专利视角下人工智能与车联网技术融合演化研究

通过对人工智能与车联网技术融合特征进行研究，能够精准识别车联网产业技术机会，为人工智能与车联网产业创新融合发展提供方向指引。选取2000—2019年相关专利数据，对人工智能与车联网领域技术融合进行识别与测度，采用Word2vec文本挖掘算法和社会网络分析法，揭示融合领域技术发展态势、技术主题关联模式、技术共现网络整体结构和节点功能演变情况。研究发现，融合过程呈现明显的阶段性特征，图像识别等相关算法是车联网领域最基础的人工智能算法，卷积神经网络等相关算法是最具潜力和发展前景的人工智能算法；智能交通平台是融合领域的重点发展方向；数据识别与数据表示技术在融合演变过程中的控制能力较强，无线通信网络技术发展受到相关人工智能技术发展水平的制约。基于研究结论，从研发方向、学科建设、平台搭建等方面提出启示和建议。

Technology Convergence Evolution of Artificial Intelligence and Internet of Vehicles Based on Patentometrics

With the rapid development of emerging technologies such as artificial intelligence and big data, the industry is transforming towards digitization and intelligence, and technology integration has become an important trend in the development of global technology and industrial innovation. The Internet of Vehicles is a typical high-tech complex involving artificial intelligence, big data, cloud computing and other technologies. Through the evolution research on the convergence characteristics of artificial intelligence and Internet of Vehicles technology, it is likely to more accurately identify the technical opportunities in the field of Internet of Vehicles industry, and provide direction guidance for the innovative and integrated development of artificial intelligence and Internet of Vehicles industry. At present, the academic community has carried out extensive research on the application of artificial intelligence in the field of Internet of Vehicles, but it is focused on the derivation and verification of specific technical solutions, and less involved in the technology convergence characteristics, trends of artificial intelligence and the Internet of Vehicles. Besides, there is a lack of application of text mining methods such as machine learning in technology convergence research based on patent data, and there are still certain limitations in technology convergence analysis only from the aspect of citation network, patent index or using patent generic information. Therefore, this paper adopts Word2vec text mining algorithm and social network analysis method to reveal the technological development trend in the fusion field, the correlation pattern of technical topics, the overall structure of the technology co-occurrence network and the evolution of node functions.#br#The research framework is built with the help of data analysis methods and technology convergence theory. Firstly, relevant patent data of artificial intelligence and Internet of Vehicles field from 2000 to 2019 are collected as the research object and the data are preprocessed to obtain information feature extraction. Next, the technology life cycle of the fusion field is divided into budding period, slow development period and rapid development period. Finally,the technical indicators such as patent application volume, technology convergence degree and technology similarity are analyzed from the three dimensions of technology convergence identification and measurement, technology theme association and technology co-occurrence network. In order to display the results of data analysis more intuitively, the convergence characteristics and evolution process of artificial intelligence and Internet of Vehicles technology are visualized in stages.#br#The results show that the development of technology convergence presents significant phasic characteristics, and the degree of technology convergence has been markedly deepened. In the field of Internet of vehicles, there are various artificial intelligence algorithms emerging and actively applied in each period, and the application scope of artificial intelligence in the field of Internet of Vehicles continues to expand. In response to users' diversified comprehensive service needs such as vehicle safety management, intelligent reception and travel arrangements, artificial intelligence has gradually been applied to the technical research of the Internet of Vehicles intelligent platform, which is an important turning point in the development of the Internet of Vehicles industry. Furthermore, the relationship between technical fields is getting closer, and the technology co-occurrence network is still relatively sparse. Data recognition and data representation, computer system based on specific computing model and other technologies have strong control ability in the process of convergence evolution. Specifically, the rapid development of technologies such as wireless communication networks, digital information transmission and signaling devices requires breakthroughs in related artificial intelligence technologies.#br#This paper innovatively proposes an identification framework for the convergence mode of artificial intelligence across the field of Internet of Vehicle，and it can be generally applied to other industrial field for predicting the trend of technology convergence. Different from the previous literature, this paper comprehensively uses text mining and network analysis methods to empirically analyze the impact of artificial intelligence algorithms on the development of Internet of Vehicles technology from multiple dimensions, and explore the most potential and promising artificial intelligence algorithms and their application scope. In addition, this paper discovers new technology opportunities for the application of artificial intelligence in the field of Internet of Vehicles, and puts forward enlightenment and suggestions from the aspects of R&D direction, discipline construction and platform construction.#br#