Hello! I'm Daalma, and I love Pokémon. As a Data Scientist, I've been working on this project in my spare time. It's something I hope reflects my love for the series and that others as passionate as I am will find interesting or appealing.

This is a complete Data Science project with three main objectives:

1: Generation of a dataset using web scraping containing information about all Pokémon (up to Generation IX), including variants and forms.

2: Preprocessing the dataset, extracting basic information, and creating informative visualizations.

3: Applying Machine Learning and AI techniques to generate higher-level insights and visualizations.

You can check out the project here: https://github.com/Daalma7/PokemonDataScience

The results of the project have been quite good, and while I reserve the right to have made mistakes, I must say I’m really pleased with the graphics and outcomes. If anyone wants to take a look and share their thoughts, I would be very grateful. Below are some images showing a sample of what I've done.

Thank you so much for reading!

Daalma

Posted by Daalma7

6 comments

  1. Id honestly give your project an A this is awesome

    Sending this to my friend

  3. Please read Rule #3 and post a **comment** on this thread clearly providing the tools used and the sources of the data. Thanks!

  4. Amazing work!

    Also in the readme file the heading formatting seems to be messed up

  5. The data source is Bulbapedia, web scrapping tools were used using Python and BeautifulSoup, and i extracted the data in csv format, the data can be consultes in the github link provided.

    For the tools I used Python (Jupyter Notebooks) as well as libraries such as pandas, numpy, metas, tensorflow, marplotlib, seaborn, plotly, markdown and seaborn. Finally, all files used are in the github link provided as well 🙂

  6. In “Pokemon type share relationships” you can see that thick branch connecting Fire and Fighting – that represents generations of GameFreak ragebaiting the fans lol

    The interconnectivity between Dragon and Ground surprised me, but Zygarde’s various forms explains the apparent significance
    here.

    Then there’s Grass/Poison. Kanto really was a thing. And Normal/Flying makes a wing-shape!

    Can you explain “Principal Components” to me? I’ve got guesses but I can’t figure out why Litwick would score lower than, say, Weedle.

    The hierarchical clustering graph is a complete mystery to me. Maybe if the images were in a higher resolution I could distinguish a pattern.

    Pretty neat stuff; mostly over my head.

Comments are closed.