Stat441-Final

Final presentation recources

Description : The link of the competition : https://www.kaggle.com/c/kkbox-churn-prediction-challenge

Theme and purpose : KKBOX, the leading music steaming platform in Asia, is looking for a new algorthim that determines if a user would continue to churn his/her current sub

Current methodology : Survival analysis techniques to determine the residual membership lifetime, in other words, if a particular user would continue his/her sub after it expired, the residual membership should be relatively long

Evaluation : Based on log loss model on page : https://www.kaggle.com/c/kkbox-churn-prediction-challenge#evaluation

Outputs : Two columns : the first column for user ids, which can be directly approached from the datasets(excel sheets)

the second column for a predicted probability that the user would continue the sub, value ranges from 0 to 1

Version 1.0 , editted on Oct 30, 2017 by Michael C Dong

Name		Name	Last commit message	Last commit date
Latest commit History 6 Commits
.gitattributes		.gitattributes
.gitignore		.gitignore
Copy of Proposal_Classification_F17.pdf		Copy of Proposal_Classification_F17.pdf
README.md		README.md
Statistical Approach 1		Statistical Approach 1

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

Stat441-Final

Final presentation recources

Description : The link of the competition : https://www.kaggle.com/c/kkbox-churn-prediction-challenge

Theme and purpose : KKBOX, the leading music steaming platform in Asia, is looking for a new algorthim that determines if a user would continue to churn his/her current sub

Current methodology : Survival analysis techniques to determine the residual membership lifetime, in other words, if a particular user would continue his/her sub after it expired, the residual membership should be relatively long

Evaluation : Based on log loss model on page : https://www.kaggle.com/c/kkbox-churn-prediction-challenge#evaluation

Outputs : Two columns : the first column for user ids, which can be directly approached from the datasets(excel sheets)

the second column for a predicted probability that the user would continue the sub, value ranges from 0 to 1

Version 1.0 , editted on Oct 30, 2017 by Michael C Dong

About

Uh oh!

Releases

Packages

Uh oh!

Contributors

Uh oh!

Folders and files

Latest commit

History

Repository files navigation

Stat441-Final

Final presentation recources

Description : The link of the competition : https://www.kaggle.com/c/kkbox-churn-prediction-challenge

Theme and purpose : KKBOX, the leading music steaming platform in Asia, is looking for a new algorthim that determines if a user would continue to churn his/her current sub

Current methodology : Survival analysis techniques to determine the residual membership lifetime, in other words, if a particular user would continue his/her sub after it expired, the residual membership should be relatively long

Evaluation : Based on log loss model on page : https://www.kaggle.com/c/kkbox-churn-prediction-challenge#evaluation

Outputs : Two columns : the first column for user ids, which can be directly approached from the datasets(excel sheets)

the second column for a predicted probability that the user would continue the sub, value ranges from 0 to 1

Version 1.0 , editted on Oct 30, 2017 by Michael C Dong

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Packages