Ment.io’s discussion analyzer is an AI-based algorithm that helps organize your discussions.
Structuring discussions helps participants see highlights and get a better understanding of every discussion.
Learn more about our algorithm
Each post, answer and/or comment, receives a Ment.io score based on the individual rating of the user posting. Each user affects the score differently based on their own unique user score.
To provide the most accurate overall score, and avoid biases, our AI-based discussion analyzer calculates the score based on the weighted agreement level of the team towards the content as well as based on understanding the “thought distance” between the team members, and patterns of agreements and disagreements between the team members.
The Ment.io score is based both on a smart peer review and knowledge analytics. It is not just the average of votes, but based on the discussion structure and similar past discussions.
Each discussion is stored in a graph database, and Bayesian and Machine Learning algorithms are utilized to decide the score. Ment.io also follows the activity of each of the participants to analyze their domain expertise and include this information in the scoring mechanism. Analysis of the typical behavior of different departments is also incorporated into the algorithm.
Each user receives a score based on the quality of their answers and comments as rated by their teammates with comments and votes.
Our AI-based discussion analyzer calculates the score based on weighted multiple factors including the level of teammates’ agreement with the user’s content, total number of contributions, recent activity level, as well as the content’s reach (number of times team members viewed the user’s answers and comments).
This score is represented as a percentile, and users are seen as top % of their teammates. Ment.io visually presents this score using a range of one to five stars. Users with no stars are in the bottom half percentile compared to their teammates.