Browse Prior Art Database

A Social Signal Driven Adaptive Framework for Bucket Testing

IP.com Disclosure Number: IPCOM000234851D
Publication Date: 2014-Feb-11
Document File: 7 page(s) / 367K

Publishing Venue

The IP.com Prior Art Database

Related People

Ariyam Das: INVENTOR [+2]

Abstract

A social signal driven adaptive framework for bucket testing is disclosed. The framework reduces the risk of brand damage through rapid propagation of negative user responses via social networking sites for a new feature that is tested through bucket testing. Additionally, the framework ensures controlled propagation of positive user responses by adaptively changing a safety threshold for selecting users for bucket testing based on the users’ social networking activities.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

A Social Signal Driven Adaptive Framework for Bucket Testing

Abstract

A social signal driven adaptive framework for bucket testing is disclosed.  The framework reduces the risk of brand damage through rapid propagation of negative user responses via social networking sites for a new feature that is tested through bucket testing.  Additionally, the framework ensures controlled propagation of positive user responses by adaptively changing a safety threshold for selecting users for bucket testing based on the users’ social networking activities.

Description

Disclosed is a social signal driven adaptive framework for bucket testing.  The framework reduces the propagation of negative responses by exposed users over social networking sites for a new feature tested through bucket testing.  Also, the framework prevents induction of negative biases among people, who have not been exposed to the new feature.  In addition, the framework ensures controlled propagation of positive responses through the social network by adaptively varying a safety threshold based on users’ social networking activities.  Here, the new feature is a feature, which is tested through bucket testing and an exposed user is a user who is being exposed to the new feature through bucket testing.

Fig. 1 illustrates a flowchart of steps executed by the framework.  

Figure 1

As shown in fig.1, in step A, a selection pool is created and then in step B a social score is assigned to each user in the selection pool.  After assigning the social score, users are marked as safe users in step C based on the safety threshold and added to the test bucket for bucket testing.  Thereafter, in step D the gross overall response is calculated for the test bucket and the safety threshold is recomputed accordingly in step E.

Fig. 2 illustrates a flowchart for creating the selection pool that includes multiple users.  

Figure 2

As shown in fig.2, a user with a registered account is considered.  Thereafter, it is determined whether the user’s social profile is publicly accessible.  If the user’s social profile is publicly accessible then the user is added to the selection pool, otherwise it is determined whether the user’s account is linked with social networking sites.  If the user’s account is linked with social networking sites then the user is also added to the selection pool.

Fig. 3 illustrates a flowchart for assigning the social score to each user in the selection pool.  

Figure 3

As shown in fig 3, a user is selected from the selection pool.  Thereafter, the social score of the selected user is calculated as a weighted function of the user’s number of friends and followers and the user’s activity in the social networks such as, but not limited to, number of posts and tweets per day.  Here, appropriate weights are also assigned to the user’s activity and user’s number of friends and followers before calculating the social score.

Fig. 4 illustrates a flowchart for...