A machine learning model for predicting innovation effort of firms

Ruchi Rani, Sumit Kumar, Rutuja Rajendra Patil, Sanjeev Kumar Pippal


Classification and regression tree (CART) data mining models have been used in several scientific fields for building efficient and accurate predictive models. Some of the application areas are prediction of disease, targeted marketing, and fraud detection. In this paper we use CART which widely used machine learning technique for predicting research and development (R&D) intensity or innovation effort of firms using several relevant variables like technical opportunity, knowledge spillover and absorptive capacity. We found that accuracy of CART models is superior to the often-used linear parametric models. The results of this study are considered necessary for both financial analysts and practitioners. In the case of financial analysts, it establishes the power of data-driven prototypes to understand the innovation thinking of employees, whereas in the case of policymakers or business entrepreneurs, who can take advantage of evidence-based tools in the decision-making process.


classification and regression tree; data mining; innovation; innovation predictors; machine learning;

Full Text:


DOI: http://doi.org/10.11591/ijece.v13i4.pp4633-4639

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

International Journal of Electrical and Computer Engineering (IJECE)
p-ISSN 2088-8708, e-ISSN 2722-2578

This journal is published by the Institute of Advanced Engineering and Science (IAES) in collaboration with Intelektual Pustaka Media Utama (IPMU).