Research on Enterprise Digital Agility Based on Machine Learning: An Evaluation of Green Financial Technology

Research on Enterprise Digital Agility Based on Machine Learning: An Evaluation of Green Financial Technology

Ying Zhang, Hong Chen, Keyi Ju
Copyright: © 2023 |Pages: 13
DOI: 10.4018/JGIM.327006
Article PDF Download
Open access articles are freely available for download

Abstract

To help enterprises quickly adapt to the environment of green finance, a technology innovation performance prediction method based on machine learning is proposed to improve digital convenience. Firstly, by analyzing scientific and technological innovation, the authors design four characteristics: the number of theses, the quantity and quality of projects, the level of technology transformation, and the value of commercialization. Then, according to the above features, a feature processing method based on improved attention mechanism is proposed to deeply explore the internal relationship between the four features. Finally, a performance evaluation method is used based on the temporal convolution network (TCN) that can predict the performance of scientific and technological innovation by inputting enhanced features. The experiment demonstrates that the proposed method can reach 0.846, 0.869, and 0.851 in terms of the precision, recall, and H value, respectively, which can help enterprises predict the performance and improve the electronic convenience of enterprises.
Article Preview
Top

Introduction

Research on Enterprise Digital Agility Based on Machine Learning: An Evaluation of Green Financial Technology

Amidst the advancements in green finance, the competition in the realm of science and technology is escalating with intensity. How to quickly gain advantages in the competitive market has become an issue in the management of the Science and Technology Innovation (STI) industry (Li et al., 2020; Umar & Safi, 2023). As the most intuitive method to evaluate the degree of development, performance appraisal is the key to assessing the scientific and technological innovation in green finance, and this has the power to promote its development.

Scientific and technological innovation stands as the sole pathway towards the development of green finance. Regular and quantitative evaluation of the performance of scientific and technological innovation can enhance both efficiency and compliance (Chen et al., 2023; Wang et al., 2021). Through the novel technologies, regulators can more quickly and accurately obtain and process a large amount of STI data and improve the efficiency and precision of regulatory work. At the same time, assessing sci-tech innovation can also support regulators in implementing more effective green finance policies and regulations for the sustainable development of the financial market. In addition, through the establishment of environmental data platforms, risk assessment models, and early warning systems, potential environmental and climate risks can be detected and responded to early, reducing the vulnerability and protecting the interests of financial institutions and investors (Ge et al., 2022). It is a complex task to assess the scientific and technological innovation capability of green finance, so there are many difficulties. First of all, scientific and technological innovation involves a wide range of fields and different types of technologies, each of which has its own unique characteristics and evaluation criteria. Therefore, the assessment of S&T innovation capability is often subjective and diverse (Wang et al., 2022). Secondly, scientific and technological innovation is a long-term process, and its results and impacts often need to be revealed over a relatively long time scale. Due to the uncertainty and evolution of technology, it is necessary to consider the future potential and development direction to assess scientific and technological innovation capability, which increases the uncertainty and difficulty of assessment. However, innovation activities often contain many unstructured and implicit factors, which are difficult to be quantified and measured (Ibrahim et al., 2022; Liu & Wang, 2023). Therefore, data acquisition and quantification are important challenges when assessing the performance of STI. Finally, due to the rapid evolution and reform of green finance, the methods for assessing the performance of STI need to suit the rapidly updated technology, and it is difficult to use a single method to evaluate the performance.

The development of machine learning provides a novel research approach for evaluating the performance of scientific and technological innovation in green finance. Based on the above difficulties (Aggarwal & Thakur, 2013), many scholars have studied it. According to different needs, different indicators are set to assess scientific and technological innovation (Curzi et al., 2019). First, key performance indicators (KPIs) are set to evaluate the performance of individuals or organizations in the field of STI. KPIs can be quantitative and measurable indicators that are used to assess the degree to which performance targets are being achieved. Second, by setting clear goals and standards, KPIs facilitate the evaluation of employees' performance in effectively carrying out their work tasks and accomplishing their goals. Employees set the goals together with management and regularly track and evaluate the achievement of the goals. Thirdly, the assessment of individuals or organizations' contributions and achievements at work centers on the results and outcomes they accomplish (Ali et al., 2019). The evaluation process places emphasis on appraising the tangible impact of employees on organizational goals and value creation. Fourthly, the evaluation of an individual or organization's performance entails observing their behavior, working methods, and exhibited skills and attitudes within their job role. This approach concentrates on assessing behaviors and working styles to discern their influence on technological innovation (Memon et al., 2019). By employing these four performance measures, the evaluation of STI performance in the context of green finance can be approached quantitatively (Al-Jedaia & Mehrez, 2020). In the realm of machine learning, researchers can utilize regression models, classification models, time series models, and association analysis mining models to address this task. For example, when the regression model is used to continuously predict scientific and technological innovation in green finance through historical data, the innovation capability can be directly quantified as the research output, the patent applications, the innovation partnerships, and other entities. When the classification model is used, the STI of individuals or organizations can be directly classified and finally quantified to different levels of innovation capability. These methods and indicators can be selected and combined according to the specific context and objectives of STI to evaluate all aspects of STI. However, there are many problems with these methods. Existing scientific and technological innovation performance assessment methods rely only on a single model for simulation, which cannot fully reflect the achievements of scientific and technological innovation (Liu et al., 2021).

To address this issue, the authors study the performance evaluation method of green fintech innovation based on machine learning. The main research contributions are as follows:

  • 1.

    According to the characteristics of green finance, the authors quantify and extract different features, which can represent the performance of STI, and propose a feature processing method based on an improved attention mechanism.

  • 2.

    According to the above features, the authors propose a performance evaluation method based on the temporal convolution network (TCN) to fully highlight the key points in the features and conduct performance evaluation.

  • 3.

    In the experiment, the authors’ method outperforms the performance of many excellent models and can accurately evaluate the performance of STI in green finance.

Complete Article List

Search this Journal:
Reset
Volume 31: 9 Issues (2023)
Volume 30: 12 Issues (2022)
Volume 29: 6 Issues (2021)
Volume 28: 4 Issues (2020)
Volume 27: 4 Issues (2019)
Volume 26: 4 Issues (2018)
Volume 25: 4 Issues (2017)
Volume 24: 4 Issues (2016)
Volume 23: 4 Issues (2015)
Volume 22: 4 Issues (2014)
Volume 21: 4 Issues (2013)
Volume 20: 4 Issues (2012)
Volume 19: 4 Issues (2011)
Volume 18: 4 Issues (2010)
Volume 17: 4 Issues (2009)
Volume 16: 4 Issues (2008)
Volume 15: 4 Issues (2007)
Volume 14: 4 Issues (2006)
Volume 13: 4 Issues (2005)
Volume 12: 4 Issues (2004)
Volume 11: 4 Issues (2003)
Volume 10: 4 Issues (2002)
Volume 9: 4 Issues (2001)
Volume 8: 4 Issues (2000)
Volume 7: 4 Issues (1999)
Volume 6: 4 Issues (1998)
Volume 5: 4 Issues (1997)
Volume 4: 4 Issues (1996)
Volume 3: 4 Issues (1995)
Volume 2: 4 Issues (1994)
Volume 1: 4 Issues (1993)
View Complete Journal Contents Listing