combine the resulting densified sketches to generate the trained recommendation model. For example, in some embodiments, the disclosed systems can combine the sketches by generating a plurality of locality sensitive hashing tables based on the sketches. In some embodiments, the disclosed systems generate a count sketch matrix based on the sketches and generate trait embeddings based on the count sketch matrix using spectral embedding. Based on the trait embeddings, the disclosed systems can utilize the recommendation model to flexibly and accurately determine the similarity between traits.