System and method for detecting fraudulent software installation activity
A method, a system, and an article are provided for detecting and managing anomalies associated with content presentations. An example computer-implemented method includes: obtaining a stream of data including a record of software installations on a plurality of client devices; processing the stream of data using a real-time anomaly detector to detect a first anomaly indicating that a client device was previously associated with fraudulent software installation activity; aggregating the stream of data to generate a batch record of software installations for each publisher from a plurality of content publishers; processing each batch record of software installations using a batch anomaly detector to detect a second anomaly indicating that at least one of the publishers is associated with fraudulent software installation activity; and facilitating an adjustment of content presentations on the client devices, based on at least one of the detected first anomaly and the detected second anomaly.
1. A method, comprising:
obtaining a stream of data comprising a record of software installations occurring on a plurality of client devices, at least a portion of the software installations having been derived from user exposure to content provided on the client devices by a plurality of publishers,
wherein a software installation occurs when a software application is first installed or first used on one of the client devices, and
wherein the record of software installations comprises a timestamp, a client device identifier, and a publisher identifier;
processing the stream of data using a real-time anomaly detector to detect a first anomaly indicating that at least one of the client devices was previously associated with fraudulent software installation activity;
aggregating the stream of data to generate a batch record of software installations for each publisher in the plurality of publishers;
processing each batch record of software installations using a batch anomaly detector to detect a second anomaly indicating that at least one of the publishers is associated with fraudulent software installation activity; and
facilitating an adjustment of content presentations on the client devices, based on at least one of the detected first anomaly and the detected second anomaly.
2. The method of claim 1 , wherein processing the stream of data comprises: sending each software installation record to the real-time anomaly detector when the software installation record is obtained.
3. The method of claim 1 , wherein processing the stream of data comprises: determining that a device blacklist identifies the at least one of the client devices.
4. The method of claim 1 , wherein the batch anomaly detector comprises a plurality of detection algorithms, and wherein each detection algorithm is configured to detect at least one distinct anomaly from a plurality of detectable anomalies.
5. The method of claim 4 , wherein processing each batch record comprises: selecting, for each batch record, at least one detection algorithm from the plurality of detection algorithms.
6. The method of claim 1 , wherein processing each batch record comprises: aggregating the batch record into a first record of organic software installations, a second record of non-incentivized software installations, and a third record of incentivized software installations, wherein (i) the incentivized software installations were derived from user exposure to content that offers a reward, (ii) the non-incentivized software installations were derived from user exposure to content that does not offer a reward, and (iii) the organic software installations were not derived from user exposure to content.
7. The method of claim 6 , wherein processing each batch record further comprises: processing the first record, the second record, and the third record using respective collections of detection algorithms, wherein each collection of detection algorithms is configured to detect a distinct set of anomalies.
8. The method of claim 1 , wherein facilitating the adjustment comprises preventing at least one publisher from presenting content on the client devices.
9. The method of claim 1 , further comprising: identifying at least one of the client devices as being associated with the second anomaly; and adding the identified at least one of the client devices to a blacklist.
10. The system of claim 1 , wherein facilitating the adjustment comprises preventing at least one publisher from presenting content on the client devices.
11. The system of claim 1 , further comprising: identifying at least one of the client devices as being associated with the second anomaly; and adding the identified at least one of the client devices to a blacklist.
12. A system, comprising:
one or more computer processors programmed to perform operations comprising:
obtaining a stream of data comprising a record of software installations occurring on a plurality of client devices, at least a portion of the software installations having been derived from user exposure to content provided on the client devices by a plurality of publishers,
wherein a software installation occurs when a software application is first installed or first used on one of the client devices; and
wherein the record of software installations comprises a timestamp, a client device identifier, and a publisher identifier;
processing the stream of data using a real-time anomaly detector to detect a first anomaly indicating that at least one of the client devices was previously associated with fraudulent software installation activity;
aggregating the stream of data to generate a batch record of software installations for each publisher in the plurality of publishers;
processing each batch record of software installations using a batch anomaly detector to detect a second anomaly indicating that at least one of the publishers is associated with fraudulent software installation activity; and
facilitating an adjustment of content presentations on the client devices, based on at least one of the detected first anomaly and the detected second anomaly.
13. The system of claim 12 , wherein processing the stream of data comprises: sending each software installation record to the real-time anomaly detector when the software installation record is obtained.
14. The system of claim 12 , wherein processing the stream of data comprises: determining that a device blacklist identifies the at least one of the client devices.
15. The system of claim 12 , wherein the batch anomaly detector comprises a plurality of detection algorithms, and wherein each detection algorithm is configured to detect at least one distinct anomaly from a plurality of detectable anomalies.
16. The system of claim 15 , wherein processing each batch record comprises: selecting, for each batch record, at least one detection algorithm from the plurality of detection algorithms.
17. The system of claim 12 , wherein processing each batch record comprises: aggregating the batch record into a first record of organic software installations, a second record of non-incentivized software installations, and a third record of incentivized software installations, wherein (i) the incentivized software installations were derived from user exposure to content that offers a reward, (ii) the non-incentivized software installations were derived from user exposure to content that does not offer a reward, and (iii) the organic software installations were not derived from user exposure to content.
18. The system of claim 17 , wherein processing each batch record further comprises: processing the first record, the second record, and the third record using respective collections of detection algorithms, wherein each collection of detection algorithms is configured to detect a distinct set of anomalies.
19. An article, comprising:
a non-transitory computer-readable medium having instructions stored thereon that, when executed by one or more computer processors, cause the computer processors to perform operations comprising:
obtaining a stream of data comprising a record of software installations occurring on a plurality of client devices, at least a portion of the software installations having been derived from user exposure to content provided on the client devices by a plurality of publishers,
wherein a software installation occurs when a software application is first installed or first used on one of the client devices; and
wherein the record of software installations comprises a timestamp, a client device identifier, and a publisher identifier;
processing the stream of data using a real-time anomaly detector to detect a first anomaly indicating that at least one of the client devices was previously associated with fraudulent software installation activity;
aggregating the stream of data to generate a batch record of software installations for each publisher in the plurality of publishers;
processing each batch record of software installations using a batch anomaly detector to detect a second anomaly indicating that at least one of the publishers is associated with fraudulent software installation activity; and
facilitating an adjustment of content presentations on the client devices, based on at least one of the detected first anomaly and the detected second anomaly.