Prediksi Indeks Inovasi Global (GII) dengan Pendekatan Hybrid Machine Learning: Analisis Komparatif Model Random Forest, XGBoost, dan LSTM Berbasis Data Time Series WIPO-GII Indonesia 2013-2022
Keywords:
informatikaAbstract
Penelitian ini mengembangkan kerangka kerja hybrid machine learning untuk memprediksi Indeks Inovasi Global (GII) Indonesia berbasis data time series WIPO-GII periode 2013–2022. Dengan membandingkan kinerja tiga model Random Forest (RF), XGBoost, dan LSTM serta arsitektur hybrid (XGBoost-LSTM), penelitian mengidentifikasi XGBoost sebagai model terbaik (RMSE=1.951; MAPE=3.7%) yang mampu menangkap hubungan non-linear antara kinerja tahun sebelumnya (lag_1) dan GII tahun berjalan. Analisis feature importance dan SHAP mengungkap dominasi mutlak lag_1 (87% varians), menunjukkan bahwa akumulasi kapasitas inovasi, seperti infrastruktur digital dan SDM, lebih menentukan keberhasilan daripada intervensi jangka pendek. Sementara itu, LSTM gagal total (RMSE=51.419) akibat keterbatasan data dan kompleksitas arsitektur, mengonfirmasi tantangan yang umum dijumpai dalam penerapan deep learning pada dataset kecil.
Temuan residual mengungkap kerentanan sistem inovasi terhadap guncangan eksternal, seperti pandemi COVID-19 tahun 2020 dan gejolak geopolitik tahun 2022. Hal ini mendorong rekomendasi kebijakan: (1) perencanaan multianual berbasis lag_1 (contoh: alokasi anggaran 5-tahun), (2) percepatan infrastruktur digital (jaringan 5G, pusat data), dan (3) integrasi variabel makroekonomi (stabilitas politik, harga energi) ke dalam model prediksi. Studi ini menekankan pentingnya pendekatan berbasis akumulasi kapasitas dan kebijakan adaptif, sekaligus menawarkan panduan teknis untuk meningkatkan ketahanan sistem melalui analisis explainable AI. Implikasi praktis mencakup pengembangan dashboard prediksi real-time berbasis SHAP, yang dapat menjadi alat strategis bagi pemangku kepentingan dalam merancang kebijakan berbasis bukti.
References
[1] K. Alqararah, “Assessing the robustness of composite indicators: the case of the Global Innovation Index,” J Innov Entrep, vol. 12, no. 1, p. 61, Sep. 2023, doi: 10.1186/s13731-023-00332-w.
[2] S. Djaballah, L. Saidi, K. Meftah, A. Hechifa, M. Bajaj, and I. Zaitsev, “A hybrid LSTM random forest model with grey wolf optimization for enhanced detection of multiple bearing faults,” Sci Rep, vol. 14, no. 1, p. 23997, Oct. 2024, doi: 10.1038/s41598-024-75174-x.
[3] Y. E. Gur, “Development and application of machine learning models in US consumer price index forecasting: Analysis of a hybrid approach,” DSFE, vol. 4, no. 4, pp. 469–513, 2024, doi: 10.3934/DSFE.2024020.
[4] M. Abumohsen, A. Y. Owda, M. Owda, and A. Abumihsan, “Hybrid machine learning model combining of CNN-LSTM-RF for time series forecasting of Solar Power Generation,” e-Prime - Advances in Electrical Engineering, Electronics and Energy, vol. 9, p. 100636, Sep. 2024, doi: 10.1016/j.prime.2024.100636.
[5] B. Kumar, Sunil, and N. Yadav, “A novel hybrid model combining ? S A R M A and LSTM for time series forecasting,” Applied Soft Computing, vol. 134, p. 110019, Feb. 2023, doi: 10.1016/j.asoc.2023.110019.
[6] W. Zhang, S. Li, Z. Guo, and Y. Yang, “A hybrid forecasting model based on deep learning feature extraction and statistical arbitrage methods for stock trading strategies,” Journal of Forecasting, vol. 42, no. 7, pp. 1729–1749, Nov. 2023, doi: 10.1002/for.2978.
