Twitter data is one of the most openly available data-set for analysis. But in the current scenario of analysis, twitter data and hashtag analytics arent enough to give the bigger picture of how a particular topic effects a given political biome. We hence wanted to grow this said search to include a more representative picture by including multifaceted information, through statistics, and hence showing a greater solution to any given problem.
What it does
Our project combines sentiment analysis using AI and various other data environments. The way our data is layered, we can see an infinite possibilities of analysis, which is very customizable by any users. Hence, we wanted to give out a product which can be used by any particular individual, to gain better and accurate insights of any given search. This project is built uses openly available tools and datasets, by using machine learning to enrich the data, to merge these data.
How we built it
Twitter Data *Using openly available twitter data-set (5 million tweets) we indexed them at scale, using elastic search. *We calculated sentiment of every tweet using machine learning model Consensus Data *Acquired various political and statistical data, of US
Merged both these data-facets using geo-spatial arrangement of real time mapping
Challenges we ran into
Finding recent and authentic datasets Indexing at scale takes a lot of time and infrastructure, for personal PCs Understanding Visualization libraries, for users to easily understand
Accomplishments that we're proud of
We acquired useful insights that are multifaceted. The impact is realized to be across industry and public use. Hence we consider our project to be of a greater usage through elastic consumption.
What we learned
Team work Various visualization requirements How to have fun in tough times
What's next for BeyondTwitter
To include time progress, to show the growth of a given tweet. Include multilayer emotional analysis. Make it real-time GET MORE DATA!! Include this product as a inter-industrial solution