Please enable JavaScript.
Coggle requires JavaScript to display documents.
Ch. 20 and 22 (Ch. 20 Communicate Model Insights ( Features that need to…
Ch. 20 and 22
Ch. 20 Communicate Model Insights
External data: While it is not clear that mining public external data will be a worthwhile exercise for this project, it is likely that this project could benefit from purchased data such as grocery store purchase data, financial data, distance to hospital, and so on
the discussion turns to different types of features that are useful in model evaluation and presentation
Internal data: In this case, it could be argued that more data on past visits should be collected for patients that are frequent visitors
Often information that helps a model make high-quality decisions is not helpful in changing practice
It is recommended to respond by stating that delving into foundational questions will not leave enough time remaining for the presentation and might not realistically help the audience follow the information displayed
There are four kinds of features to consider before going into a management presentation:
These questions represent a real problem because not answering them may suggest that one doesn’t actually know the answer, while answering them properly will take the entire time allotted for the presentation
Features that need to be changed and therefore require a re-run of the models
Features requiring further examination
Immutable features
Mutable features
models. Even with decades of experience, seasoned data scientists continue to encounter difficult lines of questioning and skepticism from their audiences
Implement the model to find the 62% of patients most likely to be readmitted
Business Problem
Model quality metrics (confusion matrix)
Areas where a model struggles
Most predictive features for model building
Feature types especially interesting to management
Recommended business actions
Institute a one-month pilot program targeting five percent of patients discharged to home
There are six types of information that should be communicated during a presentation:
Institute a data-extraction-and-purchase pilot program to explore how the patient readmission model can be improved and to what degree
Ch. 22 Document Modeling Process for Reproduction
The next step in documentation is to specify all the steps taken within DataRobot. In this case, that means not using the advanced options, but rather taking the following steps: using LogLoss as the optimization criterion, using the autopilot to build the models, selecting the best model based on cross validation, and running that model with 100% of the data
Assuming that such a system is successfully in place, the next step is to specify the business problem in the project documentation. This step should be easy, as this has already been done at the beginning of the machine learning process
Consider model documentation as an opportunity to do more of the desirable central machine learning work
By attending to the details of articulating project processes while a project remains fresh in-mind, an analyst can save future use of time on the project
proper documentation is critical for others to understand what actions were taken to accomplish project results, as well as the justification for the project to exist
Documenting the modeling process is where projects most often fail
Work under the assumption that the project will need to be revisited within a year, and think through what information would be helpful to have immediate access to at that time