[7] Y.-B. Hong and J.-D. Choi, “Prediction of KOSPI Index by Time Series based on Convergence Model using Cross-Validation of Time Series Data,” JKORMS, vol. 48, no. 4, pp. 1–21, Nov. 2023, doi: 10.7737/JKORMS.2023.48.4.001.
[8] T. Yang, S. He, X. Chen, P. Fu, L. Huang, and X. Zhang, “Combined Multi-Component Composite Time Series Power Prediction Model for Distributed Energy Systems Based on Stl Data Decomposition,” 2024. doi: 10.2139/ssrn.4823042.
[9] L. Pahuja and A. Kamal, “ENLEFD?DM?: Ensemble Learning based Ethereum Fraud Detection using CRISP?DM framework,” Expert Systems, vol. 40, no. 9, p. e13379, Nov. 2023, doi: 10.1111/exsy.13379.
[10] WIPO, “Global Innovation Index.” https://prosperitydata360.worldbank.org/en/dataset/WIPO+GII, 2024. [Online]. Available: https://www.wipo.int/en/web/global-innovation-index
[11] A. Zainuddin, M. A. Hairuddin, A. I. M. Yassin, Z. I. A. Latiff, and A. Azhar, “Time Series Data and Recent Imputation Techniques for Missing Data: A Review,” in 2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST), Miri Sarawak, Malaysia: IEEE, Oct. 2022, pp. 346–350. doi: 10.1109/GECOST55694.2022.10010499.
[12] B. Bala and S. Behal, “A Brief Survey of Data Preprocessing in Machine Learning and Deep Learning Techniques,” in 2024 8th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC), Kirtipur, Nepal: IEEE, Oct. 2024, pp. 1755–1762. doi: 10.1109/I-SMAC61858.2024.10714767.
[13] I. sibel Kervanci and F. Akay, “LSTM Hyperparameters optimization with Hparam parameters for Bitcoin Price Prediction,” Sakarya University Journal of Computer and Information Sciences, vol. 6, no. 1, pp. 1–9, Apr. 2023, doi: 10.35377/saucis...1172027.
[14] V. Cerqueira, L. Torgo, and I. Mozeti?, “Evaluating time series forecasting models: an empirical study on performance estimation methods,” Mach Learn, vol. 109, no. 11, pp. 1997–2028, Nov. 2020, doi: 10.1007/s10994-020-05910-7.
[15] A. Davydenko and R. Fildes, “Measuring Forecasting Accuracy: Problems and Recommendations (by the Example of SKU-Level Judgmental Adjustments),” in Intelligent Fashion Forecasting Systems: Models and Applications, T.-M. Choi, C.-L. Hui, and Y. Yu, Eds., Berlin, Heidelberg: Springer Berlin Heidelberg, 2014, pp. 43–70. doi: 10.1007/978-3-642-39869-8_4.
[16] S. Geng, “Analysis of the Different Statistical Metrics in Machine Learning,” HSET, vol. 88, pp. 350–356, Mar. 2024, doi: 10.54097/jhq3tv19.
[17] S. M. Lundberg et al., “From local explanations to global understanding with explainable AI for trees,” Nat Mach Intell, vol. 2, no. 1, pp. 56–67, Jan. 2020, doi: 10.1038/s42256-019-0138-9.
[18] P. Biecek and T. Burzykowski, Explanatory Model Analysis: Explore, Explain, and Examine Predictive Models. New York: Chapman and Hall/CRC, 2021. doi: 10.1201/9780429027192.
[19] Rob J Hyndman and George Athanasopoulos, Forecasting: Principles and Practice (3rd ed), 3rd ed. 2025. Accessed: Apr. 08, 2025. [Online]. Available: https://otexts.com/fpp3/
[20] M. S. D. Carvalho and G. L. D. Silva, “Inside the black box: using Explainable AI to improve Evidence-Based Policies,” in 2021 IEEE 23rd Conference on Business Informatics (CBI), Bolzano, Italy: IEEE, Sep. 2021, pp. 57–64. doi: 10.1109/CBI52690.2021.10055.
[21] Christoph Molnar, Interpretable Machine Learning: A Guide for Making Black Box Models Explainable (3rd ed.), 3rd ed. https://christophm.github.io/interpretable-ml-book/, 2025. Accessed: Apr. 08, 2025. [Online]. Available: https://christophm.github.io/interpretable-ml-book/
[22] Gopal Krushna Panda, “Sustainable Finance: Driving a Greener Future,” IJFMR, vol. 6, no. 3, p. 21066, May 2024, doi: 10.36948/ijfmr.2024.v06i03.21066.
