Introducing 

Prezi AI.

Your new presentation assistant.

Refine, enhance, and tailor your content, source relevant images, and edit visuals quicker than ever before.

Loading…
Transcript

Music Recommadation

Presented by

Neha shikha 173050080

Abhijit Patil 173050085

Kuber Gautam 173050087

Introduction

Kaggle Competition

WSDM - KKBox's Music Recommendation Challenge

Introduction

Problem

Problem Statement

More

To predict the chances of a user listening to a song repetitively after the first observable listening event within a time window was triggered.

Data Description

Data

  • KKBOX, Asia’s leading music streaming service
  • first observable listening event for each unique user-song pair within a specific time duration.
  • Metadata of each unique user and song pair is also provided.
  • More than 77Lakh of entries

Lightgbm

Techniques

LightGBM is a gradient boosting framework that uses tree based learning algorithms. It is designed to be distributed and efficient with the following advantages:

  • Faster training speed and higher efficiency
  • Lower memory usage
  • Better accuracy
  • Capable of handling large-scale data

Approaches

Approach

  • Merging
  • Handling Null values
  • Training (Gradient Boosting Decision Tree)
  • Testing on Kaggle

Challenges

  • Big Data
  • Chronological Ordered Data
  • Muti Attribute Genre
  • Feature Extraction

more

Result and Conclusion

For LightGBMs, Our best single model is 0.671 on public with a learning rate of 0.3

Conclusion

Things to learn

  • Max Depth , num_leaves

(Training Error less, Test Error More--> Overfitting)

  • Learning Rate

Learn more about creating dynamic, engaging presentations with Prezi