/Music-Recommendation-System

KKBox's Music Recommendation Challenge on Kaggle.

Primary LanguagePythonApache License 2.0Apache-2.0

ALL WE NEED IS Tree 🌲

该项目代码为本人首次参加 Kaggle 比赛的模型代码。参加的 Kaggle 比赛内容为音乐推荐系统。比赛的任务要求如下：

The main objective of the project is to predict the chances of a user listening to a song repetitively after the first observable listening event within a time window was triggered. If there are recurring listening event(s) triggered within a month after the user’s very first observable listening event, its target is marked 1, and 0 otherwise in the training set. The same rule applies to the testing set.

在比赛结束时，我所在的队伍在 Public Leaderboard 排名为 62 名，在 Private Leaderboard 排名为 59 名（参赛队伍为 1172 支）。

Requirements

  • Python 3.x
  • Numpy
  • CatBoost
  • XGBoost
  • LightBGM
  • GBDT
  • Libffm
  • g++ (with C++11 and OpenMP support)

Dataset

Get the research data

数据来源为 Kaggle 上的比赛数据集，内容为音乐推荐。

How to Kick the Ass 👾

Step 1 & 2: EDA & FE

Exploratory Data Analysis 以及 Feature Engineering 的工作部分在此不进行展开，如果对如何系统合理地处理数据感兴趣，可以参考我个人博客中的该篇文章「Music Recommendation Challenge」。

Step 3: Choose the Model

LIBFFM + GBDT

This model is called Field-aware Factorization Machines. If you want to use this model, please download LIBFFM first.

CatBoost

XGBoost

LightBGM

About Me

黄威，Randolph

SCU SE Bachelor; USTC CS Ph.D.

Email: chinawolfman@hotmail.com

My Blog: randolph.pro

LinkedIn: randolph's linkedin

Reference