Item attribute based data mining system
Data mining systems and methods are disclosed for item recommendation based on frequent attribute-values associated with items. The system may determine commonalities in item attribute-values based on user transactions and identify frequent attribute-value tuples that include attribute-values that frequently co-occur in user transactions. The system may associate user interests with the frequent attribute-value tuples and recommend items to target users based thereon. A user-interface for presenting the recommendation allows users to explore item recommendations based on modifications to one or more frequent attribute-value tuples initially recommended to the user
1. A method for item recommendation based on item attribute-value tuples that are frequent to user transactions, comprising:
obtaining item acquisition data indicating a plurality of transactions associated with a set of users, wherein individual transactions include one or more items acquired by a corresponding user;
incorporating item attribute-values into the item acquisition data, wherein individual items are associated with one or more item attribute-values;
identifying a set of attribute-value tuples, wherein individual attribute-value tuples of the set of attribute-value tuples include two or more item attribute-values that co-occur in individual transactions of a subset of the plurality of transactions;
associating user interest measures with individual attribute-value tuples of the set of attribute-value tuples, wherein associating user interest measures with individual attribute-value tuples comprises generating a user interest measure for the attribute-value tuple based, at least in part, on one or more user ratings of an item corresponding to the attribute-value tuple;
identifying a first attribute-value tuple from the set of attribute-value tuples for a target user based, at least in part, on the user interest measures;
causing presentation, to the target user, of a first recommendation based, at least in part, on the first attribute-value tuple;
obtaining, from the target user, an indication of a modification to the first attribute-value tuple;
identifying items corresponding to the modified first attribute-value tuple; and
causing presentation, to the target user, of a second recommendation based, at least in part, on the items corresponding to the modified first attribute-value tuple;
the method performed programmatically by one or more computing systems under control of executable program code.
2. The method of claim 1 , wherein individual items include a product or service represented in an electronic catalog system.
3. The method of claim 1 , wherein one or more items acquired by a corresponding user correspond to one or more items purchased, rented, licensed, downloaded, installed, added to a wish list, saved, tagged, recommended, or subscribed to by the corresponding user.
4. The method of claim 1 , wherein individual item attribute-values represent a generalization or categorization of an aspect of a corresponding item.
5. The method of claim 1 , wherein at least two transactions of the subset of transactions are associated with different users.
6. The method of claim 1 , wherein identifying a first attribute-value tuple for a target user comprises identifying an attribute-value tuple associated with a user interest measure in connection with the target user.
7. The method of claim 1 , wherein the presentation of the first recommendation includes presenting a user-interface representing the first attribute-value tuple.
8. The method of claim 7 , wherein the indication of modification to the first attribute-value tuple is obtained based, at least in part, on the target user's interaction with the user-interface.
9. The method of claim 1 , wherein the modification to the first attribute-value tuple includes changes to at least one attribute-value of the attribute-value tuple.
10. A system for item recommendation based on item attribute-value tuples that are frequent to user transactions, the system comprising:
a computing system comprising one or more hardware processors, the computing system programmed with executable instructions to perform a process that comprises:
obtaining item acquisition data indicating a plurality of transactions associated with a set of users, wherein individual transactions include one or more items acquired by a corresponding user;
incorporating item attribute-values into the item acquisition data, wherein individual items are associated with one or more item attribute-values;
identifying a set of attribute-value tuples, wherein individual attribute-value tuples of the set of attribute-value tuples include two or more item attribute-values that co-occur in individual transactions of a subset of the plurality of transactions;
associating user interest measures with individual attribute-value tuples of the set of attribute-value tuples, wherein associating user interest measures with individual attribute-value tuples comprises generating a user interest measure for the attribute-value tuple based, at least in part, on one or more user ratings of an item corresponding to the attribute-value tuple;
identifying a first attribute-value tuple from the set of attribute-value tuples for a target user based, at least in part, on the user interest measures;
causing presentation, to the target user, of a first recommendation based, at least in part, on the first attribute-value tuple;
obtaining, from the target user, an indication of a modification to the first attribute-value tuple;
identifying items corresponding to the modified first attribute-value tuple; and
causing presentation, to the target user, of a second recommendation based, at least in part, on the items corresponding to the modified first attribute-value tuple.
11. The system of claim 10 , wherein individual items include a product or service represented in an electronic catalog system.
12. The system of claim 10 , wherein one or more items acquired by a corresponding user correspond to one or more items purchased, rented, licensed, downloaded, installed, added to a wish list, saved, tagged, recommended, or subscribed to by the corresponding user.
13. The system of claim 10 , wherein individual item attribute-values represent a generalization or categorization of an aspect of a corresponding item.
14. The system of claim 10 , wherein at least two transactions of the subset of transactions are associated with different users.
15. The system of claim 10 , wherein identifying a first attribute-value tuple for a target user comprises identifying an attribute-value tuple associated with a user interest measure in connection with the target user.
16. The system of claim 10 , wherein the presentation of the first recommendation includes presenting a user-interface representing the first attribute-value tuple.
17. The system of claim 16 , wherein the indication of modification to the first attribute-value tuple is obtained based, at least in part, on the target user's interaction with the user-interface.
18. The system of claim 10 , wherein the modification to the first attribute-value tuple includes changes to at least one attribute-value of the attribute-value tuple.