Electrical computers and digital processing systems: multicomput – Computer-to-computer protocol implementing – Computer-to-computer data transfer regulating
Reexamination Certificate
1998-12-23
2002-06-25
Wiley, David (Department: 2155)
Electrical computers and digital processing systems: multicomput
Computer-to-computer protocol implementing
Computer-to-computer data transfer regulating
C710S062000
Reexamination Certificate
active
06412012
ABSTRACT:
BACKGROUND
The present invention relates generally to data processing systems, and more particularly, collaborative filtering and recommender systems.
Recommender systems predict the preferences of users based on attributes known about the user or a past history of preferences or consumption by the user. For example, a recommender system may predict that a user will like the movie “Titanic” because he previously indicated a liking for such other epic movies as “Lawrence of Arabia” or “Ben Hur”.
Present recommenders focus on making accurate recommendations of user preference. However, the most accurate recommendations, in isolation, may be the wrong items when considered collectively as a set of recommendations and as recommendations designed to accompany products that the user is buying or has bought. For example, a system recommending books may discover that a particular customer would enjoy Tolkien's
The Hobbit,
based on other books that the user did and did not like.
The Hobbit,
however, appears in more than thirty editions, from hardcover to paperback, audiobook to videocassette, editions bundled with other books, and editions illustrated by different illustrators. If the recommender system were to suggest that the customer consider several of these, it would likely annoy the user and waste the opportunity to recommend a more diverse set of books. As important, if we know that the customer already owns one edition of
The Hobbit,
we likely would prefer to avoid recommending another edition, so as to add greater value to the bookstore's recommendations. Similar examples also occur where having an item makes recommending a different item more valuable. For example, a student who has a copy of a particular textbook in her shopping cart would be well-served by a recommendation for the study guide that accompanies that textbook, even if her prior purchases do not independently indicate that the study guide is an item she would otherwise be likely to buy.
A recommender system determines its recommendations by examining previous user preference data. The preference data can be unary or numerically valued. Unary preference data is a set of customer-item pairs: a customer-item pair indicates that an event linking the customer to the item has occurred. No additional preference information is available to the recommender system about the a user-item event except that it happened. The non-existence of a customer-item pair (more generally known as a tuple) for a specific customer-item pair does not indicate a preference: it only indicates a lack of information. An example of unary customer data is purchase record data where a customer-item pair indicates that the customer has purchased the indicated item. Another example of unary data is contained in web page logs, where a customer-item pair indicates that the customer has visited a specific web page.
Binary and numerically valued preference data are generally in the form of a 3-tuples, where the three elements of the tuple are customer identifier, item identifier, and preference value. The preference value indicates, for example, the strength of the user's preference for the item or whether the user's preference is either for or against the item. To illustrate, where the preference is represented in binary form, a “0” may mean a preference against an item while a “1” means a preference for the item. Where the preference is presented as numerically valued data, the data value may represent a one-dimensional axis of preference, with the midpoint indicating an ambivalent preference for the item, a low value indicating a strong dislike for the item, and a high value indicating a strong preference for the item.
Preference data may be presented to the recommender system in explicit or implicit form. Explicit preference data are preference values that a user has supplied directly, for example by filling out a survey. Implicit preference data consist of preference values that have been inferred by observing actions that the user has taken. It can be inferred that the user has some preference for the item that she has just bought, although the act of purchasing the item is not an explicit statement of preference per se. A user's preference for a web page may be inferred, for example, by measuring the amount of time that the user spends reading the web page, or the number of times the user returns to that page.
The inputs to a recommender system are typically preference values as described above. The outputs of the recommender system are predictions of preference values for items, particularly those for which the user has not already indicated a preference. Like the input values, the output preferences may be unary, binary, or numerically valued. A system that outputs unary recommendations predicts items that will be of interest to the user, but does not attempt to predict the strength of a user's preference for each item. Binary predictions indicate items that are likely to be of high preference to the user and items that are likely to be of low preference, but again cannot provide an estimate of preference strength. Numerically valued preferences indicate a preference for or against the item and also indicate the preference strength. Note that the domain of the preference input may be different from the domain of the output preference predictions. For example, the preference input may be unary, while the output preference predictions may be numerically valued.
While unary and binary preference values do not indicate the strength of the preference, some recommender systems may additionally rank the preference predictions being returned such that the highest rank predictions have the largest probability of being correct. Numerically valued items are implicitly ranked.
Existing recommender systems generate recommendations by selecting the highest-ranking positive preference values. However, this technique does not always provide a desirable effect. In some cases, there may be strong correlative effects between a current or past purchase, and a recommendation that otherwise would have a low ranking. For example, the purchase of 35 mm film may rank low on a recommendation list, or may not even be on the recommendation list, given the current contents of a shopping basket. However, if there is knowledge in the recommender system that the user has previously purchased a 35 mm camera, then it becomes more sensible to recommend 35 mm film to the user.
There may also be strong anti-correlative effects that should be taken into account to remove a recommendation from a recommendation list. For example, if a user has purchased a pizza and a bottle of root beer made by one manufacturer, then, there is little value in recommending that the user by root beer made by another manufacturer.
Therefore, there exists a problem with existing recommender systems that, although able to recommend items with high confidence level, the recommender system is unable determine the quality of recommendations in view of other items present in the shopping basket, the recommendation set or in an historical set of past purchases. Consequently, the value of some of the recommendations made to a user may be low. There is a need to reduce the frequency of occurrence of low value recommendations.
SUMMARY OF THE INVENTION
To address the problems listed above, the present invention is directed to generating compatibility-aware recommendations for the user. In particular, the invention is directed to an electronic processing system for generating a compatibility-aware recommendation output set to a user based, at least in part, on a set of item compatibility rules. The system includes a processing system of one or more processors configured to receive applicable data, including i) user preference data, and ii) item compatibility rules, and to produce a compatibility-aware recommendation output set using the user preference data and the item compatibility rules.
In another embodiment, the invention is directed to a method of producing a compatibili
Bieganski Paul
Frankowski Dan
Konstan Joseph
Rauser John
Net Perceptions, Inc.
Wiley David
LandOfFree
System, method, and article of manufacture for making a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System, method, and article of manufacture for making a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System, method, and article of manufacture for making a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2925938