Ebay, Incorporated

Paypal, Inc.

Wenyan Hu

Alvaro Bolivar

Randall Scott Shoup

Xiaodi Zhang

Shuhui Chen

Xiaoyuan Wu

In various exemplary embodiments, a system and associated method predict auto-complete recommendations for a partial search request. A search engine receives a partial search request with one or more characters from a client device. A first set of auto-complete recommendations corresponding to the one or more characters of the partial search request are identified. A second partial search request is predicted that comprises a set of characters with one or more additional characters added to the one or more characters from the partial search request. A second set of auto-complete recommendations corresponding to the second partial search request is identified. The first set of auto-complete recommendations and the second set of auto-complete recommendations are returned to the client device as a response to the partial search request.

Search box auto-complete

In various exemplary embodiments, a system and associated method for prioritizing search results in an electronic environment is disclosed. In one embodiment, a plurality of characters is received from an end-user as a query. A plurality of equivalent keywords is selected responsive to the query with each of the equivalent keywords being evaluated based on a probability of matching the query of the end-user. A rank of each of the equivalent keywords is determined to form a response, which is then transmitted to the end-user. Other methods and related systems are disclosed.

Ranking algorithm for search queries

A method and a system to rank search results based on dwell time is provided. The system comprises a search module to identify a plurality of listings stored in a listing database as search results. A dwell time module determines a respective dwell time associated with each of the plurality of listings. The dwell time is based on an elapsed amount of time one or more buyers view a view item page associated with the listing. A ranking module ranks the listings composing the identified plurality of listings based at least in part on the respective dwell time associated with each of the plurality of listings.

Systems and methods for ranking results based on dwell time

In various exemplary embodiments, systems and associated methods to predict search results in an electronic environment are disclosed. In one embodiment, a method to provide responses to a search request includes receiving, from an end-user, one or more characters that form a portion of the search request. Prior to receiving a completed version of the search request from the end-user, a plurality of keywords is predicted based on the one or more characters. The prediction is based on the most probable keywords selected from a plurality of prior requests stored in a database. A plurality of responses is prepared based on the most probable keywords. The responses are fit within a single communications packet and returned as a reply to the search request. Other systems and methods are disclosed.

Predictive algorithm for search box auto-complete

A system and method for personalized search. The system includes a search engine that receives search input from a user and a user identification and generates a search result based on the search input; and a profiling engine to gather profile data, generate a user profile associated with a user, and rank the search result personalized to the specific user using the user profile.

System and method for personalized search

In various embodiments, a method of prioritizing search results in an electronic environment is disclosed. The method includes selecting a plurality of equivalent keywords responsive to a search query from an end-user, determining a rank of each of the plurality of equivalent keywords, selecting a plurality of most highly ranked keywords from the plurality of equivalent keywords, and preparing a response including the plurality of most highly ranked keywords.

In various exemplary embodiments, a system and associated method to predict search results in an electronic environment is disclosed. The system architecture comprises a communications module to receive a query from a remote end-user where the query includes one or more characters contained within a first single communications packet. A query engine is coupled to the communications module. The query engine uses the one or more characters from the query to predictively match keywords contained within a data structure (e.g., a database) and forms a response from the one or more characters. The response is transmitted back to the end-user within a second single TCP communications packet.

In various exemplary embodiments, a system and associated method for prioritizing search results in an electronic environment is disclosed. The system comprises a communications module to receive a search query from an end-user. The search query includes one or more characters contained within a single TCP packet. A query engine coupled to the communications module matched the one or more characters from the search query against a plurality of selected keywords contained within a database against the search query. Each keyword is then ranked within the plurality of selected keywords thus forming a subset of selected keywords. A response including the subset of selected keywords is returned to the end-user.

Ranking algorithm for search box auto-complete

A computerized method and a system for providing geographic demand distribution and forecast are provided. Example embodiments may include receiving a criterion to identify an item and searching a database for data associated with a geographic distribution of demand for the item identified by the criterion. The computerized method may also include presenting the geographic distribution of the demand for the item on a map. The map may display multiple geographic areas; each one of the multiple geographic areas may be displayed with a first and a second attribute. The first and the second attributes may be associated with the demand.

Geographic demand distribution and forecast

Potential bidding errors may be detected using a computer-implemented method comprising receiving a proposed bid for an auction item, the auction item having a current bid and a minimum bid increment, the proposed bid being greater than the sum of the current bid and the minimum bid increment. The method then determines a threshold value, the threshold value being greater than the sum of the current bid and a minimum bid increment. When the proposed bid is greater than the threshold value, then the method validates the proposed bid using the current bid and the proposed bid to provide a validation result. Additional apparatus, systems, and methods are disclosed.

Growing community of inventors

Shanghai, China

Wenyan Hu