[23] Monte dei Paschi and C. Giliberto, “Sustainable Investments and ESG factors,” RMM, vol. 19, no. 3, pp. 44–52, Dec. 2024, doi: 10.47473/2020rmm0147.
[24] O. Al-Jayyousi, H. Amin, H. A. Al-Saudi, A. Aljassas, and E. Tok, “Mission-Oriented Innovation Policy for Sustainable Development: A Systematic Literature Review,” Sustainability, vol. 15, no. 17, p. 13101, Aug. 2023, doi: 10.3390/su151713101.
[25] M. P. Hekkert, M. J. Janssen, J. H. Wesseling, and S. O. Negro, “Mission-oriented innovation systems,” Environmental Innovation and Societal Transitions, vol. 34, pp. 76–79, Mar. 2020, doi: 10.1016/j.eist.2019.11.011.
[26] OECD, “Designing Effective Governance to Enable Mission Success,” OECD Science, Technology and Industry Policy Papers, Dec. 2024. doi: 10.1787/898bca89-en.
[27] A. Sergeev, E. Baglaeva, and I. Subbotina, “Hybrid model combining LSTM with discrete wavelet transformation to predict surface methane concentration in the Arctic Island Belyy,” Atmospheric Environment, vol. 317, p. 120210, Jan. 2024, doi: 10.1016/j.atmosenv.2023.120210.
[28] B. Barz and J. Denzler, “Deep Learning on Small Datasets without Pre-Training using Cosine Loss,” in 2020 IEEE Winter Conference on Applications of Computer Vision (WACV), Snowmass Village, CO, USA: IEEE, Mar. 2020, pp. 1360–1369. doi: 10.1109/WACV45572.2020.9093286.
[29] R. Hu and Q. Liu, “Active noise control system based on the combined CNN-LSTM network,” in Workshop on Electronics Communication Engineering (WECE 2023), W. Xu, Ed., Guilin, China: SPIE, Jan. 2024, p. 15. doi: 10.1117/12.3015656.
[30] W. M. Cohen and D. A. Levinthal, “Absorptive Capacity: A New Perspective on Learning and Innovation,” Administrative Science Quarterly, vol. 35, no. 1, p. 128, Mar. 1990, doi: 10.2307/2393553.
[31] M. A. Akaka, S. L. Vargo, and H. Wieland, “Extending the Context of Innovation: The Co-creation and Institutionalization of Technology and Markets,” in Innovating in Practice, T. Russo-Spena, C. Mele, and M. Nuutinen, Eds., Cham: Springer International Publishing, 2017, pp. 43–57. doi: 10.1007/978-3-319-43380-6_3.
[32] A. Konstantinov, L. Utkin, and S. Kirpichenko, “AGBoost: Attention-based Modification of Gradient Boosting Machine,” in 2022 31st Conference of Open Innovations Association (FRUCT), Helsinki, Finland: IEEE, Apr. 2022, pp. 96–101. doi: 10.23919/FRUCT54823.2022.9770928.
[33] Y. Katz, “Government’s Role in Advancing Innovation,” Rand. Inter. Social Sci. J., vol. 2, no. 2, pp. 161–175, Apr. 2021, doi: 10.47175/rissj.v2i2.236.
[2] S. Djaballah, L. Saidi, K. Meftah, A. Hechifa, M. Bajaj, and I. Zaitsev, “A hybrid LSTM random forest model with grey wolf optimization for enhanced detection of multiple bearing faults,” Sci Rep, vol. 14, no. 1, p. 23997, Oct. 2024, doi: 10.1038/s41598-024-75174-x.
[3] Y. E. Gur, “Development and application of machine learning models in US consumer price index forecasting: Analysis of a hybrid approach,” DSFE, vol. 4, no. 4, pp. 469–513, 2024, doi: 10.3934/DSFE.2024020.
[4] M. Abumohsen, A. Y. Owda, M. Owda, and A. Abumihsan, “Hybrid machine learning model combining of CNN-LSTM-RF for time series forecasting of Solar Power Generation,” e-Prime - Advances in Electrical Engineering, Electronics and Energy, vol. 9, p. 100636, Sep. 2024, doi: 10.1016/j.prime.2024.100636.
