Browse Prior Art Database

A System for Predictively Recording Meetings of Interest Disclosure Number: IPCOM000202101D
Publication Date: 2010-Dec-03
Document File: 3 page(s) / 25K

Publishing Venue

The Prior Art Database


Disclosed is a system and method for ensuring that meeting content is available to the appropriate members of an organization. The system provides people with alerts to relevant meetings, predicts audience interest, records meetings, and provides a method of notification and recommendation to users.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 3

A System for Predictively Recording Meetings of Interest

In today's busy, multi-tasking, and global society, people are often overly committed or simply unable to attend all the meetings that they should or would like. In addition, sometimes meetings take place in a person's interest area that they don't even know about, but would want to be alerted to.

Unless a person attends a meeting, or the meeting is explicitly recorded, they and others in their organization cannot fully benefit from the information conveyed in that meeting.

To provide people with alerts to relevant meetings, a tool that predicts audience interest, records meetings, and provides a method of notification and recommendation to users
is needed. Such a tool would have the following characteristics:
· Utilizes proper safeguards, to ensure that private data is not inadvertently disclosed

· Datamines the content of meetings in real-time, and records meetings where content is of interest to the community

· Matches personal interest settings to the meeting content, to determine which recordings to retain

· Alerts people to meeting recordings for meetings with topics of interest to them
· Learns from what is actually watched to help hone the personal interest settings
· Discards recordings that are not of interest to anyone

Current technologies make it easier to record meetings, and new technologies make it possible to mine the content of meetings, without having to go through them in real time. But what's missing is a way to match up the recordings with the people who might be interested in viewing them. This disclosure is about bridging that gap -- using various types of metrics and self-learning algorithms to allow the system to determine that one
or more people might be interested in this meeting, to pro-actively record the meeting only when there is potential interest, make interested parties aware of it, and then efficiently remove it when all parties are done with it.

The invention uses user-oriented attributes, such as people in a common workgroup or network, topics in which users have indicated an interest, and a user's calendar entries, to predict interest and pro-actively record meetings, allowing users to watch or listen to recordings at their convenience.

The invention discovers the meetings by applying data such as the following:
· Meeting attendees
· A user's calendar (e.g., you were invited but could not attend a meeting)
· Content in the meeting chat transcript
· A transcription of the conversation during the meeting (using text-to-speech technologies)

· Content shown in slides
· Other information derived from a meeting


Page 02 of 3

The system has the ability to warn meeting owners before meetings are recorded, allowing them the option to o...