The Photometric LSST Astronomical Time-series Classification Challenge

Data

Pour obtenir les données vous devez posséder un compte sur Kaggle et installer l’outils en ligne de commande kaggle-api. Ensuite, entrez la commande suivante dans votre terminal:

kaggle competitions download -c PLAsTiCC-2018

References

• https://arxiv.org/abs/1810.00001
• https://plasticcblog.wordpress.com/
• https://github.com/LSSTDESC/plasticc-kit
• https://galactica.isima.fr/

Business understanding

Background

Business objectives and success criteria

Inventory of resources

Requirements, assumptions and constraints

Risks and contingencies

Terminology

Costs and benefits

Data mining goals and success criteria

Project plan

Initial assessent of tools and techniques

Data understanding

https://github.com/yafeunteun/kaggle-plasticc-astronomical-classification/tree/master/data-understanding

Data preparation

Dataset description report

Background including broad goals and plan for preprocessing

Rationale for inclusion/exclusion of datasets

For each included dataset:

• Description of the preprocessing, including the actions that were necessary to address any data quality issues
• Detailed description of the resultant dataset, table by table and field by field
• Rationale for inclusion/exclusion of attributes
• Discoveries made during preprocessing and anu implications for futher work
• Summary and conclusions

Modeling

Modeling asumption

Test design

Background - outlines the modeling undertaken and its relation to the data minig goals

For each modeling task:

• Broad description of the type of model and the training data to be used
• Explanation of how the model will be tested or assessed
• Description of any data required for testing
• Plan for production of test data if any
• Description of any planned examination of models by domain or data experts
• Summary of test plan

Model description

Overview of models produced

For each model:

• Type of model and relation to data mining goals
• Parameter settings used to produce the model
• Detailed description of the model and any special features (see p. 66)
• Conclusions regarding patterns in the data (if any);

Summary of conclusions

Model assessment

Overview of assessments process and results including any deviations from the plan

For each model:

• Detailed assessment of model including measurements such as acuracy and interpretation of behavior
• Any comments on models by domain or data experts
• Summary assessment of model
• Insights into why a certain modeling technique and certain parameter settings led to good/bad results
• Summary assessment of complete model set

Evaluation

Assessment of data mining results with respect to business success criteria

• Review of Business Objectives and Business Success Criteria (which may have changed during and/or as a result of data mining)
• Review of Project Success; has the project achieved the original Business Objectives?
• Are there new business objectives to be addresses later in the project or in new projects?
• Conclusions for future data mining projects

Review of process

List of possible actions

Deployment

Deployment plan

Summary of deployable results

Description of deployment plan

Monitoring and maintenance plan

Overview of results deployment and indication of which may require updating (and why)

For each deployed result:

• Description of how updating will be triggered
• Description of how updating will be performed

Summary of the results updating process

Final report

• Summary of Business Understanding: background, objectives and success criteria.
• Summary of data mining process.
• Summary of data mining results.
• Summary of results evaluation.
• Summary of deployment and maintenance plans.
• Cost/benefit analysis.
• Conclusions for the business.
• Conclusions for future data mining.