Method for relevancy ranking of products in online shopping

A method for determining a goodness value for a product in online shopping, the goodness value being determined independent of a keyword search query, the method comprising the steps of:

• a) identifying, one or more attributes related to the product based on a product category associated with the product, the one or more attributes comprising at least one of a brand name, a store name, a style, and price;
• b) identifying one or more features associated with each of the corresponding one or more attributes;
• c) computing a normalized score for each of an attribute instance of the one or more attributes for each of the product in the product category based on a value associated with each of the one or more features and the product category, the value of the one or more features being associated with market demand-supply characteristics for the product, wherein the normalized score for each of the attribute instance is computed independent of an user input, the normalized score being computed using one or more mathematical functions, wherein computing the normalized score for each of an attribute instance of the one or more attributes comprises:
• i) selecting one of the one or more identified attributes as a base attribute based on the product category,
• ii) computing a normalized score of each of an attribute instance of the base attribute for each product in the product category, and
• iii) computing normalized scores of the attribute instances of the attributes other than the base attribute based, at least in part, on an attribute affinity matrix, wherein the attribute affinity matrix comprises one or more values, for the attribute instances of the attributes other than the base attribute, calculated based on the normalized scores of the attribute instances of the base attribute;
• d) combining the scores of each attribute instance of the one or more attributes by using combinational functions to generate the goodness value for the product, wherein the goodness value is an indication of market and bulk user preferences;
• and
• e) storing the goodness value in a database;
• wherein one or more products in response to a key word search query are retrieved from the database based on the corresponding goodness value, the steps of the method (a-e) being performed by a microprocessor and independent of the search terms in the key word search query.

A system for ranking one or more products in online shopping, the system comprising:

• a) a microprocessor;
• b) a query manager, the query manager configured for interacting with the microprocessor to receive a keyword search query for the one or more products, the one or more products being associated with a product category;
• c) a feature extraction module, the feature extraction module configured for interacting with the microprocessor to identify one or more attributes related to the product category and to identify one or more features corresponding to each of the one or more attributes for each of the one or more products based on receiving the keyword search query, the one or more attributes comprising at least one of a brand name, a store name, a style, and price, the one or more features comprising query independent features and query dependent features;
• d) a computing module, the computing module configured for interacting with the microprocessor to compute:
• (i) a normalized score of each of an attribute instance of a base attribute for each product in the product category, the normalized score being computed based on the value associated with each of the one or more features associated with the base attribute, wherein the value of the one or more features is determined based on a query dependent and a query independent analysis of the one or more features, the normalized score being computed using one or more mathematical functions, wherein the base attribute is selected from the one or more attributes based on the product category;
• and
• (ii) a normalized score of each of an attribute instance of the one or more attributes other than the base attribute for each product in the product category, the normalized score being computed based at least in part on the normalized score of the corresponding base attribute instance of each product, the normalized score being computed using one or more mathematical functions and an attribute affinity matrix, wherein the attribute affinity matrix comprises one or more values, for the attribute instance of the one or more attributes other than the base attribute, calculated based on normalized scores of the attribute instances of the base attribute;
• e) a weight assignment module, the weight assignment module configured for interacting with the microprocessor to assign weights to each of the one or more attributes based on the relative importance of each of the one or more attributes for the product category and the keyword search query;
• and
• f) a product ranking module, the product ranking module configured for interacting with the microprocessor to generate a product rank for each of the one or more products in the product category based on the assigned weights of the one or more attributes and the scores of each attribute instance using a predefined relevancy-ranking algorithm, wherein the rank ordered list of the one or more products are displayed.

A computer program product for use with a computer, the computer program product comprising a non-transitory computer usable medium having a computer readable program code embodied therein for ranking of one or more products in online shopping, the computer readable program code performing the steps of:

• a) receiving a keyword search query for the one or more products, the one or more products being associated with a product category;
• b) identifying one or more attributes, based on receiving the keyword search query, related to the product category, the one or more attributes comprising at least one of a brand name, a store name, a style, and a price;
• c) selecting one of the one or more identified attributes as a base attribute, wherein the base attribute is selected based on the product category;
• d) identifying one or more features corresponding to the base attribute, the one or more features comprising query independent features and query dependent features;
• e) computing a normalized score of each of an attribute instance of the base attribute for each product in the product category, the normalized score being computed based on the value associated with each of the one or more features associated with the base attribute, wherein the value of the one or more features is determined based on a query-dependent and a query-independent analysis of the one or more features, the normalized score being computed using one or more mathematical functions;
• f) computing a normalized score of each of an attribute instance of the one or more attributes other than the base attribute for each product in the product category, the normalized score being computed based at least in part on the normalized score of the corresponding base attribute instance of each product, the normalized score being computed using one or more mathematical functions and an attribute affinity matrix, wherein the attribute affinity matrix comprises one or more values, for the attribute instance of the one or more attributes other than the base attribute, calculated based on normalized scores of the attribute instances of the base attribute;
• g) assigning weights to each of the one or more attributes based on the relative importance of each of the one or more attributes for the product category and the keyword search query;
• h) generating a product rank for each of the one or more products in the product category based on the assigned weights of the one or more attributes and the scores of each of the attribute instance using a predefined relevancy-ranking algorithm;
• and
• i) displaying a rank ordered list of the one or more products, wherein the computer program product interacts with a microprocessor for executing the computer readable program code.

A computer program product for use with a computer, the computer program product comprising a non-transitory computer usable medium having a computer readable program code embodied therein for determining a goodness value for a product in online shopping, the goodness value being determined independent of a keyword search query, the computer readable program code performing the steps of:

• a) identifying one or more attributes related to the product based on a product category associated with the product, the one or more attributes comprising at least one of a brand name, a store name, a style, and price;
• b) identifying, one or more features associated with each of the corresponding one or more attributes;
• c) computing a normalized score for each of an attribute instance of the one or more attributes for each of the product in the product category based on a value associated with each of the one or more and the product category, the value of the one or more features being associated with market demand-supply characteristics for the product, wherein the normalized score for each of the attribute instance is computed independent of an user input, the normalized score being computed using one or more mathematical functions, wherein computing the normalized score for each of an attribute instance of the one or more attributes comprises:
• i) selecting one of the one or more identified attributes as a base attribute based on the product category,
• ii) computing a normalized score of each of an attribute instance of the base attribute for each product in the product category, and
• iii) computing normalized scores of the attribute instances of the attributes other than the base attribute based, at least in part, on an attribute affinity matrix, wherein the attribute affinity matrix comprises one or more values, for the attribute instances of the attributes other than the base attribute, calculated based on the normalized scores of the attribute instances of the base attribute;
• d) combining the scores of each attribute instance of the one or more attributes by using combinational functions to generate the goodness value for the product, wherein the goodness value is an indication of market and bulk user preferences;
• and
• e) storing the goodness value in a database, wherein one or more products in response to a keyword search query are retrieved from the database based on the corresponding goodness value, the steps of the method (a-e) being performed independent of the search terms in the key word search query, wherein the computer program product interacts with a microprocessor for executing the computer readable program code.