Covering Disruptive Technology Powering Business in The Digital Age

A driven entrepreneur with a passion for Data Analytics. That would best describe the Managing Director at Malaysian born company, Berkshire Media SDN BHD, En. Shahid Shayaa.

Big Community had the pleasure of a quick catch up with En. Shahid here at MDEC headquarters in Cyberjaya and have a one-on-one interview before he was whisked off to his next appointment.

In his recent article titled ‘Social Analytics Powers Highly Accurate Election Result Prediction’, En. Shahid made heads turn with the results from the company’s Data Analysis-As-A-Service (DAAS), when he was able to achieve an accuracy rate of 97% in his prediction.

The company embarked on an opportunity to predict a by-election result for a small constituency of 42,000 voters. When the official results were out, the difference between actual vs. predicted results was 97%.

“Since there is no magic bullet in predicting election results, the best representation of a predictive outcome (i.e. probability of it happening) will fall under a best case, worst case and base case scenario. For the by-election predictive modelling, we predicted the incumbent will win 37% (worst case), 52% (base case) and 57% (best case)”, he shares in his article.

Although his team comprise of mostly Social Data analysts, the data that is scraped together can come from social media posts, tweets, blogs, and any other social interface easily accessible and frequently accessed by the public.

The analysis he says, uses behavioural psychology of the 6 basic emotions. Joy, love, anger, sadness, surprise and fear. It goes further in-depth by looking at the different words and phrases used to describe those emotions. “For languages like mandarin, we need to understand where they are coming from because the language is based on phrases that describe the emotion and not just on single words”. They also take into account dialects from different states, colloquial terms, current trending short forms and also the location from where the posts are made.

When asked how does he approach his research on social media with issues around privacy, he was quick to inform that only posts made public are used and no privacy will be breached. Since only 2-5% fall into the category of people who have private profiles, that number does very little to affect the accuracy of the findings as was shown.

“Everyone wants to be connected these days, especially the younger generation.”

Is there a possibility to pick up incorrect sentiment from social media? En Shahid says it comes down to the nuances. With the detailed analysis that they incorporate into their data, together with experts with industry know-how, the margin for error is very limited.

Currently the company is focused on DAAS. However they have plans in the pipeline to venture into having their own Data Analytics software within 2 years. He shared this when asked about their software offerings.

Do watch this space as Big Community will be keeping close tabs on En Shahid and Co. in the coming weeks.

(0)(0)