Privacy policy-driven emotion detection
This disclosure describes techniques for protecting privacy of a user with respect to emotion detection via a computer network. The techniques may include receiving sensed data associated with a user. A privacy policy of the user may be used with processing of the sensed data. For example, based at least in part on the privacy policy, a private subset of the sensed data may be filtered from remaining sensed data. The remaining sensed data may be used to determine an emotion classification result. The emotion classification result may indicate a sharable emotion of the user, for instance.
1. A server device comprising:
one or more processors; and
one or more non-transitory computer-readable media storing computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to:
access a privacy policy of a user, the privacy policy associated with a first emotion of the user;
based at least in part on the privacy policy, determine a mask related to the first emotion of the user;
in an initial stage of an emotion inference process, apply the mask to filter out a private subset of sensed data that corresponds to the first emotion of the user, creating filtered sensed data, the filtered sensed data corresponding to a second emotion of the user;
in a subsequent stage of the emotion inference process, determine, using the filtered sensed data, an emotion classification result indicating the second emotion of the user; and
send the emotion classification result indicating the second emotion of the user to a remote device.
2. The server device of claim 1 , wherein the computer-executable instructions further cause the one or more processors to:
cause the applying the mask to filter out the private subset of the sensed data corresponding to the first emotion to be performed at an edge device; and
receive the filtered sensed data corresponding to the second emotion from the edge device, the filtered sensed data excluding the private subset of the sensed data corresponding to the first emotion.
3. The server device of claim 2 , wherein the applying the mask to filter out the private subset of the sensed data corresponding to the first emotion is performed in conjunction with a feature extraction process.
4. The server device of claim 2 , wherein determining the emotion classification result comprises applying a classification process to the filtered sensed data received from the edge device.
5. The server device of claim 2 , wherein causing the applying the mask to filter out the private subset of the sensed data to be performed at the edge device comprises provisioning the edge device with processing capabilities associated with filtering the private subset.
6. The server device of claim 1 , wherein the computer-executable instructions further cause the one or more processors to:
receive a user selection associated with the first emotion of the user; and
cause an indication of the user selection to be associated with the privacy policy of the user.
7. The server device of claim 1 , wherein the mask to filter out the private subset of the sensed data that corresponds to the first emotion of the user is applied in the initial stage of the emotion inference process before determination of the emotion classification result indicating the second emotion of the user in the subsequent stage of the emotion inference process.
8. The server device of claim 7 , wherein the mask is applied in the initial stage of the emotion inference process at an edge device, and the determination of the emotion classification result indicating the second emotion of the user is made in the subsequent stage of the emotion inference process at the server device.
9. A computer-implemented method comprising:
accessing a privacy policy of a user, the privacy policy associated with a first emotion of the user;
based at least in part on the privacy policy, determining a mask related to the first emotion of the user;
in an initial stage of an emotion inference process, applying the mask to filter out a private subset of sensed data that corresponds to the first emotion of the user, creating filtered sensed data, the filtered sensed data corresponding to a second emotion of the user;
in a subsequent stage of the emotion inference process, determining, using the filtered sensed data, an emotion classification result indicating the second emotion of the user; and
sending the emotion classification result indicating the second emotion of the user to a remote device.
10. The computer-implemented method of claim 9 , further comprising:
causing the applying the mask to filter out the private subset of the sensed data corresponding to the first emotion to be performed at an edge device; and
receiving the filtered sensed data corresponding to the second emotion from the edge device, the filtered sensed data excluding the private subset of the sensed data corresponding to the first emotion.
11. The computer-implemented method of claim 10 , wherein the applying the mask to filter out the private subset of the sensed data corresponding to the first emotion is performed in conjunction with a feature extraction process.
12. The computer-implemented method of claim 10 , wherein the determining the emotion classification result comprises applying a classification process to the filtered sensed data received from the edge device.
13. The computer-implemented method of claim 10 , wherein the causing the applying the mask to filter out the private subset of the sensed data to be performed at the edge device comprises provisioning the edge device with processing capabilities associated with filtering the private subset.
14. The computer-implemented method of claim 9 , further comprising:
receiving a user selection associated with the first emotion of the user; and
causing an indication of the user selection to be associated with the privacy policy of the user.
15. The computer-implemented method of claim 9 , wherein the mask to filter out the private subset of the sensed data that corresponds to the first emotion of the user is applied in the initial stage of the emotion inference process before determination of the emotion classification result indicating the second emotion of the user in the subsequent stage of the emotion inference process.
16. The computer-implemented method of claim 15 , wherein the mask is applied in the initial stage of the emotion inference process at an edge device, and the determination of the emotion classification result indicating the second emotion of the user is made in the subsequent stage of the emotion inference process at a server device.
17. One or more non-transitory computer-readable media storing instructions that, when executed, cause one or more processors to perform operations comprising:
accessing a privacy policy of a user, the privacy policy associated with a first emotion of the user;
based at least in part on the privacy policy, determining a mask related to the first emotion of the user;
in an initial stage of an emotion inference process, applying the mask to filter out a private subset of sensed data that corresponds to the first emotion of the user, creating filtered sensed data, the filtered sensed data corresponding to a second emotion of the user;
in a subsequent stage of the emotion inference process, determining, using the filtered sensed data, an emotion classification result indicating the second emotion of the user; and
sending the emotion classification result indicating the second emotion of the user to a remote device.
18. The one or more non-transitory computer-readable media of claim 17 , the operations further comprising:
receiving a user selection associated with the first emotion of the user; and
causing an indication of the user selection to be associated with the privacy policy of the user.
19. The one or more non-transitory computer-readable media of claim 17 , wherein the mask to filter out the private subset of the sensed data that corresponds to the first emotion of the user is applied in the initial stage of the emotion inference process before determination of the emotion classification result indicating the second emotion of the user in the subsequent stage of the emotion inference process.
20. The one or more non-transitory computer-readable media of claim 19 , wherein the mask is applied in the initial stage of the emotion inference process at an edge device, and the determination of the emotion classification result indicating the second emotion of the user is made in the subsequent stage of the emotion inference process is performed by the one or more processors.