دانلود رایگان مقاله تنوع کنسانتره مبتنی بر توانایی تکنولوژیکی

abstract

This research responds to the needs for concentric diversification by focusing on how firms can find new business opportunities based on their technological capabilities. We propose a systematic approach to identifying potential areas for concentric diversification at a product level via link analysis of products and technologies. For this, first, text mining is utilised to construct an integrated patent-product database from the US patent and trademark database. Second, association rule mining is employed to construct a product ecology network using directed technological relationships between products. Third, a link prediction analysis is conducted to identify potential areas for concentric diversification at a product level. Finally, three quantitative indicators are developed to assess the characteristics of the areas identified. Our case study employs a total of 850,676 patents and 328,288 products in the integrated patent-product database from 2010 to 2014 and shows that the proposed approach enables a wide-ranging search for potential areas for concentric diversification and the quick assessment of their characteristics, with statistically significant results. We believe that the proposed approach will be useful as a complementary tool for decision making for small and medium-sized high-tech companies that are considering entering new business areas, but which have little domain knowledge.

5. Conclusions

This study has proposed a systematic approach to identifying potential areas for concentric diversification. The proposed approach is based on the premise that significant technological relationships between products extracted from large-scale quantitative databases can provide valuable information on the feasibility of concentric diversification from one area to another. Our case study shows that the proposed approach enables a wide ranging search for potential areas for concentric diversi- fication and the quick assessment of their characteristics, with analysis results that are statistically significant.