Automatic detection of online commercial intention

• US 20070136457A1
• Filed: 12/14/2005
• Published: 06/14/2007
• Est. Priority Date: 12/14/2005
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Nie, Zaiqing, Li, Ying, Wen, Ji-Rong, Wang, Lee, Zhao, Lingzhi, Dai, Honghua
• Granted Patent: US 7,831,685 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Features extracted from network browser pages and/or network search queries are leveraged to facilitate in detecting a user'"'"'s browsing and/or searching intent. Machine learning classifiers constructed from these features automatically detect a user'"'"'s online commercial intention (OCI). A user'"'"'s intention can be commercial or non-commercial, with commercial intentions being informational or transactional. In one instance, an OCI ranking mechanism is employed with a search engine to facilitate in providing search results that are ranked according to a user'"'"'s intention. This also provides a means to match purchasing advertisements with potential customers who are more than likely ready to make a purchase (transactional stage). Additionally, informational advertisements can be matched to users who are researching a potential purchase (informational stage).

Claims

1. A system that facilitates determination of intent, comprising:

a component that receives network associated information comprising, at least in part, search information and/or network browsing information; and

a classifier component that determines intent of the network associated information and provides an online commercial intention type for the search information and/or network browsing information.

2. The system of claim 1, the classifier component employs an intent model trained from extracted features and/or content of the network browsing information.

3. The system of claim 1, the classifier component employs an intent model trained from a search result page and/or a top N list of search results and combine coefficients, where N is an integer from one to infinity.

4. The system of claim 1, the online commercial intention type comprising a non-commercial intent, an informational intent, or a transactional intent.

5. The system of claim 1, the classifier component utilizes a network browser information classifier to facilitate in classifying at least one search query.

6. The system of claim 1, the classifier component employs machine learning to facilitate in determining intent.

7. The system of claim 1, the network browsing information comprising at least one web page and/or at least one search result page.

8. A search engine that employs the system of claim 1 to facilitate in ranking search results based on search intent.

9. An online advertisement system that employs the system of claim 1 to facilitate in determining pricing of online advertisements and/or associating advertisements to network associated information.

20. A device employing the system of claim 1 comprising at least one selected from the group consisting of a computer, a server, and a handheld electronic device.

10. A method for facilitating determining intent, comprising:

obtaining browsing training information from a network to facilitate model training;

extracting intent differentiating features and/or content from the network browsing training information;

applying a machine learning process to train an intent model based on the extracted features and/or content; and

constructing a network browsing information intent classifier based on the intent model.

11. The method of claim 10 further comprising:

constructing the network browsing information intent classifier to include, at least in part, an intent classifier that facilitates in processing a structure characteristic of a search result page.

12. The method of claim 10 further comprising:

obtaining browsing information from a network for intent classifying;

extracting at least one intent feature from the network browsing information; and

employing the network browsing information intent classifier to determine the intent of the network browsing information.

13. The method of claim 12 further comprising:

automatically classifying web pages into an online commercial intention (OCI) type.

14. The method of claim 10 further comprising:

obtaining intent labeled query information to facilitate in search intent model training;

processing the intent labeled query information with a search engine to determine a search result page and/or a list of top N query results, where N is an integer from one to infinity;

determining an intent of browsing information associated with the search result page and/or each top N query result via employment of the network browsing information intent classifier; and

combining the intentions from the search result page and/or the top N query results utilizing, at least in part, combine coefficients to construct a search intention classifier.

15. The method of claim 14 further comprising:

obtaining query information for intent classifying;

processing the query information with a search engine to determine a search result page and/or a top P query result list, where P is an integer from one to infinity;

determining an intent of network browsing information associated with the search result page and/or each top P query result via employment of the network browsing information intent classifier; and

determining a search intention by combining the intentions from the search result page and/or the top P query results utilizing, at least in part, the combine coefficients.

16. The method of claim 15 further comprising:

employing a machine learning algorithm to facilitate in optimizing performance in detecting intent of search queries.

17. The method of claim 15 further comprising:

automatically classifying the query information into an online commercial intention (OCI) type.

18. An advertising sales method employing, at least in part, the method of claim 15.

19. A system that facilitates determination of intent, comprising:

means for receiving network associated information comprising, at least in part, search information and/or network browsing information;

means for determining an intent of the network associated information; and

means for providing an online commercial intention type for the search information and/or network browsing information.