Method to randomize online activity
A computer-implemented method for digital fingerprint obfuscation is disclosed. The computer-implemented method includes training a machine learning model to classify web traffic data into one or more personas. The computer-implemented method further includes identifying, using the trained machine learning model, a particular persona of a user based, at least in part, on a user's real traffic data generated during a current user session. The computer-implemented method further includes generating synthetic traffic data based, at least in part, on the identified particular persona of the user.
1 . A computer-implemented method for digital fingerprint obfuscation, the computer-implemented method comprising:
sanitizing training data by stripping the training data of information that identifies source individuals, the source of training data being online user activity or synthetically generated online user activity;
training a machine learning model using the training data to classify a user's real traffic data into one or more personas;
collecting and marking a user's real traffic data to run classification algorithms to generate a base-persona, the base persona being main attributes connected to the user, the base persona having similar traits, likes, or dislikes from the user's real traffic data;
identifying, using the trained machine learning model, a particular persona of a user based, at least in part, on a user's real traffic data generated during a current user session;
generating synthetic traffic data based, at least in part, on a different persona than the particular persona of the user identified using the trained machine learning model; and
combining the user's real traffic data of the particular persona with the synthetic traffic data of the different persona to generate combined traffic data and plotting the combined traffic data as a traffic leveling graph to generate a visualization of the flattening of the user's traffic patterns.
2 . The computer-implemented method of claim 1 , further comprising:
transmitting the generated synthetic traffic data after the current user session has ended.
3 . The computer-implemented method of claim 2 , wherein transmitting the generated synthetic traffic data further comprises:
simultaneously transmitting the generated synthetic traffic data simultaneously with the real user traffic data generated during the current user session.
4 . The computer-implemented method of claim 1 , further comprising:
performing traffic leveling during the current user session, wherein performing traffic leveling includes transmitting the generated synthetic traffic data that is complementary to the user traffic data.
5 . The computer-implemented method of claim 4 , wherein traffic leveling is further based, at least in part, on an amount and persona of generated synthetic traffic data which is based, at least in part, on real traffic data.
6 . The computer-implemented method of claim 1 , wherein generating synthetic traffic is further based, at least in part, on:
generating a first score for the identified particular persona of the user;
determining a second persona and a second score for the second persona with a difference in the first score and the second score being above a predetermined threshold; and
generating synthetic traffic data based, at least in part, on the second persona.
7 . The computer-implemented method of claim 1 , further comprising:
generating a first score for the identified particular persona of the user;
determining a second persona and a second score for the second persona with a difference in the first score and the second score being below a predetermined threshold; and
generating synthetic traffic data based, at least in part, on the second persona.
8 . The computer-implemented method of claim 1 , wherein training the machine learning model to classify web traffic data into one or more personas is further based, at least in part on one or more of: active time of day of a user's web browsing history, web pages visited, web browsers accessed, applications accessed, and local contents present in the users' web browser or computer.
9 . The computer-implemented method of claim 1 , wherein generating synthetic traffic data is based, at least in part, on anonymizing a respective traffic pattern by traffic leveling using the synthetic traffic data to create a complimentary frequency of automated traffic to level an original traffic volume curve of a respective user plotted against time vs volume.
10 . The computer-implemented method of claim 1 , wherein generating synthetic traffic data is further based, at least in part on:
generating synthetic traffic data using the identified particular persona of the user to form a curated web surfing pattern based on a mixture of universal resource locators (URLs) accessed by users having different associated personas from the particular persona of the user.
11 . A computer program product for digital fingerprint obfuscation, the computer program product comprising one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the program instructions including instructions to:
sanitize training data by stripping the training data of information that identifies source individuals, the source of training data being online user activity or synthetically generated online user activity;
train a machine learning model using the training data to classify web a user's real traffic data into one or more personas;
collect and mark a user's real traffic data to run classification algorithms to generate a base-persona, the base persona being main attributes connected to the user, the base persona having similar traits, likes, or dislikes from the user's real traffic data;
identify, using the trained machine learning model, a particular persona of a user based, at least in part, on a user's real traffic data generated during a current user session;
generate synthetic traffic data based, at least in part, on a different persona than the particular persona of the user identified using the trained machine learning model; and
combine the user's real traffic data of the particular persona with the synthetic traffic data of the different persona to generate combined traffic data and plot the combined traffic data as a traffic leveling graph to generate a visualization of the flattening of the user's traffic patterns.
12 . The computer program product of claim 11 , further comprising instructions to:
transmit the generated synthetic traffic data after the current user session has ended.
13 . The computer program product of claim 12 , wherein the instructions to transmit the generated synthetic traffic data further comprises instructions to:
simultaneously transmit the generated synthetic traffic data simultaneously with the real user traffic data generated during the current user session.
14 . The computer program product of claim 11 , further comprising instructions to:
perform traffic leveling during the current user session, wherein performing traffic leveling includes transmitting the generated synthetic traffic data that is complementary to the user traffic data.
15 . The computer program product of claim 14 , wherein traffic leveling is further based, at least in part, on an amount and persona of generated synthetic traffic data which is based, at least in part, on real traffic data.
16 . The computer program product of claim 11 , wherein generating synthetic traffic is further based, at least in part, on instructions to:
generate a first score for the identified particular persona of the user;
determine a second persona and a second score for the second persona with a difference in the first score and the second score being above a predetermined threshold; and
generate synthetic traffic data based, at least in part, on the second persona.
17 . The computer program product of claim 11 , further comprising instructions to:
generate a first score for the identified particular persona of the user;
determine a second persona and a second score for the second persona with a difference in the first score and the second score being below a predetermined threshold; and
generate synthetic traffic data based, at least in part, on the second persona.
18 . The computer program product of claim 11 , wherein training the machine learning model to classify web traffic data into one or more personas is further based, at least in part on one or more of: active time of day of a user's web browsing history, web pages visited, web browsers accessed, applications accessed, and local contents present in the users' web browser or computer.
19 . The computer program product of claim 11 , wherein generating synthetic traffic data is based, at least in part, on anonymizing a respective traffic pattern by traffic leveling using the synthetic traffic data to create a complimentary frequency of automated traffic to level an original traffic volume curve of a respective user plotted against time vs volume.
20 . A computer system for digital fingerprint obfuscation, comprising:
one or more computer processors;
one or more computer readable storage media;
computer program instructions;
the computer program instructions being stored on the one or more computer readable storage media for execution by the one or more computer processors; and the computer program instructions including instructions to:
sanitize training data by stripping the training data of information that identifies source individuals, the source of training data being online user activity or synthetically generated online user activity;
train a machine learning model using the training data to classify a user's real traffic data into one or more personas
collect and mark a user's real traffic data to run classification algorithms to generate a base-persona, the base persona being main attributes connected to the user, the base persona having similar traits, likes, or dislikes from the user's real traffic data;
identify, using the trained machine learning model, a particular persona of a user based, at least in part, on a user's real traffic data generated during a current user session;
generate synthetic traffic data based, at least in part, on a different persona than the particular persona of the user identified using the trained machine learning model; and
combine the user's real traffic data of the particular persona with the synthetic traffic data of the different persona to generate combined traffic data and plot the combined traffic data as a traffic leveling graph to generate a visualization of the flattening of the user's traffic patterns.