# Data Project 2, Election Prediction : A Data Project Will Definitely Get You a Job Offer

In 2024, More Than 60 Nations to Hold National Elections.

Created with Abidin Dino AI, available for paid subscribers on substack, here.

2024 is an election year because more than 60 Nations, including the United States, will hold national elections.

This holds great opportunities for you if you are pursuing a related career. Because the majority of the world population will vote this year, and if the recruiter sees this project on your portfolio, I assure you he/she will look at your CV in detail.

And that will give you the opportunity to show yourself. Now, in this one, I’ll direct you to the steps to do this data project.

# Step 1 Finding Dataset

I highly suggest you scrape this dataset through Twitter by using Twitter API. If you don’t know how to use it, this might be your chance to develop yourself. I assure you, it is very easy. Watch this one if you want to learn it :

However, if you are new and don’t want to learn this one, you can do your data projects using this Kaggle dataset.

#Step 2 Data Exploration

Here is just one data that we’ll use. It contains 970K tweets collected from Twitter.

We have 2 different datasets; one is the tweets about Trump, and the other one is for Biden.

We notice how the datasets for Trump and Biden are structured, focusing on key columns like tweet content, likes, and retweets, which might hint at public sentiment. We also observe geographical data like user locations, providing a glimpse into the diverse origins of these tweets.
Interestingly, we consider the potential of this data to unlock insights into public opinion during the election.

#Step 3 Data Preparation and Sentiment Analysis

Here, you should make your data ready to do sentiment analysis.

This includes;

Preprocessing step

• To clean the tweet

• Remove regular expressions

• Changing it’s type

Then comes the exciting part: we employ sentiment analysis, assigning each tweet a positive, negative, or neutral score. The anticipation builds as we prepare to infer public sentiment towards each candidate—will the tweets lean towards optimism or skepticism?

#Step 4 Data Visualization

This is the most important step, along with sentiment analysis, because the visuals you’ll create will be the face of your data project.

I spent most of my time creating these data visualizations because the tweets are from 2020', so there is no chance I can predict this result correctly for 2024 because ideas can change.

However, creating impressive visuals will give you a heads-up in your job interviews as data-related jobs. I’ve created the following map with folium.

But to reach the map seen in the news, you only need to paste state names for each state.

One hint: You can also use geopandas to locate states or folium libraries.

Final Thoughts

Now, this will be our second data project. If you struggle with any steps, paste my explanation to the ChatGPT and ss; it will help you do this correctly.
Here is the pdf of this project.

Thanks for reading!