Skip to content

EDA analysis and a couple of models from classical machine learning on actual data as of 12.07.2023 about video games. Dataset from kaggle link in readme.

Notifications You must be signed in to change notification settings

Ch3ekiBr3eki/popular_videogames

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 

Repository files navigation

popular_videogames

python kaggle

About

This project aimed to examine the relationship between game ratings and the availability of games on the PC platform. The dataset included attributes such as game title, release date, developers, summary, platforms, genres, and ratings. The analysis revealed several interesting findings:

  • Games added to wishlists tend to make it to the top, indicating their popularity.
  • Reviews are more frequent than wishlist additions, suggesting a weaker correlation between game lists and reviews.
  • Highly rated games are more likely to be in backlogs, possibly due to purchases during sales.
  • The number of reviews and wishlist additions correlates with the number of players.
  • As player count increases, more games are bought but not played, possibly due to impulse buying or sales.
  • The correlation between backlogs and wishlist additions may be attributed to anticipation.
  • Game ratings have a slight correlation with wishlist additions, with a significant jump at the 4.5 rating.
  • Machine learning models were trained to predict game ratings, with the Lineral Regression model performing the best.

Целью данного проекта было изучение взаимосвязи между рейтингами игр и их доступностью на платформе PC. Набор данных включал такие атрибуты, как название игры, дата выхода, разработчики, краткое описание, платформы, жанры и рейтинги. Анализ показал несколько интересных результатов:

  • Игры, добавленные в списки желаний, как правило, попадают в топ, что свидетельствует об их популярности.
  • Обзоры встречаются чаще, чем добавления в списки желаний, что говорит о слабой корреляции между списками игр и обзорами.
  • Игры с высоким рейтингом чаще попадают в бэклоги, что, возможно, связано с покупками во время распродаж.
  • Количество обзоров и добавлений в список желаемого коррелирует с количеством игроков.
  • С увеличением числа игроков все больше игр покупается, но не играется, возможно, из-за импульсивных покупок или распродаж.
  • Корреляция между количеством отзывов и пополнением списка желаний может быть связана с ожиданием.
  • Рейтинги игр слабо коррелируют с пополнением списка желаний, причем значительный скачок наблюдается при оценке 4,5.
  • Для прогнозирования рейтингов игр были обучены модели машинного обучения, при этом наилучшие результаты показала модель линейной регрессии.

Developers

Kaggle link

About

EDA analysis and a couple of models from classical machine learning on actual data as of 12.07.2023 about video games. Dataset from kaggle link in readme.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published