Systematic identification and masking of private data for replaying user sessions
Techniques are described herein for identifying and masking instances of private information in tracked user interactions with a network site during a network session. A scanning module can receive set of captured user interactions (e.g., movements between portions of the network site and data provided in one or more fields on the network site) with the network site during the network session. Instances of private information can be identified and a number of nodes in a document object model (DOM) for the network site that correspond to each identified instance of private information. Information about each corresponding node can be provided to update a capture agent to mask each corresponding node in future user interactions.
1. A method for masking private information captured during user interactions with a network site during a network session, the method comprising performing by a server:
obtaining, from a capture agent operating on a user device, a set of captured user interactions with the network site during the network session, wherein the set of captured user interactions include (1) movements between portions of the network site and (2) data provided in one or more fields on the network site;
identifying, by a scanning module of the server, instances of private information being included in the set of captured user interactions with the network site;
assigning, for each of the instances of private information, a timestamp indicating a time of each instance of private information provided during the set of captured user interactions with the network site;
loading, into a browser module, changes of a document object model (DOM) for the network site, the changes corresponding to the set of captured user interactions;
identifying, for each of the instances of private information, a corresponding node in the document object model (DOM) for the network site using the timestamp assigned to each of the instances of private information; and
providing information about each corresponding node to an additional module as part of a process for updating the capture agent to mask the corresponding node in future user interactions sent to the server.
2. The method of claim 1 , further comprising:
identifying, for each of the instances of private information, an information type included in each instance of private information; and
identifying, for each of the instances of private information, a masking operation that corresponds with the information type of each instance of private information, wherein the information about each corresponding node includes masking operations for each of the instances of private information.
3. The method of claim 2 , wherein the masking operation includes an encryption operation to encrypt private information such that the encrypted private information can be subsequently decrypted to securely access the private information.
4. The method of claim 2 , wherein the masking operation includes a removal operation to remove private information.
5. The method of claim 1 , wherein the instances of private information include text-based data provided in a field displayed on the network site.
6. The method of claim 1 , further comprising:
determining, for each corresponding node, that each corresponding node corresponds with a threshold number of previous network sessions identifying each corresponding node in the DOM for the network site; and
adding each corresponding node that corresponds with the threshold number of previous network sessions to a blacklist to mask data in each corresponding node in subsequent network sessions.
7. The method of claim 1 , wherein providing the information about each corresponding node to the additional module includes:
sending the information to a user verification module capable of presenting the information about each corresponding node for confirmation of each corresponding node as including private information; and
updating the capture agent to mask corresponding nodes that are confirmed as including private information by the user verification module.
8. The method of claim 1 , further comprising:
identifying a first field as relating to private information;
determining whether a first value of the first field is common across a threshold number of network sessions; and
ignoring the first field for masking in future sessions.
9. The method of claim 1 , further comprising:
filtering the set of captured user interactions using a filtering module, wherein the filtering module determines which of the set of captured user interactions satisfy at least one criterion, thereby determining filtered user interactions; and
sending the filtered user interactions to the scanning module.
10. The method of claim 9 , wherein the least one criterion includes one or more criteria of a page, the one or more criteria comprising a number of page views across users in a timeframe, an engagement metric providing a measure of user interaction with the page, a customer-specified criterion specifying a property of the page of the network site, or combinations thereof.
11. A system comprising:
one or more processors; and
a non-transitory computer-readable medium including instructions that, when executed by the one or more processors, cause the one or more processors to perform a method for masking private information captured during user interactions with a network site during a network session, the method including:
obtaining, from a capture agent operating on a user device, a set of captured user interactions with the network site during the network session, wherein the set of captured user interactions include (1) movements between portions of the network site and (2) data provided in one or more fields on the network site;
identifying, by a scanning module, instances of private information being included in the set of captured user interactions with the network site;
assigning, for each of the instances of private information, a timestamp indicating a time of each instance of private information provided during the set of captured user interactions with the network site;
loading, into a browser module, changes of a document object model (DOM) for the network site, the changes corresponding to the set of captured user interactions;
identifying, for each of the instances of private information, a corresponding node in the document object model (DOM) for the network site using the timestamp assigned to each of the instances of private information; and
providing information about each corresponding node to an additional module as part of a process for updating the capture agent to mask the corresponding node in future user interactions sent to the system.
12. The system of claim 11 , wherein the method further includes:
identifying, for each of the instances of private information, an information type included in each instance of private information; and
identifying, for each of the instances of private information, a masking operation that corresponds with the information type of each instance of private information, wherein the information about each corresponding node includes masking operations for each of the instances of private information.
13. The system of claim 12 , wherein the masking operation includes an encryption operation to encrypt private information such that the encrypted private information can be subsequently decrypted to securely access the private information.
14. The system of claim 12 , wherein the masking operation includes a removal operation to remove private information.
15. The system of claim 11 , wherein the instances of private information include text-based data provided in a field displayed on the network site.
16. The system of claim 11 , wherein the method further includes:
determining, for each corresponding node, that each corresponding node corresponds with a threshold number of previous network sessions identifying each corresponding node in the DOM for the network site; and
adding each corresponding node that corresponds with the threshold number of previous network sessions to a blacklist to mask data in each corresponding node in subsequent network sessions.
17. The system of claim 11 , wherein providing the information about each corresponding node to the additional module includes:
sending the information to a user verification module capable of presenting the information about each corresponding node for confirmation of each corresponding node as including private information; and
updating the capture agent to mask corresponding nodes that are confirmed as including private information by the user verification module.
18. The system of claim 11 , wherein the method further includes:
identifying a first field as relating to private information;
determining whether a first value of the first field is common across a threshold number of network sessions; and
ignoring the first field for masking in future sessions.
19. The system of claim 11 , wherein the method further includes:
filtering the set of captured user interactions using a filtering module, wherein the filtering module determines which of the set of captured user interactions satisfy at least one criterion, thereby determining filtered user interactions; and
sending the filtered user interactions to the scanning module.
20. The system of claim 19 , wherein the least one criterion includes one or more criteria of a page, the one or more criteria comprising a number of page views across users in a timeframe, an engagement metric providing a measure of user interaction with the page, a customer-specified criterion specifying a property of the page of the network site, or combinations thereof.