| Autor: |
Sangli, Mohan, Kadadevaramath, Rajeshwar S., Joseph, Jerin, Kadadevaramath, Akarsha, Edinbarough, Immanuel |
| Předmět: |
|
| Zdroj: |
Journal of Mines, Metals & Fuels; Jun2023, Vol. 71 Issue 6, p857-866, 10p |
| Abstrakt: |
Predictive models are built by learning the combined effects of several independent variables that directly or indirectly influence the outcome. H. Response or dependent variable. In practice, data collection has data on a large number of independent variables that are outcome-sensitive and may or may not be related to the outcome. Some independent variables have a large impact on the results, while others may have little or no impact on the results. The presence of some independent variables that are irrelevant to the outcome can affect the performance of the predictive model. In this context, it is desirable and essential to identify the independent variables that most influence the forecast model to keep it lean and efficient. In this work, we used a dataset containing employee turnover rates and explored how to identify a subset of outcome-sensitive variables, thus eliminating variables that hinder the development of effective predictive models. By partially selectively influencing the independent variables, we developed lean and efficient predictive models that enabled us to act on an actionable subset of the variables to reduce staff turnover, thereby improving corporate save effort and cost. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
