Back to results
US 11,698,908 B2Granted
Content inversion for user searches and product recommendations systems and methods
AssigneePaypal, Inc.
InventorBoris A. Galitsky
Life of this patent
Estimating the term…
Abstract
There is provided systems and method for content inversion for use in user searches and product recommendations. The methods include receiving a first content, wherein the first content includes a first sentence having at least one first sentiment expression, creating a syntactic parse tree of the first sentence, identifying a first sentiment in the at least one first sentiment expression, wherein the first sentiment corresponds to a first polarity, determining a first needs expression corresponding to the first sentiment, wherein the first needs expression includes a first subject. The method may further include creating a sub-tree corresponding to the first needs expression, and grouping the content with other content having similar sub-trees.
Claims20
A system, comprising:
- one or more processors, one or more computer-readable memories, with program instructions stored on the one or more computer-readable memories, the one or more processors configured to execute the program instructions to cause the system to perform operations comprising:
- detecting a search query corresponding to a user;
- creating a user syntactic parse tree for the search query, wherein the user syntactic parse tree comprises units of at least one word corresponding to content of the search query;
- determining a user sentiment and a search subject corresponding to the search query based on the user syntactic parse tree, wherein the user sentiment comprises a user needs expression for the search subject;
- performing an online search utilizing the user sentiment and the search subject with the search query;
- analyzing a plurality of search results corresponding to the online search;
- determining a subject and a sentiment corresponding to each of the plurality of search results;
- determining alignments between words in the search query and each of the plurality of search results;
- based on the alignments:
- applying weights to each of the words in the plurality of search results, and
- calculating generalization scores associated with phrases for the plurality of search results based further on a weighted sum from the weights applied to each of the words;
- determining whether the subject and the sentiment of each of the plurality of search results correspond to the search subject and the user sentiment, respectively;
- determining a rank corresponding to each of the plurality of search results based on the generalization scores and determining whether the subject and the sentiment of each of the plurality of search results correspond to the search subject and the user sentiment, respectively;
- generating a ranked listing of the plurality of search results based on the rank corresponding to each of the plurality of search results, wherein the ranked listing comprises a user interface output for a computing device of the user;
- and
- causing to be displayed, on the computing device of the user, the ranked listing.
13 more elements
A method comprising:
- detecting, by a computer system, a search query corresponding to a user;
- creating, by the computer system, a user syntactic parse tree for the search query, wherein the user syntactic parse tree comprises units of at least one word corresponding to content of the search query;
- determining, by the computer system, a user sentiment and a search subject corresponding to the search query based on the user syntactic parse tree, wherein the user sentiment comprises a user needs expression for the search subject;
- performing, by the computer system, an online search utilizing the user sentiment and the search subject;
- analyzing, by the computer system, a plurality of search results corresponding to the online search;
- determining, by the computer system, a subject and a sentiment corresponding to each of the plurality of search results;
- determining, by the computer system, alignments between words in the search query and each of the plurality of search results;
- determining one or more generalization scores associated with phrases for the plurality of search results based on a weighted sum from weights of each of the words;
- determining, by the computer system, whether the subject and the sentiment of the plurality of search results correspond to the search subject and the user sentiment, respectively;
- determining, by the computer system, a rank corresponding to each of the plurality of search results based on the generalization scores and determining whether the subject and the sentiment of each of the plurality of search results correspond to the search subject and the user sentiment, respectively;
- generating, by the computer system, a ranked listing of the plurality of search results based on the rank corresponding to each of the plurality of search results, wherein the ranked listing comprises a user interface output for a computing device of the user;
- and
- causing to be displayed, by the computer system on the computing device of the user, the ranked listing.
10 more elements
A non-transitory machine-readable medium having stored thereon machine-readable instructions executable to cause a machine to perform operations comprising:
- detecting a search query corresponding to a user;
- creating a user syntactic parse tree for the search query, wherein the user syntactic parse tree comprises units of at least one word corresponding to content of the search query;
- determining a user sentiment and a search subject corresponding to the search query based on the user syntactic parse tree, wherein the user sentiment comprises a user needs expression for the search subject;
- performing an online search utilizing the user sentiment and the search subject;
- analyzing a plurality of search results corresponding to the online search;
- determining a subject and a sentiment corresponding to each of the plurality of search results;
- determining alignments between words in the search query and each of the plurality of search results;
- based on the alignments:
- applying weights to each of the words in the plurality of search results, and
- calculating generalization scores associated with phrases for the plurality of search results based further on a weighted sum from the weights applied to each of the words;
- determining whether the subject and the sentiment of each of the plurality of search results correspond to the search subject and the user sentiment, respectively;
- determining a rank corresponding to each of the plurality of search results based on the generalization scores and determining whether the subject and the sentiment of each of the plurality of search results correspond to the search subject and the user sentiment, respectively;
- generating a ranked listing of the plurality of search results based on the rank corresponding to each of the plurality of search results, wherein the ranked listing comprises a user interface output for a computing device of the user;
- and
- causing to be displayed, on the computing device of the user, the ranked listing.
12 more elements
Dependencies are inferred from the claim text.