Twitter Feedback Segregation System performs the analysis on the feedback and opinions given by the customers all over the world regarding any business product by segregation of Twitter mentions according to their positivity or negativity. After successful processing and performing sentimental analysis on the twitter mentions, it displays the analysis in the form of classification of positive and negative tweets along with graphical analysis of the same. At last, it enables the user to download the final report which contains detailed analysed data performed by the system.
Twitter allows businesses to engage personally with consumers. However, there’s so much data on Twitter that it can be hard for brands to prioritize mentions that could harm their business. We wanted to create a web-application where companies could perform sentimental analysis on their brands/products. Carefully listening to voice of the customer on Twitter using sentiment analysis allows companies to understand their audience, keep on top of what’s being said about their brand – and their competitors – and discover new trends in the industry.
- We would start implementing on a smaller scale. Twitter's Tweepy API allows us to mine all data there is, but under the duration of 1 week. Whatever we do next, would be constrained within this time period.
- For Multiple Weeks of data we would require large storage and high computing power. Once satisfied with our outputs, we will try to expand it on a larger scale.
- Our Application would work for any small or large company, brand or organisation.
- The applications should segregate tweets into positive and negative feedbacks.
- Graph should be shown according to the feedback received.
- Data should be available to the user in a downloadable format.
- Must Be a Responsive Web Based Applications
The website welcomes you with a home page. The first thing your eyes would be attracted to would be the title, 'Twitter Feedback Segregation System'. We hand-coded animations on each page to make our website seem visually appealing. Nav bar turns into a Hamburger Menu when you access the site using a smartphone. 'Get Started' button leads you to the Search Page.
The core of our website, the Search page, simply requires two inputs from the user; the Mention you need to search and Number of Tweets you need the website to process.
One fantastic functionality of this website is it doesn't wait for all the tweets to be processed at once and then finally give you your output. The process is quite Dynamic once atleast 100 tweets are fetched and you are presented with a loader view at the top to keep a track of how many tweets have been processed upto now.
As the tweets are fetched dynamically, our Bar Graph adjusts itself. Graph is a display of sentimental analysis that has been performed on the fetched tweets. The user can observe whether the Twitter Mention (s)he has searched for has a Negative, Positive or a Neutral Bias. At the bottom an Analysis Report would be made available to be downloaded for the user.
If you fancy looking at the demography of the users using this particular Mention in their tweets, you are also presented with a World Map.
Below the Map View, the user can have a look at the 'Top 5' Negative and Postive Tweets, out of the many tweets the website just processed. This would give our user an idea about what the customers/users usually appreciate and what they dislike.
For Privacy Reasons The Team Page has been made Unavailable.
Twitter Feedback Analysis is about knowing the overall feedback and opinions given by the customers all over the world regarding any business product by segregation of Twitter mentions according to their positivity or negativity. This system would enable the organization or company to know the responses of the customers towards their products, their advantages and disadvantages, and how well their products are accepted by the society which would help them to increase their scope of improvement and profits for a particular product.
I would like to thank my fellow members who helped with this project's growth and success.
Below you would find their Github Profiles: