Discovering firm technology opportunities by machine learning prediction method based on firm's technology portfolio

Abstract

Identifying technology opportunities is crucial for firms to maintain competitiveness and drive innovation. However, existing methods often overlook the interaction between external technological trends and firm's internal capabilities. To address this gap, this study proposes a novel approach for identifying firm-specific technology opportunities. Our methodology leverages a technology convergence network and the connectivity between nodes to identify a firm's portfolio. These candidate technologies in firm's portfolio are then evaluated using a systematic set of indices that measure both their external attributes in the domain and their internal relevance to the target firm, assessed by social network and link prediction. The Support Vector Machine (SVM) algorithm is then dynamically applied to learn from the features of past technology opportunities and predict future promising opportunities among the candidate technologies. We apply this approach to the wind energy sector and the largest firm in this field as a target firm. From this, 102 potential technology opportunities are identified and validated from a dataset of 5034 collected green patents. This practical applicability demonstrates our method's efficacy in guiding future technological directions for firms.