Browse Prior Art Database

Automatic meetings organizer with a cognitive solution Disclosure Number: IPCOM000247278D
Publication Date: 2016-Aug-18
Document File: 7 page(s) / 166K

Publishing Venue

The Prior Art Database


Method to schedule meetings through correlations and computation. One key difference is about scoring. Here, this is not the time slot as usual but all the components making a meeting including importance of the subject and weight on participants depending of the needs for them to attend or not.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 21% of the total text.

Page 01 of 7

Automatic meetings organizer with a cognitive solution

I.1 Context

This invention is in the area of agenda optimizer, especially meeting planning, and provide a mean to compute a score for each meeting in order to determine the best slot for a meeting without intervention of people and considering the context of the meeting and the availability of attendees.

I.2 Problem

    Organizing a meeting is often a fastidious task. Depending of the number of people, it can take several minutes to several decades to finally find the best slot. And thus, without taking account the fact that others meetings can be scheduled after. Depending of the priority of each, agenda can be changed and make people repeat the process once again, creating waste of time.

    While it exists some tools to automatically find the first available slot, it does not allow necessarily find the best slot or take account the context, neither take them account the importance of meeting or people in it.

    That way, an important meeting M1 can be scheduled in few days while another M2 less important occurring the day M1 is planed could be rescheduled to optimize process.

    Thus it can provoke delay in decision taking and a lot of latency in procedures and have business impacts.

    We propose a way to optimize agenda planning and find best slot for a meeting, even if it means reschedule some. For this, at each meeting is associated a score. This score allows to sort the meeting by importance and so organize them with the best way.

    Our invention follows these different steps: 1) Analyse the environment:
The environment consists of the context, the kind of meeting and the people involved in the meeting.

    For each of these 3 cases (contexts, kind of meeting and people), some information will be retrieved called "properties " and "criteria ".

2) Computation

From the properties and criteria from step 1, a compute engine will put in relation

criteria and properties and compute a correlation value.

    Then, from all the correlation value computed, a meeting score corresponding to the importance of the meeting and a participate score corresponding to the importance of the person for the considering meeting will be computed by the compute engine.

    These different values can also be adapted with the time. From a machine learning algorithm, the engine can modify its way of computing correlation value.

3) Schedule

From the different participants scores and the meeting score computed in step 2,, the

schedule engine will find the best slot in the agenda.

    For this, it is based on a threshold system. Depending of the value of the meeting score and participant score, the schedule engine will be able to determine the period of time the meeting must occur and which people need to be present or no.

State of the implementation

Only the theory was done. No prototype was made currently.

We demonstrate the concept with an example.

Let the current time the end of the quarter. It is a closing period for many compani...