Predictive conversion systems and methods
First Claim
1. A computer-implemented method of generating and presenting interactive search results, the method comprising:
- providing, by a computer system, a user interface that comprises functionality that enables a user to interactively search for and select items;
maintaining, by the computer system, one or more electronic data stores that store information relating to a plurality of items each comprising a plurality of attributes and a category, the stored information comprising predicted conversion factors each configured to indicate a predicted likelihood of conversion of one of the plurality of items;
accessing, by the computer system via a computer network, a plurality of network-accessible feeds to obtain information related to the plurality of items;
generating, by the computer system, metadata based on the information obtained from the plurality of network-accessible feeds, the metadata comprising Uniform Resource Locators (URL'"'"'s) for webpages at which the plurality of items are for sale,wherein the metadata further comprises stated metadata and derived metadata, the stated metadata comprising attributes obtained from item descriptions provided by a source or promoter of items relating to the item descriptions, the derived metadata comprising attributes obtained from a third party source,database-linking, in an electronic data store, the generated metadata to the plurality of items;
generating, by the computer system, the predicted conversion factors by inputting the generated metadata into a logistic regression formula, the logistic regression formula generated based at least in part on an analysis of historical conversion activity and stated metadata and derived metadata related to attributes associated with items different than the plurality of items to determine statistically how each attribute associated with the items different than the plurality of items affected the historical conversion activity, the logistic regression formula configured to apply a result of the analysis to the inputted metadata to output the predicted conversion factors,wherein the items different than the plurality of items comprise a same category as the plurality of items but comprise one or more different attribute values;
receiving, by the computer system, a user search request generated via the user interface, the user search request comprising an item search criteria;
searching, by the computer system, the one or more electronic data stores for a plurality of items relating to the item search criteria;
generating, by the computer system, for the user a personalized user interface data that provides functionality for the user to select from a result set of items based on the searching, the result set being prioritized based on the predicted conversion factors;
causing, by the computer system, rendering of a search results output, based on the personalized user interface data, the search results output presents the prioritized result set and enables interaction with, by the user, the plurality of items relating to the item search criteria;
electronically monitoring, by the computer system, over a computer network, user interactions with the search results output to determine a user response to the search results output, the user response comprising at least a selection of a URL associated with an item;
adapting, by the computer system, the logistic regression formula based on the user response, the adapting comprising at least including an indication of the selection of the URL in the historical conversion activity analyzed to generate the logistic regression formula;
regenerating, by the computer system, one or more of the predicted conversion factors by inputting the generated metadata into the adapted logistic regression formula;
receiving, by the computer system, a second search request that comprises the item search criteria; and
generating, by the computer system, a second result set of items based on the second search request, the second result set comprising a different prioritization based at least in part on the regenerated predicted conversion factors,wherein the computer system comprises one or more physical servers.
1 Assignment
0 Petitions
Accused Products
Abstract
In one embodiment, a system and method of predicting sale transaction conversion rate of an item operates through a search of information in response to a query over a network. The item can be included in a category of items. Information for other relevant items of the category is available through network query and historical data, among others. Respective information for the other items of the category is available to the method. The system and method includes discovering available information of the item of interest, extracting certain of the available information of the item, analyzing the certain information for the item by comparing the information to other item information for the category of items, weighting the information for the commercial item in comparison to other items of the category, calculating a predictive score for the commercial item of interest, and presenting the information of the commercial item of interest ranked according to the predictive score as compared to other items of the category.
