Method and system for listing categorization

• US 8,010,410 B2
• Filed: 12/29/2006
• Issued: 08/30/2011
• Est. Priority Date: 12/29/2006
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Zinchenko, Vitaliy, Sundaresan, Neelakantan
• Primary Examiner(s): Airapetian; Mila
• Application Number:
  US11/618,701
• Publication Number:
  US 20080162302A1
• Time in Patent Office: 1,705 Days
• Field of Search: 705/26, 705/27
• US Class Current: 705/26.1
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0603 Catalogue ordering
• CPC Class Codes: G06Q 30/0633 Lists, e.g. purchase orders, compilation or processing

Abstract

Embodiments of a method and system for listing categorization are disclosed. A category structure may be accessed. The category structure may include a plurality of categories for items. A set of training data may be accessed from a plurality of listings from at least one of supply data and/or demand data. The supply data may be generated from seller activity of a plurality of users in a networked system. The demand data may be generated from buyer activity of the plurality of users in the networked system. Each listing may include a category from the category structure. The set of training data may be provided to a categorization application for training. The categorization application may be capable of building listing statistics by applying a classifier to the set of training data and recommending a category from the category structure for a new listing by utilizing the listing statistics.

Claims

1. A method comprising:

accessing a category structure, the category structure including a plurality of categories for items;

accessing a set of training data from a plurality of listings from at least one of supply data or demand data, the supply data generated from seller activity of a plurality of users in a networked system, the demand data generated from buyer activity of the plurality of users in the networked system, each listing including a category from the category structure; and

providing the set of training data to a categorization application for training, the categorization application capable of building listing statistics by applying a classifier to the set of training data and recommending a category from the category structure for a new listing by utilizing the listing statistics.

2. The method of claim 1, wherein the classifier is a Bayesian classifier.

3. The method of claim 1, wherein the buyer activity is at least one of:

at least one buyer from the plurality of users browsed to a category from the plurality of listings within the networked system,

the at least one buyer searched for a listing of an item for sale,

the at least one buyer selected the listing of the item for sale,

the at least one buyer bid on the item for sale, and

the at least one buyer bought the item for sale.

4. The method of claim 1, wherein the seller activity is at least one of:

at least one seller from the plurality of users created a listing for an item within the networked system,

the at least one seller received a bid for the item, and

the at least one seller made a sale of the item.

5. The method of claim 1, wherein accessing a set of training data from a plurality of listings from at least one of supply data or demand data comprises:

accessing listing data from a plurality of listings from at least one of supply data or demand data, the supply data generated from seller activity of a plurality of users in a networked system, the demand data generated from buyer activity of the plurality of users in the networked system, each listing including a category from the category structure; and

automatically generating a set of training data for the category structure by applying a classifier to the listing data.

6. The method of claim 5, wherein automatically generating a set of training data comprises:

automatically generating a set of training data by applying a classifier to a title of the plurality of listings of the listing data.

7. The method of claim 5, wherein automatically generating a set of training data comprises:

automatically generating a set of training data by applying a classifier to at least one of a title and title relevant information within a description of the plurality of listings.

8. A machine-readable non-transitory storage medium comprising instructions, which when executed by a machine, cause the machine to:

access a category structure, the category structure including a plurality of categories for items;

access a set of training data from a plurality of listings from at least one of supply data or demand data, the supply data generated from seller activity of a plurality of users in a networked system, the demand data generated from buyer activity of the plurality of users in the networked system, each listing including a category from the category structure; and

provide the set of training data to a categorization application for training, the categorization application capable of building listing statistics by applying a classifier to the set of training data and recommending a category from the category structure for a new listing by utilizing the listing statistics.

9. The machine-readable medium non-transitory storage of claim 8, wherein causing the machine to access a set of training data from a plurality of listings from at least one of supply data or demand data causes the machine to:

access listing data from a plurality of listings from at least one of supply data or demand data, the supply data generated from seller activity of a plurality of users in a networked system, the demand data generated from buyer activity of the plurality of users in the networked system, each listing including a category from the category structure; and

automatically generate a set of training data for the category structure by applying a classifier to the listing data.

10. The machine-readable non-transitory storage medium of claim 8, wherein the classifier is a Bayesian classifier.

11. The machine-readable non-transitory storage medium of claim 8, wherein the buyer activity is at least one of:

at least one buyer from the plurality of users browsed to a category from the plurality of listings within the networked system,

the at least one buyer searched for a listing of an item for sale,

the at least one buyer selected the listing of the item for sale,

the at least one buyer bid on the item for sale, and

the at least one buyer bought the item for sale.

12. The machine-readable non-transitory storage medium of claim 8, wherein the seller activity is at least one of:

at least one seller from the plurality of users created a listing for an item within the networked system,

the at least one seller received a bid for the item, and

the at least one seller made a sale of the item.

13. The machine-readable non-transitory storage medium of claim 9, wherein automatically generating a set of training data comprises:

automatically generating a set of training data by applying a classifier to a title of the plurality of listings of the listing data.

14. The machine-readable non-transitory storage medium of claim 9, wherein automatically generating a set of training data comprises:

automatically generating a set of training data by applying a classifier to at least one of a title and title relevant information within a description of the plurality of listings.

15. A system comprising:

a memory;

at least one processor coupled to the memory;

a first module to access, using the at least one processor, a category structure, the category structure including a plurality of categories for items;

a second module to access, using the at least one processor, a set of training data from a plurality of listings from at least one of supply data or demand data, the supply data generated from seller activity of a plurality of users in a networked system, the demand data generated from buyer activity of the plurality of users in the networked system, each listing including a category from the category structure; and

a third module to provide the set of training data to a categorization application for training, the categorization application capable of building listing statistics by applying a classifier to the set of training data and recommending a category from the category structure for a new listing by utilizing the listing statistics.

16. The system of claim 15, wherein the classifier is a Bayesian classifier.

17. The system of claim 15, wherein the buyer activity is at least one of:

at least one buyer from the plurality of users browsed to a category from the plurality of listings within the networked system,

the at least one buyer searched for a listing of an item for sale,

the at least one buyer selected the listing of the item for sale,

the at least one buyer bid on the item for sale, and

the at least one buyer bought the item for sale.

18. The system of claim 15, wherein the seller activity is at least one of:

at least one seller from the plurality of users created a listing for an item within the networked system,

the at least one seller received a bid for the item, and

the at least one seller made a sale of the item.

19. The system of claim 15, wherein the second module is to:

access listing data from the plurality of listings from the at least one of supply data or demand data, the supply data generated from seller activity of a plurality of users in a networked system, the demand data generated from buyer activity of the plurality of users in the networked system, each listing including a category from the category structure; and

automatically generate the set of training data for the category structure by applying a classifier to the listing data.

20. The system of claim 19, wherein the second module is to automatically generate the set of training data by applying a classifier to a title of the plurality of listings of the listing data.

21. The system of claim 19, wherein the second module is to automatically generate the set of training data by applying a classifier to at least one of a title and title relevant information within a description of the plurality of listings.