Unveiled last week and available here, The Malta Independent has teamed up with ICON and Minely to generate a social media scoreboard which is driven by Artificial Intelligence.
Today we have added another chart which visualises sentiment registered by unique active users each day. Social sentiment is a measurement based on aggregated social media data which gives an inkling of a person's frame of mind about a specific topic. This visualisation plots the sentiment distribution of 137,603 Facebook users and normalises their data in such a way that multiple actions by one user are attributed singly and do not influence the general data based on volume. In this analysis, we can see that the PL is recovering from a consecutive number of days in which it was scoring far lower than the PN.
Speaking to The Malta Independent, Dr. Gege Gatt, Director of ICON stated: "Elections empower citizens to choose their Government and true democracy is about giving a voice to every person in society. This big data project allows us to study the sentiment towards parties in Malta across a channel which has changed our society: Facebook. We are interested in how public views develop over time and how the political narrative influences the sentiment-score. It's extremely insightful to then analyse what is triggering changes in emotional reaction and detect whether the two campaign-teams are acting on such triggers. Searching for people's opinions and sentiment via polls has been an expensive and time-consuming task - this changes with our tool".
Commenting on this project Mr. Mauro Pirrone, CTO of Minely and chief data scientist on this project explained that: "a large and diverse dataset is used to chart the fluctuating trends in this election. We have mined - with the help of Artificial Intelligence - almost 2,000,000 Facebook likes and almost 30,000 posts from across 13 news publishers. This has allowed us to find patterns within social data and extract sentiment preference for Malta's two largest political parties. This unique approach is coupled with a 'noise filter to allow us to exclude stories which are non-political and thus should be excluded from our sample".
This project is unique as it uses a large digital data set based on the public's likes, emotions and reactions, as opposed to the mainstream polls which occur with small numbers. Whilst the analysis is not intended to provide a forecast of the election winner, it can give useful insights into trends which emerge prior to that. The project focuses principally on following the activity of weak PN/PL loyalists who are known to determine the election outcome. The infrastructure used to run this project is based on Apache Spark which is the fastest engine available for large-scale data processing.
It is difficult for humans or pollsters to read the deluge of social content which exists and thus ICON and Minely have created a tool which analysis this information in an automated way based on machine-learning techniques trained on specific words, emotions or actions. Commenting on this, Dr. Gege Gatt stated: "It is evident that both parties are using social channels as a key element of their campaigns. It can be expected that social sentiment analysis such as this project will be part of every campaign in the future."
from The Malta Independent http://ift.tt/2rAvTW3
via IFTTT
No comments:
Post a Comment