The pursuit of a sustainable and clean energy future has emerged as a paramount global imperative of the 21st century. Achieving this transition is a multifaceted and complex endeavour that requires a harmonious interplay of factors: effective policy frameworks, cleantech firms, and the transformative power of data science. By focusing on the European context, this paper advances the field in several directions. First, it explores the use of machine learning (ML) techniques to identify cleantech firms by analysing their mission statements and addressing the weaknesses of the existing methods. Second, it collects a unique and comprehensive dataset of national-level policies addressing the different topics covered by the European Green Deal. Third, in a regression analysis at country level, it examines the interplay between the national regulatory framework and the birth and growth of the cleantech landscape, by distinguishing between innovators (firms which develop the cleantech) and ecosystem firms (which adopt the cleantech). Our results indicate that the introduction of policies favours by itself the birth of cleantech innovator firms and their growth in the country. An increasing number of policies has a regulatory effect in the cleantech ecosystem limiting the number of newborn firms while favouring the growth of existing ones.

中文翻译：

---

欧洲的清洁技术和政策框架：机器学习方法

追求可持续和清洁能源的未来已成为 21 世纪全球的首要任务。实现这一转变是一项多方面、复杂的努力，需要多种因素的和谐相互作用：有效的政策框架、清洁技术公司和数据科学的变革力量。通过关注欧洲背景，本文在几个方向上推进了该领域的发展。首先，它探讨了如何使用机器学习（ML）技术通过分析清洁技术公司的使命宣言并解决现有方法的弱点来识别清洁技术公司。其次，它收集了一个独特且全面的国家级政策数据集，涉及欧洲绿色协议涵盖的不同主题。第三，在国家层面的回归分析中，通过区分创新者（开发清洁技术的公司）和生态系统公司（采用清洁技术），研究了国家监管框架与清洁技术领域的诞生和发展之间的相互作用。我们的结果表明，政策的出台本身就有利于清洁技术创新公司的诞生及其在该国的发展。越来越多的政策对清洁技术生态系统产生监管作用，限制新生企业的数量，同时有利于现有企业的增长。