PL-PN neck and neck in public sentiment on social media

The Malta Independent has teamed up with Maltese award-winning technology company ICON and Minely, a European start-up founded in 2013, with a mission to empower businesses to simplify Big Data processing to easily build data-driven solutions using leading-edge technologies for business analytics with timely and relevant decision-making actions.

This initiative intends to provide readers with a unique big data analysis of how people are interacting on social media as regards the 2017 electoral campaign. This collaboration was possible thanks to The Malta Independent’s IT developers Deloitte Digital, formerly known as Alert Communications.

This Big Data project by ICON and Minely analyses trends in social media activity across major channels in Malta. It provides an insight into how online-users are interacting with the election 2017 coverage they read.

A quick run-through of the analysis so far provided some interesting insights. The Big Data Project came into effect in February 2017. When analysing the way people were reacting to political posts on social media back in February the PL scored 63% in public sentiment while the PN trailed behind at 37% of public sentiment towards political posts. This includes mainly ‘likes’ and ‘shares’ count among other factors.

In March engagement was somewhat less for both parties however the PL peaked at 71.1% while the PN had sentiment reactions as low as 28.9%. However in April activity amongst political posts on social media began to pick up considerably with the PL dropping to 61% of public sentiment while the PN achieving 39% of engagement.

Towards the end of April, when the Egrant revelations came to light, activity skyrocketed with PN surpassing the PL for the first time since the project went live generating public sentiment at 55.2% while the PL registered 44.8%.

Come May with the announcement of a snap election in a month’s time and during the first week of the campaign the PN hit 39.8% while the PL gained ground again to peak at 60.2% of public sentiment.

All through the second week of the campaign the PL retained a steady lead over the PN in public sentiment to political posts on social media. However during the third week of the campaign a tide is clearly turning in public sentiment towards political posts with the parties running neck and neck as regards public sentiment. The PN is currently at 52% while the PL stands at 48% of public sentiment.

It should be amply clear that this tool cannot predict who will win the general election in Malta but it provides an insight into activity and engagement across social channels, which carry a bearing on how people express their feelings.

It cannot substitute scientific surveys such as the iSurvey conducted by Business Leaders for The Malta Independent, yet it provides insight into how people really feel, something they can skew in a survey by simply not saying the truth. The validity of The Big Data Project lies in the fact that it gathers real data on how people express themselves publicly.

Amongst the features provided in The Big Data Project one can find analysis over 25,000 articles and over 2 million reactions automatically. Clustering is used to determine whether a post is nonpolitical or to classify as PL or PN. This model utilises the political parties’ social pages as initial sentiment signal for building the predictive model.

The resulting model is used to automatically classify articles from all portals’ including independent media. A machine learning model represents a realworld scenario. It abstracts from the real relationships by simplifying variables which are not too important.

This simplification process causes a margin of error which is considered as noise.

Minely offers a full range of social connectors including Facebook, Twitter, Instagram etc. Using a Facebook Application Id, a connection was made to the social network. Minely retrieves public data periodically from the social pages of each news portal. Using a combination of natural language processing and machine learning data is extracted and analysed into Minely.