CAPTURING CANDIDATE PROFILES
Some embodiments of the invention provide novel methods for capturing the profiles of employment candidates. To start the process for capturing such profiles, some embodiments provide a thin browser agent that can be executed within a browser when the browser is displaying an employment candidates profile on a social media webpage. After selection of the browser agent (e.g., a bookmarklet or a browser extension), the browser agent in some embodiments sends to a set of servers the URL (uniform resource locator) of the webpage that the browser is displaying when the agent is selected. The server set can include one server or more than one server. The server set determines whether the URL is part of a set of one or more social media communities from which the server set can capture employment candidate profiles. When the server set cannot identify a social media community for the webpage at issue, the server set in some embodiments returns to the browser agent a message to indicate that it cannot capture a profile from this page. On the other hand, when the server set matches the URL to a social media community, the server set in some embodiments informs the browser agent to send a copy of the webpage to the server set. In other embodiments, the server set does not immediately ask the browser agent to send a copy of the webpage after matching the URL to a social media community. For instance, even after the server set matches the URL to a social media community, the server set will not request the profile webpage until the user that is viewing the webpage requests the candidate profile to be captured.
1 . A method of building candidate profiles comprising:
receiving a webpage that contains a candidate's profile;
identifying a category to which the webpage belongs;
based on the identified category, selecting a parsing process from a plurality of different parsing processes for a plurality of different categories of webpages; and
using the selected parsing process to parse the retrieved webpage into a plurality of data tuples and mapping the parsed data tuples to a uniform data tuple structure of a uniform data store that stores candidate profiles from different webpage categories.
2 . The method of claim 1 , wherein receiving the webpage comprises:
receiving a URL (uniform resource locator) of the webpage that contains the candidate's profile; and
using the URL to determine whether the webpage should be retrieved.
3 . The method of claim 2 , wherein the URL is received after a bookmarklet is selected while the webpage is being displayed in a web browser.
4 . The method of claim 3 further comprising providing the bookmarklet to a device that is used to display the webpage
5 . The method of claim 2 ,
wherein after determining that the webpage should be retrieved, providing data to a browser agent that is displaying the webpage that the webpage is a candidate webpage for capturing;
wherein the webpage is received after the browser agent is directed to capture the webpage.
6 . The method of claim 2 , wherein after determining that the webpage should be retrieved, directing a browser agent that is displaying the webpage to capture the webpage.
7 . The method of claim 2 , wherein the URL is received from a browser extension that a web browser executes while displaying the webpage.
8 . The method of claim 1 , wherein the different categories of webpages are different social media communities, wherein the received webpage is a social media webpage.
9 . A method of building candidate profiles comprising:
through a network, receiving an invocation of a browser agent of a browser while the browser is displaying a webpage that contains a candidate's profile;
through the network, directing the browser agent to send a copy of the webpage;
parsing the webpage copy into a plurality of data tuples; and
storing the parsed data tuples in a data storage that stores candidate profiles that are gathered from a plurality of webpages.
10 . The method of claim 9 further comprising providing the browser agent to a device that executes the browser in order to the install the browser agent.
11 . The method of claim 9 further comprising:
identifying the candidate whose page is being displayed by the browser; and
to the browser agent, providing information about the candidate, said information being stored before the browser started to display the webpage.
12 . The method of claim 11 , wherein the provided information comprises data tuples previously parsed from a set of webpages related to the candidate.
13 . The method of claim 12 , wherein the provided information comprises at least one prior communication with the candidate.
14 . The method of claim 12 further comprising
creating an initial candidate profile after receiving an initial set of data tuples for the candidate and storing an initial set of parsed data values relating to the candidate; and
referring the candidate to a position;
wherein the provided information comprises at least one prior position to which the candidate was referred.
15 . The method of claim 14 , wherein the previously parsed information is stored in the data storage before receiving the data tuples.
16 . The method of claim 9 , wherein the browser agent is bookmarklet.
17 . The method of claim 9 , wherein the browser agent is a browser extension.
18 . The method of claim 9 ,
wherein the invocation of the browser agent is an invocation of the browser agent to capture a webpage,
wherein before the invocation, receiving a URL(uniform resource locator) of the webpage from the browser agent, determining that the URL is for a candidate-profile page, and informing the browser agent that the webpage is a candidate-profile page.
19 . A non-transitory machine readable medium storing a program for capturing candidate profiles, the program comprising sets of instructions for:
receiving a webpage that contains a candidate's profile;
identifying a category to which the webpage belongs;
based on the identified category, selecting a parsing process from a plurality of different parsing processes for a plurality of different categories of webpages;
using the selected parsing process to parse the retrieved webpage into a plurality of data tuples and mapping the parsed data tuples to a uniform data tuple structure of a uniform data store that stores candidate profiles from different webpage categories.
20 . The non-transitory machine readable medium of claim 19 , wherein the set of instructions for receiving the webpage comprises a set of instructions for:
receiving a URL (uniform resource locator) of the webpage that contains the candidate's profile; and
using the URL to determine whether the webpage should be retrieved.
21 . The non-transitory machine readable medium of claim 20 , wherein the program further comprises a set of instructions for providing data to a browser agent that is displaying the webpage that the webpage is a candidate webpage for capturing, after determining that the webpage should be retrieved, wherein the webpage is received after the browser agent is directed to capture the webpage.
22 . The non-transitory machine readable medium of claim 20 , wherein the program further comprises a set of instructions for directing a browser agent that is displaying the webpage to capture the webpage after determining that the webpage should be retrieved.