• Journal of Technology Innovation
• /
• v.31 no.3
• /
• pp.139-169
• /
• 2023

This study analyzes the effects of firms' technology innovation(patent applications) on employment(number of workers and proportion of high-skilled workers) and corporate performance(sales per worker), while considering the two-way causal relationships between these variables. We used the three-stage least squares(3SLS) estimation to examine system of equations in which the dependent variables affect each other with a two-year lag wherever relevant, and applied it to firm-level panel data of Korean manufacturers with 100 or more workers. Our data covered the period of 2005-2017. Exogenous variables, such as firms' managerial and other characteristics, were controlled as explanatory variables. The identification variables for each equation included firms' R&D intensity, labor cost per worker(or operation of firms' own R&D center), and investment on worker training. We find that firms' patent applications increased number of workers, proportion of high-skilled workers, and sales per worker; the causal relationships in the opposite direction were also significant. Evidently, firms' technology innovation is critical to the growth and quality improvement of employment as well as sustainable corporate growth.

• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.251-258
• /
• 2003

A support vector machine (SVM) is supposed to provide a good generalization performance, but the actual performance of a actually implemented SVM is often far from the theoretically expected level. This is largely because the implementation is based on an approximated algorithm, due to the high complexity of time and space. To improve this limitation, we propose ensemble of SVMs by using Bagging (bootstrap aggregating) and Boosting. By a Bagging stage each individual SVM is trained independently using randomly chosen training samples via a bootstrap technique. By a Boosting stage an individual SVM is trained by choosing training samples according to their probability distribution. The probability distribution is updated by the error of independent classifiers, and the process is iterated. After the training stage, they are aggregated to make a collective decision in several ways, such ai majority voting, the LSE(least squares estimation) -based weighting, and double layer hierarchical combining. The simulation results for IRIS data classification, the hand-written digit recognition and Face detection show that the proposed SVM ensembles greatly outperforms a single SVM in terms of classification accuracy.