Iranian Research Organisation for Science and Technology Journal of Technology Development Management 2008-5060 6 3 2018 11 22 Application of Artificial Neural Networks in Predicting Macro Indicators of Science and Technology Application of Artificial Neural Networks in Predicting Macro Indicators of Science and Technology 129 155 762 10.22104/jtdm.2019.2841.1960 FA Iman Raeesi Vanani Department of Information Technology Management, Allameh Tabataba'i University Naeima Mirzamomen M.Sc. Student of information technology management, Allameh Tabataba'i University, Tehran,Iran. Journal Article 2018 04 04 The R&D evaluation and the relation between the production of science and technology at the macro level of the countries have been limited due to the high volume of information and rapid rate of changes in this area. This research is aimed at understanding the relationship and performance of technology development in relation to science production activities across countries, which is a descriptive-applied research type. The goal of this study is to build a model using advanced algorithms that can predict the scientometric indicators based on the production of science in countries. Also, the effect of each of the scientometric indicators on the technology index was determined using the sensitivity analysis method of the neural network. The research method is CRISP-DM and the data were extracted from the SCImago Journal & Country Rank SJCR data base and the World Intellectual Property Organization (WIPO) during the time period from 2001 to 2015. According to the results, neural network has more accuracy and ability to model than the regression. The results of the sensitivity analysis showed that the most important parameter for scientometric indicators is H-index and the process of referring to the international papers. Policy makers can use the research results to identify the influential variables of science production that lead to the creation of technology The R&D evaluation and the relation between the production of science and technology at the macro level of the countries have been limited due to the high volume of information and rapid rate of changes in this area. This research is aimed at understanding the relationship and performance of technology development in relation to science production activities across countries, which is a descriptive-applied research type. The goal of this study is to build a model using advanced algorithms that can predict the scientometric indicators based on the production of science in countries. Also, the effect of each of the scientometric indicators on the technology index was determined using the sensitivity analysis method of the neural network. The research method is CRISP-DM and the data were extracted from the SCImago Journal & Country Rank SJCR data base and the World Intellectual Property Organization (WIPO) during the time period from 2001 to 2015. According to the results, neural network has more accuracy and ability to model than the regression. The results of the sensitivity analysis showed that the most important parameter for scientometric indicators is H-index and the process of referring to the international papers. Policy makers can use the research results to identify the influential variables of science production that lead to the creation of technology https://jtdm.irost.ir/article_762_2c4439fe691abe85750c77b3dadbfe9d.pdf