Real-time analysis of predictive audience feedback during content creation

• US 10,169,713 B2
• Filed: 02/01/2018
• Issued: 01/01/2019
• Est. Priority Date: 03/08/2017
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Blanco, Yenny B., Butkus, Petras, Harpur, Liam, Mariani, Francesco, Silveira, Larissa A., Yadav, Nagesh
• Primary Examiner(s): Misir, Dave
• Application Number:
  US15/885,974
• Publication Number:
  US 20180260727A1
• Time in Patent Office: 334 Days
• Field of Search: 706 11, 706 46
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Providing predictive feedback during content creation in real-time. A server computer receives from a user-computing device an electronic document during creation of the electronic document in real-time. An intended audience specification is also received by the server computer for the received electronic document. The server computer calculates a similarity score for a plurality of viewer profiles. The server computer determines which viewer profiles have similarity score equal to or exceeding the similarity score threshold and then stores the determined viewer profiles. An individual interest level and an individual emotional response is calculated for the determined viewer profiles for a portion of the received electronic document. The server computer then transmits to the user-computing device the individual interest level and the individual emotional response of the determined viewer profiles.

Claims

1. A method for providing and utilizing predictive feedback during content creation in real-time comprising:

obtaining demographic data for profile generation from a viewer profile database;

receiving, by a server computer via a network adapter or interface from a user-computing device, an author-created electronic document during creation of the electronic document in real-time, the electronic document selected from a group consisting of a text file, a video file, and an audio file;

receiving, by the server computer, an intended audience specification for the received electronic document;

receiving, by the server computer from the user-computing device, a similarity score threshold;

calculating by the server computer a similarity score for a plurality of viewer profiles, the viewer profiles generated via obtaining of the demographic data and a viewer profile member assay via each viewer wearing one or more wearable sensors while consuming sample digital content;

determining by the server computer which viewer profiles have similarity score equal to or exceeding the similarity score threshold and storing the determined viewer profiles;

calculating by the server computer an individual interest level and an individual emotional response of the determined viewer profiles for a portion of the received electronic document by presentation of a graphic user interface, the individual interest level indicating a level of interest each viewer profile of the determined viewer profiles is likely to express regarding the portion of the received electronic document and the individual emotional response indicting indicating an emotional response anticipated for the portion of the electronic document for each viewer profile of the determined viewer profiles;

transmitting, by the server computer to the user-computing device, the individual interest level and the individual emotional response of the determined viewer profiles, weighting the individual interest level and individual emotional response of the determined viewer profiles according to the similarity score; and

modifying automatically by the server computer in real-time the portion of the received electronic document to improve the individual interest level of the determined viewer profiles and the individual emotional response of the determined viewer profiles by automatically altering objectionable language by the server computer including text, if the received electronic document is the text file, or by automatically bleeping out offensive words by the server computer if the received electronic document is the video file or the audio file.