Knowing a user’s purchase history can be helpful in refining, rewriting, or correcting a shopping related search query the user enters since users often purchase products that are related to or same as products bought previously. However, the text input in searches for products is often short, ambiguous, or underspecified. Even in cases where a user includes relevant detail in a query, manual formulation of the search terms, often from memory, can result in errors. Current shopping search mechanisms do not help detect and refine underspecified or erroneous queries. This disclosure describes techniques to improve keyword-based product search by appropriate query reformulation and search result annotation. With user permission, the reformulation and annotation is based on the user’s inferred product purchase history.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Sharifi, Matthew and Carbune, Victor, "Refining Shopping Search Queries and Results Based On Inferred Product Purchases", Technical Disclosure Commons, (July 02, 2020)