Data Science and Analysis ,Computer Science

Creating the Final deliverables

1. DATA SCIENCE AND ANALYTICS Submitted by , Malarvizhi V II Msc .,cs Nadar Sarawathi college of arts and science

2. INTRODUCTION TO DELIVERABLES : Deliverables must be defined during the project planning phase the project scope forms part of a documents known as project charter which also list the projects objectives and stakeholders in the addition to the project charter you may want to create a work breakdown structure also known as WBS Deliverables can includes the final result of an initivate and the individual execution required throught various project stages to produce completed work these deliverable can be internal or external Tangible or intangible depending on the project scope and its definition of success

3. Creating the final deliverables Some project components have dual use Create core materials used foe both analyst and business audiencess Final deliverables Project Goals Approach Model Description Key points Supported by Data Model Details Recommendation Additional Tips on the Final Presentation Providing Technical Specification and code

4. Project Goals The project goals portion of the final presentation is generally the same for sponsors and analysis The project goals are described first to lay the groundwork for the solution and recommendation Generally the goals are agreed on early in the project

5. Approach(for Sponsors) Developed chum models to identify customer most likely to leave the bank Identify most influential factors Provide greater explanatory power for analuzing impact of different factors on churm Minded and added social media data to the model to improve predictive power

6. Development chum model in R using a Generalized Addictive Modeling Technique Minimize variable transformations and binning provide greater explanatory power for analyzing impact of different factors on churn Collaborate with IT to Identify relevant data set and assess data quality and availability Approach (for Analysts)

7. Key points Supported with data Identify key points based on insights and observation from the data and model results This information lays the foundation for the coming recommendations

8. MODEL DETAILS: Model details comparing two data variables

9. Recommendations Implementation the model as a pilot before more wide scale rollout Addressing these promptly can potentially save more customer from churning over time and also prevent more networking that seems to drive additional churn Run the predictive model daily In database scorer can score large datasets in a matter of minutes and can be run daily Develop targeted customer surveys to investigate the causes of churn Which will make the collection of data for investigation into the causes of chru easier

10. Use Imaginary and visual representation Picture are better than words Ensure Text Mutually exclusive or collectively exhaustive Over key points dont repeat unnecessary Measure and Quality the benefits of the project Required effort to do this well Make a project benefits clear and conspicuous Details Provide Sufficient context for recommendation spell out acrohyms avoid excessive jargon Additional Tips on Final Presentation

11. Providing Technical Specification and code Delivery code plus documentation user manual Add extensive comments in the code How computationally expensive to run the model ? Describe expecting handling Data outsides expected data ranges Null values Zeros

12. Conclusion we hope the teams that continue working can improve their effectiveness by utilizing our recommendations and arifacts Code Use manual Spotlight demo recording Knowledege transfer sessions Technical Tranning Confluence documentation Final demo and Q&A

13. THANK YOU