Our AI Analytics Reports
We have been regaling you with our AI analysed studies that are all the rage on the internet. We deploy our AI algorithms to extract the most relevant insights around the topic of discussion.
This is an informative article on what we do and how we do it.
Every tweet is made in one context or the other. Our tool analyses the sentiment of the text in the tweets. The cumulative sentiment of all the tweets is calculated to give a final score that in turns becomes the overall sentiments of the issue on track.
The overall sentiment analysis gives you the simplest form of approval or rejection of the topic of discussion. If the majority sentiments are positive, the topic is favoured by the internet crowd or else it’s a red flag
For the concerned topic, two separate word clouds displaying the trending hashtags and keywords are generated. The word cloud contains the popular keywords and hashtags of different sizes and colours representing the weight and context of the topic.
The words in the green are positively spoken for while the red ones are the negative tale carriers.
Emotions should not be confused with the sentiments. The overall sentiments give the vague idea of the overall context of the topic. However, the emotions are very specific contextual metrics. For instance, happy, excited and even funny emotions are counted as positive. While angry, sad, fear would be negative sentiments.
As you can see in the image, 85.58%% of the total tweets have a happy emotion, 2.02% exhibit anger and so on.
The open nature of social media platforms lets anyone post anything with any intent. Not to be surprised that majority of the tweets are made with a marketing intention to drive engagement or can be random posts. The following bar graph segregates the intent into random, marketing, news, feedback, opinion. And, our insights are based on the posts with relevant intention.
For our next insight, we pull out the most popular tweets made on the subject. A tweet is deemed as popular on the basis of the re-tweets (when the tracking starts) and the likes it gets. All the tweets that are tracked are analysed on the this criteria returning a list of tweets sorted in descending order of popularity.
These tweets are classified into positive and negative using a classifier. Again, you get the list of positive and negative tweets sorted in descending order of popularity.
The influencers are like the twitter celebrities who contribute the maximum to the topic of discussion. The twitter users whose tweets gain hefty likes and are significantly re-tweeted become the influencers. Different people can be influencers for different topics.
Since, tweets have the context of their own, the influence can be positive or negative. And, so can be the influencers. Hence, we have separate categories for positive and negative influencers.
This was all about the standard insights we dig out through our AI analytics tool Karna. We also create custom studies when required so. If you are looking for getting a customized study done, drop a line at [email protected]
Karna.ai is a division of ParallelDots. Karna is a social media marketing research platform. We collect and analyse millions of mentions from News, Twitter, Facebook and Instagram and deliver AI driven in-depth insights through automated reports and custom analysis.
Want to get deep consumer insights. Click here to schedule a free demo.