Embedding calendar knowledge in event-driven inventory forecasting

US8620720B2 · kind B2 · utility

Assignee

• YAHOO HOLDINGS, INC. · US

Inventors

• Konstantin Salomatin · San Francisco, US
• Datong Chen · Sunnyvale, US
• Qi He · San Jose, US
• Peiji Chen · Saratoga, US

Key dates

• Filing date: Apr 28, 2011
• Grant date: Dec 31, 2013
• Expiry date: Nov 14, 2031

Classification

• Technology area (CPC G): Physics
• CPC primary: G06Q30/0241
• WIPO field: IT methods for management
• WIPO sector: Electrical engineering

Abstract

Systems and methods for automatically forecasting the future availability of one or more resources, such as Internet advertising opportunities, are described herein. In accordance with various embodiments, a forecasting model that accounts for event-driven resource availability is trained based both on historical supply data and calendar information specifying events and event duration. The trained forecasting model is then used to forecast the availability of resources at one or more specified future time periods. In accordance with certain embodiments, the forecasting model comprises a Gaussian process model that has an event-driven kernel as a covariance function.