Community

I dont see any such inference even from public models. However, such insights even though can have a basis that cannot be given much weigt in a decision in traditional underwriting. Moreover, getting indicators that usually is not possible through AI/ML is a reality already. These help push the models far left to the value chain and assist in taking better decisions like an Income estimation.

The first Data Quality challenge is most often the acquisition of right data for Machine Learning Enterprise Use cases.

Even though the business objective is clear, data scientists may not be able to find the right data to use as inputs to the ML service/algorithm to achieve the desired outcomes.

As any data scientist will tell you, developing the model is less complex than understanding and approaching the problem/use-case the right way. Identifying appropriate data can be a significant challenge. You must have the “right data.”

More broadly speaking, Coverage, can be categorized under the Completeness Dimension of Data Quality and called the Record Population concept within the Conformed Dimensions standard. This should be one of the first checks to be performed before proceeding to other Data Quality checks.

Niall, Good thoughts articulated well! What caught me in your article is the phrase " To do this, certain protocols and rules need to be put in place to ensure good governance over this process." To ensure this we need to progress on Governing the data in the cloud as well. Your further statements clearly bring out the classifications of data to be hosted in accordance to their risk profile.

This fundametally requires a change to approach in which data is logically and physically classified. http://www.dataversity.net/integrate-data-privacy/ is an article that briefs on how we can define requirements for data to be hosted on various platforms including cloud.