초록 ▼

A system for determining selections that a user is likely to be interested in. The determination is made based on the user's prior indicated preferences. The user designates his or her preferred selections as entries in a user's preference list. Entries in the user's list are compared with entries i

A system for determining selections that a user is likely to be interested in. The determination is made based on the user's prior indicated preferences. The user designates his or her preferred selections as entries in a user's preference list. Entries in the user's list are compared with entries in the other users' lists. When a significant number of matches have been found between two lists, the unmatched entries of the other user's preference list are extracted. The unmatched entries are further processed. Those unmatched entries with a high correlation to the user's preference list are presented to the user as selections in which the user is likely to be interested.

대표청구항 ▼

What is claimed is: 1. A method to be executed on a computer system, wherein the computer system includes a processor and a database, and wherein the database includes a plurality of datafiles each containing a plurality of predetermined preferences, the method comprising: receiving signals that in

What is claimed is: 1. A method to be executed on a computer system, wherein the computer system includes a processor and a database, and wherein the database includes a plurality of datafiles each containing a plurality of predetermined preferences, the method comprising: receiving signals that indicate a plurality of user preferences; comparing at least a subset of the indicated user preferences against the plurality of datafiles in the database to identify matching datafiles containing preferences matching at least a first and a second threshold number of the at least a subset of the indicated user preferences, the second threshold number being smaller than the first threshold number; selecting preferences from the identified matching datafiles, wherein the selected preferences do not match the at least a subset of the indicated user preferences; and outputting the selected preferences. 2. The method of claim 1, wherein the preferences comprise at least either artists' names or movie titles. 3. The method of claim 1, wherein the method further comprises compiling a list of preferences not matching the at least a subset of the indicated user preferences, based at least in part on a subset of the identified matching datafiles containing preferences matching at least the first threshold number of the at least a subset of the indicated user preferences, and determining for each non-matching preference of the list, a frequency of occurrences in the identified matching datafiles, and the selecting comprises selecting preferences based at least in part on the list, in view of the list's preferences' determined frequencies of occurrences in the identified matching datafiles. 4. The method of claim 3, wherein the method further comprises determining for each not matching preference of the list, a frequency of occurrences of the not matching preference in the database, and said selecting is further based on the list's preferences' frequencies of occurrences in other datafiles of the database. 5. The method of claim 1, wherein the selecting is further based on frequencies of occurrences of the not matching preferences in at least the identified matching datafiles. 6. The method of claim 1, wherein the method further comprises compiling at least a first and a second list of preferences not matching the at least a subset of the indicated user preferences, with the first and second lists of preferences being compiled based at least in part on the identified matching datafiles containing preferences matching at least the first and second threshold numbers of the at least a subset of the user preferences, respectively, and the selecting is performed based at least in part on the at least first and second lists. 7. The method of claim 6, wherein the preferences of each of the at least first and second lists are ranked relative to other preferences of the list, and the selecting is based at least in part on weighted ranks of the preferences, the weighted rank of a preference being determined based at least in part on the relative ranks of the preferences in the at least first and second lists. 8. A method for recommending music selections based on a user's preferred music selections, the method comprising: storing a plurality of associated music selections in a database; receiving signals from an input device that indicate a plurality of a user's preferred music selections; identifying a first and a second number of the plurality of associated music selections as being similar to the preferred music selections, based at least in part on a first and a second degree of similarity, the first and second degrees of similarity being different degrees of similarity, the similar associated music selections and the preferred music selections being different music selections; selecting a number of the first and second number of similar associated music selections; and outputting, via an output device, the selected associated music selections. 9. An apparatus for recommending music selections based on a user's preferred music selections, the apparatus comprising: a computer system including a database; means for storing a plurality of associated music selections in the database; means for accepting signals from a user input device to indicate a plurality of a user's preferred music selections; means for identifying a first and a second number of the associated music selections as being similar to the preferred music selections, based at least in part on a first and a second degree of similarity, respectively, the first and second degrees of similarity being different degrees of similarity, the similar associated music selections and the preferred music selections being different music selections; and means for selecting a number of the first and second similar associated music selections. 10. A method for outputting an ordered list of recommended objects based on an input object, the method comprising: storing pairs of ranked objects in a database; assigning a ranking number to each of the pairs of ranked objects and storing the ranking number in association with the pair; receiving signals from an input device that indicate an object; using a processor to find occurrences of the indicated object in the pairs of ranked objects; for each pair in which the indicated object occurs, determining a non-matching object in the pair that does not match the indicated object; ordering non-matching objects into a list according to the ranking number for the pair that the non-matching object belongs to; and outputting the list as an ordered list of recommended objects. 