Okay, so today I’m gonna spill the beans on my little project: ‘wwe ava’. It’s not what you think, hahaha. It’s actually a personal project involving some data wrangling and visualization. Let’s dive in!
Phase 1: The Data Hunt
First thing’s first, I needed data. I spent hours scraping different sites, wrestling sites mainly. Looked at match results, wrestler bios, social media engagement, all that jazz. It was messy, trust me. Different formats, inconsistent naming conventions – a total nightmare. I even messed around with some APIs, which were a bit of a headache to get working right.
Phase 2: Cleaning Up the Mess
Then came the glorious task of cleaning. Oh boy. I started with Python and Pandas. If you haven’t used Pandas, it’s a lifesaver for this kind of stuff. I standardized names, handled missing values (a lot of those!), and converted data types. Like, dates were all over the place. Some were strings, some were numbers, a total mess. I had to wrangle ’em into a consistent format.
Phase 3: Feature Engineering – Let’s Get Creative
Next up, feature engineering! This is where things got fun. I started calculating things like win percentages, average match duration, social media follower growth. I even tried to calculate a “hype” score based on social media mentions before and after matches. Some of these were total flops, but some turned out to be pretty interesting.
Phase 4: Visualization Time!
Now for the eye candy. I used Matplotlib and Seaborn for the visualizations. Nothing fancy, just basic bar charts, line graphs, scatter plots. I wanted to see trends, correlations, outliers. For example, I looked at how win percentages correlated with social media followers. Surprisingly, not always a strong correlation. Some wrestlers were just good at building their brand, regardless of their in-ring record.
Phase 5: Iteration and Refinement
This wasn’t a one-and-done deal. I kept going back and forth, refining my data cleaning, tweaking my feature engineering, and experimenting with different visualizations. I noticed some weird anomalies, dug deeper, and found even more inconsistencies in the data. It’s a never-ending cycle, really.
Key Learnings:
- Data is always messier than you think. Always.
- Pandas is your best friend. Learn it, love it.
- Don’t be afraid to experiment with feature engineering. Some will fail, some will surprise you.
- Visualization is key to understanding your data.
What’s Next?
I’m thinking of trying some machine learning stuff next. Maybe predict match outcomes or something. But honestly, I’m mostly just having fun exploring the data and seeing what I can find. It’s a good way to learn and improve my skills.
