Journal of IKEEE (전기전자학회논문지)

Institute of Korean Electrical and Electronics Engineers (한국전기전자학회)
권호 표지
DOI QR코드

DOI QR Code

A patent application filing forecasting method based on the bidirectional LSTM

양방향 LSTM기반 시계열 특허 동향 예측 연구

  • Received : 2022.10.19
  • Accepted : 2022.10.24
  • Published : 2022.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

The number of patent application filing for a specific technology has a good relation with the technology's life cycle and future industry development on that area. So industry and governments are highly interested in forecasting the number of patent application filing in order to take appropriate preparations in advance. In this paper, a new method based on the bidirectional long short-term memory(LSTM), a kind of recurrent neural network(RNN), is proposed to improve the forecasting accuracy compared to related methods. Compared with the Bass model which is one of conventional diffusion modeling methods, the proposed method shows the 16% higher performance with the Korean patent filing data on the five selected technology areas.

특정 분야의 특허출원수는 기술의 수명주기 및 산업의 활성화 정도와 밀접한 관계를 가지고 있다. 따라서 사전에 사업을 준비하는 기업들과 미래 유망 기술을 초기 단계에서 선발하여 투자하고자 하는 정부 기관들은 미래의 특허 출원수 예측에 대해 큰 관심을 가지고 있다. 본 논문에서는 시계열 데이터에 적합한 RNN의 기법 중 하나인 양방향 LSTM 기법을 이용하여 기존 예측 방법들보다 정확도를 높이는 방법을 제안한다. 5개 분야의 대한민국 특허 출원 데이터에 대해서 제안된 방법은 기존에 사용되던 확산 모델 중 하나인 Bass 모델과 비교하여 평균 절대 백분율 오차(MAPE)의 값이 약 16퍼센트 향상된 결과를 보여준다.

Keywords

Acknowledgement

This research was supported by the research fund of Hanbat National University in 2019.

References

  1. R. Haupt, M. Kloyer and M. Lange, "Patent indicators for the technology life cycle development," Research Policy, vol.36, no.3, pp.387-398, 2007. DOI: 10.1016/j.respol.2006.12.004.
  2. C.-Y. Liu and J.-C. Wang, "Forecasting the development of the biped robot walking technique in Japan through S-curve model analysis," Scientometrics, vol.82, no.1, pp.21-36, 2010. DOI: 10.1007/s11192-009-0055-5
  3. P. Hingley and M. Nicolas, "Methods for forecasting numbers of patent applications at the European Patent Office," World Patent Information, vol.26, no.3, pp.191-204, 2004. DOI: 10.1016/j.wpi.2003.12.006.
  4. DH. Kim, SS. Park, YG. Shin and DS. Jang, "Forecasting the Diffusion of Technology using Patent Information: Focused on Information Security Technology for Network-Centric Warfare," Journal of the Korea Contents Association, vol.9, no.2, pp.125-132, 2009. DOI: 10.5392/JKCA.2009.9.2.125
  5. J. Hong, T. Kim and H. Koo, "A Parameter Estimation of Bass Diffusion Model by the Hybrid of NLS and OLS," Journal of the Korean Institute of Industrial Engineers, vol.37, no.1, pp.74-82, 2011. https://doi.org/10.7232/JKIIE.2011.37.1.074
  6. GJ. Kim, DH. Yoon, JH. Hwang and DJ. Sun. "Discovering the emerging technologies through patent topic modeling and growth curve model," Journal of Korean Institute of Intelligent Systems, vol.27, no.4, pp.357-363, https://doi.org/10.5391/JKIIS.2017.27.4.357
  7. D. Nam and G. Choi, "Technology Trend Analysis in the Automotive Semiconductor Industry using Topic Model and Patent Analysis," Journal of Korea Technology Innovation Society, vol.21, no.3, pp.1155-1178, 2018.
  8. M. N. Kyebambe, G. Cheng, Y. Huang, C. He, and Z. Zhang. "Forecasting emerging technologies: A supervised learning approach through patent analysis," Technological Forecasting and Social Change, vol.125, pp.236-244, 2017. DOI: 10.1016/j.techfore.2017.08.002.
  9. R. Dutt, P. Rathi, and V. Krishna, "Novel mixed-encoding for forecasting patent grant duration," World Patent Information, vol.64, pp.102007, 2021. DOI: 10.1016/j.wpi.2020.102007.
  10. G. Kim and J. Bae. "A novel approach to forecast promising technology through patent analysis," Technological Forecasting and Social Change, vol.117, pp.228-237, 2017. DOI:10.1016/j.techfore.2016.11.023.
  11. J. H. Cho, J. Lee, and S. Y. Sohn, "Predicting future technological convergence patterns based on machine learning using link prediction," Scientometrics, vol.126, pp.5413-429, 2021. DOI: 10.1007/s11192-021-03999-8
  12. C. L. Giles, G. M. Kuhn and R. J. Williams, "Dynamic recurrent neural networks: Theory and applications," IEEE Transactions on Neural Networks, vol.5, no.2, pp.153-156, DOI: 10.1109/TNN.1994.8753425.
  13. S. Hochreiter and J. Schmidhuber "Long Short-Term Memory," Neural Computation, vol.9, no.8, pp.1735-1780, 1997. DOI: 10.1162/neco.1997.9.8.1735
  14. Y. Bengio, P. Simard and P. Frasconi, "Learning long-term dependencies with gradient descent is difficult," IEEE Transactions on Neural Networks, vol.5, no.2, pp.157-166, 1994. DOI: 10.1109/72.279181.
  15. A. Sherstinsky, "Fundamentals of Recurrent Neural Network(RNN) and Long Short-Term Memory(LSTM) network," Physica D: Nonlinear Phenomena, vol.404, pp.132306, 2020. DOI: 10.1016/j.physd.2019.132306.
  16. M. Schuster and K. K. Paliwal, "Bidirectional recurrent neural networks," IEEE Transactions on Signal Processing, vo.45, no.11, pp.2673-2681. 1997. DOI: 10.1109/78.650093.
  17. H. Sak, A. Senior and F. Beaufays, "Long short-term memory recurrent neural network architectures for large scale acoustic modeling," International Speech Communication Association (INTERSPEECH). pp.338-342,
  18. C. Olah, "Understanding LSTM Networks," http://colah.github.io/posts/2015-08-Understanding-LSTMs/
  19. D. Lee, "Exploratory research on the analysis of national R&D programs using growth model," Korea Institute of Science and Technology Evaluation and Planning, vol.27, 2014. DOI: 10.23000/TRKO201400012780