[5] B. Kumar, Sunil, and N. Yadav, “A novel hybrid model combining ? S A R M A and LSTM for time series forecasting,” Applied Soft Computing, vol. 134, p. 110019, Feb. 2023, doi: 10.1016/j.asoc.2023.110019.
[6] W. Zhang, S. Li, Z. Guo, and Y. Yang, “A hybrid forecasting model based on deep learning feature extraction and statistical arbitrage methods for stock trading strategies,” Journal of Forecasting, vol. 42, no. 7, pp. 1729–1749, Nov. 2023, doi: 10.1002/for.2978.
[7] Y.-B. Hong and J.-D. Choi, “Prediction of KOSPI Index by Time Series based on Convergence Model using Cross-Validation of Time Series Data,” JKORMS, vol. 48, no. 4, pp. 1–21, Nov. 2023, doi: 10.7737/JKORMS.2023.48.4.001.
[8] T. Yang, S. He, X. Chen, P. Fu, L. Huang, and X. Zhang, “Combined Multi-Component Composite Time Series Power Prediction Model for Distributed Energy Systems Based on Stl Data Decomposition,” 2024. doi: 10.2139/ssrn.4823042.
[9] L. Pahuja and A. Kamal, “ENLEFD?DM?: Ensemble Learning based Ethereum Fraud Detection using CRISP?DM framework,” Expert Systems, vol. 40, no. 9, p. e13379, Nov. 2023, doi: 10.1111/exsy.13379.
[10] WIPO, “Global Innovation Index.” https://prosperitydata360.worldbank.org/en/dataset/WIPO+GII, 2024. [Online]. Available: https://www.wipo.int/en/web/global-innovation-index
[11] A. Zainuddin, M. A. Hairuddin, A. I. M. Yassin, Z. I. A. Latiff, and A. Azhar, “Time Series Data and Recent Imputation Techniques for Missing Data: A Review,” in 2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST), Miri Sarawak, Malaysia: IEEE, Oct. 2022, pp. 346–350. doi: 10.1109/GECOST55694.2022.10010499.
[12] B. Bala and S. Behal, “A Brief Survey of Data Preprocessing in Machine Learning and Deep Learning Techniques,” in 2024 8th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC), Kirtipur, Nepal: IEEE, Oct. 2024, pp. 1755–1762. doi: 10.1109/I-SMAC61858.2024.10714767.
[13] I. sibel Kervanci and F. Akay, “LSTM Hyperparameters optimization with Hparam parameters for Bitcoin Price Prediction,” Sakarya University Journal of Computer and Information Sciences, vol. 6, no. 1, pp. 1–9, Apr. 2023, doi: 10.35377/saucis...1172027.
[14] V. Cerqueira, L. Torgo, and I. Mozeti?, “Evaluating time series forecasting models: an empirical study on performance estimation methods,” Mach Learn, vol. 109, no. 11, pp. 1997–2028, Nov. 2020, doi: 10.1007/s10994-020-05910-7.
[15] A. Davydenko and R. Fildes, “Measuring Forecasting Accuracy: Problems and Recommendations (by the Example of SKU-Level Judgmental Adjustments),” in Intelligent Fashion Forecasting Systems: Models and Applications, T.-M. Choi, C.-L. Hui, and Y. Yu, Eds., Berlin, Heidelberg: Springer Berlin Heidelberg, 2014, pp. 43–70. doi: 10.1007/978-3-642-39869-8_4.
[16] S. Geng, “Analysis of the Different Statistical Metrics in Machine Learning,” HSET, vol. 88, pp. 350–356, Mar. 2024, doi: 10.54097/jhq3tv19.
[17] S. M. Lundberg et al., “From local explanations to global understanding with explainable AI for trees,” Nat Mach Intell, vol. 2, no. 1, pp. 56–67, Jan. 2020, doi: 10.1038/s42256-019-0138-9.
[18] P. Biecek and T. Burzykowski, Explanatory Model Analysis: Explore, Explain, and Examine Predictive Models. New York: Chapman and Hall/CRC, 2021. doi: 10.1201/9780429027192.
