Using web-mining to enrich directory service databases and soliciting service subscriptions
A system and method are provided for augmenting information on business directory databases and communicating with businesses is disclosed. Using the enriched business directory database and Web mining technology, customized email message are sent inviting businesses to enter their enriched business information into the directory or even subscribe to other paid services provided by the directory service.
1. A method for enriching a directory service database by soliciting additional business information from businesses listed in the database, the method comprising:
by a processor in a business directory service provider server, serving a business listing on a business directory Website, the business listing containing basic information about a listed business;
by a processor, determining a URL of the listed business by conducting an Internet search using a name of the listed business as a search term;
by a processor, classifying the listed business in a category by processing data located at the URL in a pre-trained classifier model;
by a processor, mining the data located at the URL for the email address of the listed business; and
by a processor, transmitting from the business directory service provider server to the email address of the listed business mined from the data located at the URL, an email message soliciting, from the listed business, additional business information including hours of operation to enrich the basic information about the listed business, the message being tailored to the category of the listed business.
2. The method of claim 1 , wherein the determining a URL of a listed business comprises:
receiving results of the Internet search including a plurality of URLs; and
selecting the URL from the plurality of URLs.
3. The method of claim 2 , wherein the selecting the URL from a plurality of URLs further comprises:
determining confidence measures for the URLs, each confidence measure indicating a probability that the URL is sponsored by the business; and
selecting the URL based on the confidence measures.
4. The method of claim 1 , wherein the determining a URL of a listed business comprises:
obtaining the URL from an on line domain name management company.
5. The method of claim 1 , wherein the determining a URL of a listed business comprises
crawling the Web to determine whether a Web page at the URL belongs to the business.
6. The method of claim 1 , wherein the pre-trained classifier model is a statistical/rules based classifier selected from the group consisting of a naive Bayes classifier, an AdaBoost algorithm and a support vector machine.
7. The method of claim 1 , wherein the mining for an email
address of the listed business further comprises:
searching the data located at the URL for a link containing a term selected from the list consisting of “contact” and “location”; and
extracting the email address from text associated from a link found in the searching.
8. The method of claim 1 , further comprising:
validating the mined email address by checking whether a domain name in the email is a domain name in the URL.
9. The method of claim 1 , further comprising:
validating the mined email address by checking whether at least a portion of the business name is contained in a user part of an email address hosted by an email hosting company.
10. The method of claim 1 , wherein soliciting the listed business by delivering a message to the email address further comprises inviting the business to subscribe to additional paid services offered by the business directory service.
11. A tangible computer-readable medium having computer readable instructions stored thereon for execution by a processor to perform a method for enriching a directory service database by soliciting additional business information from businesses listed in the database, the method comprising:
in a business directory service provider server, serving a business listing on a business directory Website, the business listing containing basic information about a listed business;
in the business directory service provider server, determining a URL of a listed business by conducting an Internet search using a name of the listed business as a search term;
classifying the listed business in a category by processing data located at the URL in a pre-trained classifier model;
mining the data located at the URL for the email address of the listed business; and
transmitting from the business directory service provider server to the email address of the listed business an email message mined from the data located at the URL, soliciting, from the listed business, additional business information including hours of operation to enrich the basic information about the listed business, the message being tailored to the category of the listed business.
12. The tangible computer-readable medium of claim 11 , wherein the determining a URL of a listed business comprises:
receiving results of the Internet search including a plurality of URLs; and
selecting the URL from the plurality of URLs.
13. The tangible computer-readable medium of claim 12 , wherein the selecting the URL from a plurality of URLs further comprises:
determining confidence measures for the URLs, each confidence measure indicating a probability that the URL is sponsored by the business; and
selecting the URL based on the confidence measures.
14. The tangible computer-readable medium of claim 11 , wherein the determining a URL of a listed business comprises:
obtaining the URL from an on line domain name management company.
15. The tangible computer-readable medium of claim 11 , wherein the determining a URL of a listed business comprises
crawling the Web to determine whether a Web page at the URL belongs to the business.
16. The tangible computer-readable medium of claim 11 , wherein the pre-trained classifier model is a statistical/rules based classifier selected from the group consisting of a naive Bayes classifier, an AdaBoost algorithm and a support vector machine.
17. The tangible computer-readable medium of claim 11 , wherein the mining for an email address of the listed business further comprises:
searching the data located at the URL for a link containing a term selected from the list consisting of “contact” and “location”; and
extracting the email address from text associated from a link found in the searching.
18. The tangible computer-readable medium of claim 11 , wherein the method further comprises:
validating the mined email address by checking whether a domain name in the email is a domain name in the URL.
19. The tangible computer-readable medium of claim 11 , wherein the method further comprises:
validating the mined email address by checking whether at least a portion of the business name is contained in a user part of an email address hosted by an email hosting company.
20. The tangible computer-readable medium of claim 11 , wherein the soliciting the listed business by delivering a message to the email address further comprises inviting the business to subscribe to additional paid services offered by the business directory service.