Artificial intelligence has begun to solve more and more unsolved problems of human beings, and has achieved good results. However, in the past few years, the number of scientific research in the field of artificial intelligence has grown exponentially, making it difficult for scientists and practitioners to track these developments in a timely manner. Data shows that the number of research papers in the field of machine learning doubles every 23 months. One reason is that AI is being used in different disciplines such as mathematics, statistics, physics, medicine and biochemistry. By gaining insights from scientific literature, tools to propose new personalized research directions and ideas can significantly accelerate scientific progress. In the process of the intersection of artificial intelligence and other fields, how should people identify which directions are meaningful and worth doing? To this end, an international team led by Mario Krenn, an artificial intelligence scientist at the Max Planck Institute of Optical Sciences (MPL), released a study on "high-quality link prediction in knowledge networks with exponential growth". The relevant research papers were entitled "Predicting the Future of AI with AI: High Quality link prediction in an exponentially growing knowledge network" and published on the preprint website arXiv. (Source: arXiv) The purpose of this research work is to design a program that can "read, understand, and then act" on AI related literature, so as to open the door to predict and suggest interdisciplinary research ideas. The research team believes that in the long run, this will improve the productivity of AI researchers, open up new research approaches, and guide progress in this field. Past practice has proved that new research ideas are often generated by establishing new connections between seemingly unrelated topics/fields. This prompted the research team to formulate the evolution of AI literature as a time network modeling task, and create a semantic network that can describe the content and evolution of AI literature since 1994. At the same time, the research team also discussed a network containing 64000 concepts (also known as nodes) and 18 million connections between nodes, and used semantic networks as inputs for 10 different statistical and machine learning methods. One of the most basic tasks -- building a semantic network -- helps to extract knowledge from the network and then use computer algorithms to process it. Figure | In this work, the research team used 143000 papers on artificial intelligence and machine learning published on arXiv from 1992 to 2020, and built a concept list using RAKE and other NLP tools. These concepts form the nodes of the semantic network. When two concepts appear in the title or abstract of a paper at the same time, the edge will be drawn. In this way, they built