Browse Prior Art Database

Personalized social network video message generation with occasion-relevant social content and captions Disclosure Number: IPCOM000252809D
Publication Date: 2018-Feb-14
Document File: 3 page(s) / 71K

Publishing Venue

The Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Personalized social network video message generation with occasion-relevant social content and captions Social networks create automated video messages for users to celebrate an occasion. For example, a one-year anniversary of two people connected as friends on a social network can result in an automated video message showing pictures and posts between these friends during this period. However, there is no existing art for identifying occasion-relevant social network content ahead of an occasion and then contacting the posters of that content to provide captions for a personalized video. The novel contribution is a method to generate an automated video message, to send to a social network user on a special occasion, that features social network postings relevant to the occasion. Further, the novel approach includes a method to personalize this video message, whereby the system automatically requests individual social network connections to provide captions to social content that the system has derived as relevant to the occasion. The result is a video on the date of the occasion with personalized captions from friends what were retrieved prior to the occasion occurring. The core novel features of the system include:

 Analysis of content (e.g., text, pictures, videos, and events) posted on social networks to derive content that is relevant to an upcoming occasion for a user on the social network

 Generation of requests for occasion-relevant social content to be captioned by social network connections for an upcoming occasion. These requests are sent to social network connections based on predicted response time.

 Assembly of a video message of captioned occasion-relevant social content, presented to a user on the date of an occasion

 Contextual validation of the video message content and removal of inappropriate content

Figure: System overview

Stage 1: Social network identifies upcoming occasion and retrieves related social network posts

1. A social network analyzes a user’s profile to derive an upcoming occasion (e.g., birthday, anniversary, festival/celebration, etc.)

2. In anticipation of the upcoming occasion, the social network analyzes posts of a user’s social network connections to derive who would be the most appropriate connections to help celebrate this occasion. This analysis utilizes the following cognitive services:

A. Natural Language Understanding: analyzes text to extract meta-data from content such as concepts, entities, keywords, categories, sentiment, emotion, relations, and semantic roles

B. Tone Analyzer: uses linguistic analysis to detect communication tones in written text

C. Visual Recognition: understands the contents of images (e.g., visual concepts tagged in the image, recognize food, find human faces, approximate age and gender, and find similar images in a collection)

3. Through this cognitive analysis, the system derives which social content posted by which social connections are most appropriate to the upcomi...