Spotle AI-thon - The AI Global Challenge had 7000+ participants from best campuses in India, Singapore worked on addressing the mental health challenge with AI. Top 10 teams from IIT Roorkee, CMI, NIT, IIM Indore, Charotar University, DIAT made it to the final round. This is a showcase Top 10 presentation from Shivam Kumar Giri - NIT Puducherry
2. INTRODUCTION
PROBLEM STATEMENT
• COVID-19 is a humanitarian crisis on a global scale.
• Covid-19 pandemic has severely affected countries around the world.
• sudden outbreaks of such pandemics affect public mental states and emotions.
• results in either constructive or destructive behavioral changes among people.
• Anger, Sadness, fear are the most common emotions witnessed among the people during pandemics.
• Social media platform like Twitter and others have rich sources of information from people.
Here we are going to build a Machine Learning application which can understand emotions of people and classify it
based on their sentiments of tweets.
Study the twitter data to understand the emotions of people against Covid-19 and
classify the emotion based on tweets.
4. #corona, #covid19, #coronavirus are
among Leading Hashtags
HASHTAGS
Use of hashtags
increases with Time
Total 1360194 Hashtags Used
7675 Unique Hashtag
LOCATION
Maximum Tweet Location are from
India, Switzerland and UK & US
Total 3439 Locations of Activity
54,213 Tweets from India, Switzerland and UK
5. DATES
Number of tweets
increases with Dates
TIME
Maximum Tweets are
between 12 PM to 4 PM
Total 24 Hours of Activity
123,969 Tweets between 12 PM to 4 PM
Total 10 Days of Activity
86558 Tweets on September 21
6. MAJOR EMOTIONS
Emotion
Max Tweet of a
Location and Date
Remarks Date/
Location
Pattern with
total count &
‘Percentage %’
WORRY
Sep 21
32000+
India
22000+
212541 42.8%
Dates of the Worry tweets increases constantly
India clearly Emotion tops the list of Worry with 22000+ tweets
5000+ from India and almost same from Switzerland are being Neutral about COVID
NEUTRAL
Sep 21
26000+
India
5000+
133096 26.8%
Neutral Emotion peaks on 21 Sep in the Tweets
7. HATE
Sep 21
4800+
India
1000+
28991 5.8%
Emotion
Max Tweet of a
Location and Date
Remarks Date/
Location
Pattern with
total count &
‘Percentage %’
Hate Tweets are on its peak on 21 and 22 September
India leads the row with 1000 tweets in Hate Emotion, followed by
Switzerland
SADNESS
India
1100+
Sep 21
5500+
27667 5.6%
21st September is the day with Sad Tweets
India followed by Switzerland has the most sad sentiments in Tweets
SURPRISE
Sep 21
800+
India
1000+
26232 5.2%
Surprise tweets are on its exceptional peak with 800+ (which is double the
average of 300+) on 21st September.
India and Switzerland are leading in surprise emotion
8. CONCLUSION
Hence, we analyzed the given tweets
and try to carry out most of the Analysis with
different aspects of location, time, tweets, date,
hashtags and Emotions.
Data is well Visualized and interpolated
in various graphs, data is well studied and
concluded that Mental Health of Indian During
COVID is Well prone to fear, sadness and hate,
with a good amount to be neutral to COVID
Epidemic. Fear(Worry) can also be visualized as
top trending hashtags are related to corona or
covid-19.