11. The method of claim 10, wherein each object comprises an artist's name. 12. The method of claim 10, wherein each object comprises a title of a movie. 13. An apparatus for outputting an ordered list of recommended objects based on an input object, the apparatus comprising: a processor; a database accessible to the processor; pairing means for storing pairs of ranked objects in the database; ranking means for assigning a ranking number to each of the pairs of ranked objects and storing a given ranking number in association with a given stored pair; input means for receiving an indication from a human user that indicates a selected object; finding means for finding occurrences of the selected object in the pairs of ranked objects; determining means for determining, for each pair, an object in the pair that does not match the selected object; ordering means for ordering non-matching objects into a list according to the ranking number for the pair that the non-matching object belongs to; and output means for outputting the list as an ordered list of recommended objects. 14. In a multi-user computer system that provides user access to a database of objects, a method of recommending objects to a user computer, the method comprising: identifying on a remote computer, a first set of objects of interest to a user; using a processor to access one or more data structures representing a plurality of objects of interest to a community, to identify at least a first and a second set of objects having respectively at least a first and a second threshold of similarities in common with the first set of objects, the first and second thresholds being different thresholds; generating a combined set of objects from the identified at least first and second sets of objects; and transmitting to the user computer, the combined set of objects. 15. The method of claim 14, wherein the generating is further based on frequencies of occurrences of the objects of the combined set represented in the data structures having the at least first and second thresholds of similarities. 16. The method of claim 14, wherein the objects comprise a plurality of digital audio selections. 17. The method of claim 14, wherein the generating is further based on frequencies of occurrences of the objects of the combined set represented in the data structures. 18. The method of claim 14, wherein the objects of each of the at least first and second set are ranked relative to other objects of the list, and the selecting is based at least in part on weighted ranks of the objects, the weighted rank of an object being determined based at least in part on the relative ranks of the objects in the at least first and second lists. 19. A machine readable medium having stored thereon machine executable instructions, which when executed operate to implement a method comprising: identifying a first set of objects determined to be of interest to a first user; accessing one or more data structures representing a plurality of objects of interest to a community, to identify at least a first and a second set of objects having respectively at least a first and second threshold of similarities in common with the first set of objects, the first and second thresholds being different thresholds; generating a combined set of objects from the identified at least first and second sets of objects; and displaying the combined set of objects. 20. The machine readable medium of claim 19, wherein the generating is further based on frequencies of occurrences of the objects of the combined set represented in the data structures having the at least first and second thresholds of similarities. 21. The machine readable medium of claim 19, wherein said objects comprise a plurality of digital audio selections. 22. The machine readable medium of claim 19, wherein the generating is further based on frequencies of occurrences of the object of the combined set represented in the data structure having the at least first and second thresholds of similarities. 23. The machine readable medium of claim 19, wherein the object of each of the at least first and second set are ranked relative to other object of the list, and the selecting is based at least in part on weighted ranks of the objects, the weighted rank of an object being determined based at least in part on the relative ranks of the objects in the at least first and second lists. 24. The method of claim 8, wherein said identifying comprises identifying a first and a second number of digital audio music titles stored in the database, as being similar to a number of preferred digital audio music titles, based at least in part on a first and a second degree of similarity. 25. The apparatus of claim 9, further comprising means for outputting the selected associated music selections. 26. The apparatus of claim 13, wherein the recommended objects comprise artists' names. 27. In a multi-user computer system that provides user access to a database of objects, a method of recommending objects to a user, the method comprising: generating on a remote computer, a data structure which stores groupings of objects known to be of interest to a community of users; identifying a first set of objects that are known to be of interest to a first user; using a processor to access the data structure to identify at least a first and a second set of objects having respectively at least a first and a second threshold of similarities in common with the first set of objects, the first and second thresholds being different thresholds; and providing to a user computer, at least a subset of the at least first and second sets of objects. 28. The method of claim 27, wherein the providing is based at least in part on frequencies of occurrences of the first and second sets of objects in at least the data. 29. The method of claim 27, wherein the objects comprise a plurality of digital audio selections. 30. A program product comprising a plurality of programming instructions adapted to program an apparatus to enable the apparatus to practice the method of claim 1. 31. A program product comprising a plurality of programming instructions adapted to program an apparatus to enable the apparatus to practice the method of claim 8. 32. A program product comprising a plurality of programming instructions adapted to program an apparatus to enable the apparatus to practice the method of claim 14. 33. A program product comprising a plurality of programming instructions adapted to program an apparatus to enable the apparatus to practice the method of claim 27.