156 Citations
15 Claims
-
1. A computer-implemented method of generating and presenting interactive search results, the method comprising:
-
providing, by a computer system, a user interface that comprises functionality that enables a user to interactively search for and select items; maintaining, by the computer system, one or more electronic data stores that store information relating to a plurality of items each comprising a plurality of attributes and a category, the stored information comprising predicted conversion factors each configured to indicate a predicted likelihood of conversion of one of the plurality of items; accessing, by the computer system via a computer network, a plurality of network-accessible feeds to obtain information related to the plurality of items; generating, by the computer system, metadata based on the information obtained from the plurality of network-accessible feeds, the metadata comprising Uniform Resource Locators (URL'"'"'s) for webpages at which the plurality of items are for sale, wherein the metadata further comprises stated metadata and derived metadata, the stated metadata comprising attributes obtained from item descriptions provided by a source or promoter of items relating to the item descriptions, the derived metadata comprising attributes obtained from a third party source, database-linking, in an electronic data store, the generated metadata to the plurality of items; generating, by the computer system, the predicted conversion factors by inputting the generated metadata into a logistic regression formula, the logistic regression formula generated based at least in part on an analysis of historical conversion activity and stated metadata and derived metadata related to attributes associated with items different than the plurality of items to determine statistically how each attribute associated with the items different than the plurality of items affected the historical conversion activity, the logistic regression formula configured to apply a result of the analysis to the inputted metadata to output the predicted conversion factors, wherein the items different than the plurality of items comprise a same category as the plurality of items but comprise one or more different attribute values; receiving, by the computer system, a user search request generated via the user interface, the user search request comprising an item search criteria; searching, by the computer system, the one or more electronic data stores for a plurality of items relating to the item search criteria; generating, by the computer system, for the user a personalized user interface data that provides functionality for the user to select from a result set of items based on the searching, the result set being prioritized based on the predicted conversion factors; causing, by the computer system, rendering of a search results output, based on the personalized user interface data, the search results output presents the prioritized result set and enables interaction with, by the user, the plurality of items relating to the item search criteria; electronically monitoring, by the computer system, over a computer network, user interactions with the search results output to determine a user response to the search results output, the user response comprising at least a selection of a URL associated with an item; adapting, by the computer system, the logistic regression formula based on the user response, the adapting comprising at least including an indication of the selection of the URL in the historical conversion activity analyzed to generate the logistic regression formula; regenerating, by the computer system, one or more of the predicted conversion factors by inputting the generated metadata into the adapted logistic regression formula; receiving, by the computer system, a second search request that comprises the item search criteria; and generating, by the computer system, a second result set of items based on the second search request, the second result set comprising a different prioritization based at least in part on the regenerated predicted conversion factors, wherein the computer system comprises one or more physical servers. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. A computer-implemented method of generating and presenting interactive search results, the method comprising:
-
receiving, by a computer system, via a plurality of network-accessible feeds, data relating to a plurality of items each comprising a plurality of attributes and a category; generating, by the computer system, metadata based on the data received via the plurality of network-accessible feeds, the metadata comprising Uniform Resource Locators (URL'"'"'s) for webpages at which the plurality of items are for sale, wherein the metadata further comprises stated metadata and derived metadata, the stated metadata comprising attributes obtained from item descriptions provided by a source or promoter of items relating to the item descriptions, the derived metadata comprising attributes obtained from a third party source, database-linking, in an electronic data store, the generated metadata to the plurality of items; calculating, by the computer system, predicted conversion factors for the plurality of items, the predicted conversion factors calculated at least in part by inputting the stated metadata and the derived metadata into a logistic regression formula, the logistic regression formula generated based at least in part on an analysis of historical conversion activity and stated metadata and derived metadata related to attributes associated with items different than the plurality of items to determine statistically how each attribute associated with the items different than the plurality of items affected the historical conversion activity, the logistic regression formula configured to apply a result of the analysis to the inputted metadata to output the predicted conversion factors, wherein the items different than the plurality of items comprise a same category as the plurality of items but comprise one or more different attribute values; providing, by the computer system, a user interface that comprises functionality that enables a user to interactively search for and select items; receiving, by the computer system, a user search request generated via the user interface, the user search request comprising an item search criteria; searching, by the computer system, the electronic data store for a plurality of items relating to the item search criteria; generating, by the computer system, a result set of items based on the searching, the result set being prioritized based on the predicted conversion factors; causing, by the computer system, rendering of a search results output that presents the prioritized result set and enables interaction with, by the user, the plurality of items relating to the item search criteria; electronically monitoring, by the computer system, over a computer network, user interactions with the search results output to determine a user response to the search results output, the user response comprising at least a selection of a URL associated with an item; adapting, by the computer system, the logistic regression formula based on the user response, the adapting comprising at least including an indication of the selection of the URL in the historical conversion activity analyzed to generate the logistic regression formula; regenerating, by the computer system, one or more of the predicted conversion factors by inputting the generated metadata into the adapted logistic regression formula; receiving, by the computer system, a second search request that comprises the item search criteria; and generating, by the computer system, a second result set of items based on the second search request, the second result set comprising a different prioritization based at least in part on the regenerated predicted conversion factors, wherein the computer system comprises one or more physical servers. - View Dependent Claims (9, 10, 11, 12, 13, 14, 15)
-
Specification