Inspiration
We are a team of four graduate students who are interested in data science and data visualization. It was interesting and challenging for us to experience this data visualization event. We’ve recently become familiar with Tableau and wanted to assess our ability to work with it. We were thrilled by the challenge of creating a data visualization in a short time during the hackathon. The fast-paced and dynamic nature of hackathons, where quick thinking and rapid development are key, motivated us to join in.
What it does
We tried to analyze the dataset and also got the actual data from social media to see the scale of China’s state media presence and its effect on social media platforms. We reached interesting results, and we visualised them using Tableau and some Python libraries. Also, we deployed a machine learning model that classifies tweets given to it with 96% accuracy.
How we built it
We analyzed several different aspects of the given dataset. Additionally, we used scraping libraries (Selenium) using Python to extract more information about accounts on different platforms, especially Twitter. Then we used all of that data to create interactive visualizations using Tableau and Python. In the next part, we tried to cluster tweets into real and fake and try to differentiate them.
Challenges we ran into
Time Constraints: Managing time effectively to meet the hackathon deadline. Limited Data: the provided dataset was relatively small, with some limited information about 759. We needed to gain more data using those limited accounts and URLs. Tweet scraping: the most important challenge we ran into was doing the scraping of tweets because of heavy restrictions on Twitter APIs.
Accomplishments that we're proud of
- Making a webpage illustrating the plots and figures interactively 2. Training the machine learning model with good accuracy 3. Deploying the ML model on the server and making a user-friendly UI for users to test it easily and unlimitedly 4: Accomplishing the task of doing more than 10 analyses and explaining the interpretations about them in detail.
What we learned
This project helped us to investigate and learn more about data analysis and visualization, approach designing a user-friendly and intuitive interface, and improve our teamwork skills and time management. Also, as we went through the accounts or tweets, we noticed more odd points about them and accordingly made some assumptions. After doing the analysis, we realized our assumptions were true, and the visualizations proved so many interesting facts about China's media. This process was so engaging, making us dive more into the details and explore for more odd facts.
What's next for China's impact?
During this hackathon, we gained lots of technical and international knowledge, which made us more enthusiastic about data analysis. We will definitely try to accomplish more achievements in this field and gain more skills for future challenges.
Log in or sign up for Devpost to join the conversation.