Method and System for Recommending Products Based on a Ranking Cocktail

• US 20170337614A1
• Filed: 08/08/2017
• Published: 11/23/2017
• Est. Priority Date: 05/11/2012
• Status: Active Grant

• Current Assignee: Fredhopper BV (CrownPeak Technology, Inc.)
• Original Assignee: Fredhopper BV (CrownPeak Technology, Inc.)
• Inventors: Costa, David, Kohn, Andreas, Penchev, Pavel, Bellis, Erica, Coutarel, Adrien
• Granted Patent: US 10,572,928 B2
• Time in Patent Office: Days
• CPC Class Codes: G06Q 30/0631 Item recommendations

Abstract

Systems and methods for e-commerce personalization and merchandising are provided herein. In some instances, methods may include determining triggers for a consumer, where the triggers being associated with objective consumer preferences and subjective consumer preferences for the consumer. Also, the method includes selecting a ranking cocktail for the consumer that includes a plurality of attributes that each includes a weight. The method also includes utilizing the ranking cocktail to select recommended products from an inventory of products in a database of a merchant, and providing the recommended products for display to the consumer.

Claims

1-20. -20. (canceled)

21. A method for generating a ranking cocktail for a product recommendation system, the method comprising:

receiving selections of a plurality of attributes from a consumer;

calculating a weight for at least one of the plurality of attributes, wherein some of the plurality of attributes are non-weighted, the weighting based on the selections;

automatically distributing unallocated weight to the non-weighted attributes such that a total weight of the ranking cocktail equals 100%; and

storing the ranking cocktail in a database.

22. The method according to claim 21, further comprising selecting recommended products from a database using the ranking cocktail, the ranking cocktail allowing for the recommended products to be selected based on the preferences of the consumer, the preferences comprising both objective and subjective preferences.

23. The method according to claim 21, further comprising normalizing the plurality of attributes to scale the weight of each of the plurality attributes in such a way that higher weighted attributes have higher weighted normalized values.

24. The method according to claim 21, wherein normalizing comprises any of monotonically increasing and monotonically decreasing the weight of each of the plurality attributes.

25. The method according to claim 21, further comprising:

presenting recommended products for display on a website that have been selected using the ranking cocktail;

receiving crowdsource feedback on the recommended products;

adjusting the weighting of at least one attribute of the ranking cocktail in response to the crowdsource feedback to alter the selection of recommended products from the inventory of products.

26. The method according to claim 21, further comprising:

selecting, by the computing device, recommended products from an inventory of products in a database of a merchant using the ranking cocktail.

27. The method according to claim 26, further comprising:

determining, by the computing device, that applying the ranking cocktail to a query to obtain recommended products would cause a latency that meets or exceeds a specified latency threshold; and

returning, by the computing device, raw product results in response to the query.

28. The method according to claim 27, further comprising providing, by the computing device, the recommended products for display to the consumer when the latency does not meet or exceed the latency threshold after the raw product results are returned.

29. The method according to claim 21, wherein the ranking cocktail comprises any of a seasonal ranking cocktail, a category-specific ranking cocktail, a consumer-specific ranking cocktail, a deal hunter ranking cocktail, and a geek ranking cocktail.

30. The method according to claim 21, wherein the ranking cocktail comprises a social recommendation ranking cocktail that includes a social recommendation attribute, a web analytics attribute, and an enterprise resource planning (ERP) attribute.

31. A method for generating a ranking cocktail for a product recommendation system, the method comprising:

determining objective and subjective preferences of a consumer;

generating a ranking cocktail by;

calculating a weight for a portion of a plurality of attributes, wherein the attributes are based on the objective and subjective preferences of the consumer, wherein each weight is normalized; and

applying the ranking cocktail to select recommended products to display to the consumer.

32. The method according to claim 31, further comprising inverting a normalization of the ranking cocktail.

33. The method according to claim 32, wherein the normalization is monotonically increasing with respect to a value a recommended product has for that attribute while inverted normalization is monotonically decreasing.

34. The method according to claim 31, further comprising automatically distributing unallocated weight to the non-weighted attributes such that a total weight of the ranking cocktail equals 100%.

35. The method according to claim 34, further comprising proportionally dividing and distributing the unallocated weight to each attribute.

36. A method for generating a ranking cocktail for a product recommendation system, the method comprising:

determining objective and subjective preferences of a consumer;

segmenting the consumer according to their objective and subjective preferences;

selecting a ranking cocktail for a segment to which the consumer belongs, wherein the ranking cocktail comprises weightings for a portion of a plurality of attributes, wherein the attributes are based on the objective and subjective preferences of the consumer and based on the segment of the consumer;

determining that one or more triggers have been met based on actions performed by the consumer;

applying the ranking cocktail to select recommended products to display to the consumer; and

selecting one or more complementary products based on the recommended products selected for the consumer.

37. The method according to claim 36, further comprising receiving crowdsourced feedback on the recommended products, the one or more complementary products and combinations thereof.

38. The method according to claim 36, wherein the objective preferences comprise behaviors selected from any of websites visited, click through data, average time spent on a webpage, search histories, shopping cart items, past purchasing data, buying preferences, buying patterns, and any combinations thereof.

39. The method according to claim 36, wherein the selection of the recommended products is further based on upon product selections made by other consumers with substantially similar preferences to that of the consumer.

40. The method according to claim 36, wherein the objective and subjective preferences comprise demographic information of the consumer, psychographic information of the consumer, as well as subjective preferences which comprise favorite brand of the consumer, favorite designer of the consumer, favorite color of the consumer, and any combinations thereof.