[19] Rob J Hyndman and George Athanasopoulos, Forecasting: Principles and Practice (3rd ed), 3rd ed. 2025. Accessed: Apr. 08, 2025. [Online]. Available: https://otexts.com/fpp3/
[20] M. S. D. Carvalho and G. L. D. Silva, “Inside the black box: using Explainable AI to improve Evidence-Based Policies,” in 2021 IEEE 23rd Conference on Business Informatics (CBI), Bolzano, Italy: IEEE, Sep. 2021, pp. 57–64. doi: 10.1109/CBI52690.2021.10055.
[21] Christoph Molnar, Interpretable Machine Learning: A Guide for Making Black Box Models Explainable (3rd ed.), 3rd ed. https://christophm.github.io/interpretable-ml-book/, 2025. Accessed: Apr. 08, 2025. [Online]. Available: https://christophm.github.io/interpretable-ml-book/
[22] Gopal Krushna Panda, “Sustainable Finance: Driving a Greener Future,” IJFMR, vol. 6, no. 3, p. 21066, May 2024, doi: 10.36948/ijfmr.2024.v06i03.21066.
[23] Monte dei Paschi and C. Giliberto, “Sustainable Investments and ESG factors,” RMM, vol. 19, no. 3, pp. 44–52, Dec. 2024, doi: 10.47473/2020rmm0147.
[24] O. Al-Jayyousi, H. Amin, H. A. Al-Saudi, A. Aljassas, and E. Tok, “Mission-Oriented Innovation Policy for Sustainable Development: A Systematic Literature Review,” Sustainability, vol. 15, no. 17, p. 13101, Aug. 2023, doi: 10.3390/su151713101.
[25] M. P. Hekkert, M. J. Janssen, J. H. Wesseling, and S. O. Negro, “Mission-oriented innovation systems,” Environmental Innovation and Societal Transitions, vol. 34, pp. 76–79, Mar. 2020, doi: 10.1016/j.eist.2019.11.011.
[26] OECD, “Designing Effective Governance to Enable Mission Success,” OECD Science, Technology and Industry Policy Papers, Dec. 2024. doi: 10.1787/898bca89-en.
[27] A. Sergeev, E. Baglaeva, and I. Subbotina, “Hybrid model combining LSTM with discrete wavelet transformation to predict surface methane concentration in the Arctic Island Belyy,” Atmospheric Environment, vol. 317, p. 120210, Jan. 2024, doi: 10.1016/j.atmosenv.2023.120210.
[28] B. Barz and J. Denzler, “Deep Learning on Small Datasets without Pre-Training using Cosine Loss,” in 2020 IEEE Winter Conference on Applications of Computer Vision (WACV), Snowmass Village, CO, USA: IEEE, Mar. 2020, pp. 1360–1369. doi: 10.1109/WACV45572.2020.9093286.
[29] R. Hu and Q. Liu, “Active noise control system based on the combined CNN-LSTM network,” in Workshop on Electronics Communication Engineering (WECE 2023), W. Xu, Ed., Guilin, China: SPIE, Jan. 2024, p. 15. doi: 10.1117/12.3015656.
[30] W. M. Cohen and D. A. Levinthal, “Absorptive Capacity: A New Perspective on Learning and Innovation,” Administrative Science Quarterly, vol. 35, no. 1, p. 128, Mar. 1990, doi: 10.2307/2393553.
[31] M. A. Akaka, S. L. Vargo, and H. Wieland, “Extending the Context of Innovation: The Co-creation and Institutionalization of Technology and Markets,” in Innovating in Practice, T. Russo-Spena, C. Mele, and M. Nuutinen, Eds., Cham: Springer International Publishing, 2017, pp. 43–57. doi: 10.1007/978-3-319-43380-6_3.
[32] A. Konstantinov, L. Utkin, and S. Kirpichenko, “AGBoost: Attention-based Modification of Gradient Boosting Machine,” in 2022 31st Conference of Open Innovations Association (FRUCT), Helsinki, Finland: IEEE, Apr. 2022, pp. 96–101. doi: 10.23919/FRUCT54823.2022.9770928.
[33] Y. Katz, “Government’s Role in Advancing Innovation,” Rand. Inter. Social Sci. J., vol. 2, no. 2, pp. 161–175, Apr. 2021, doi: 10.47175/rissj.v2i2.236.
Additional Files
Published
2025-07-31
Issue
Section
Articles
