The knowledge map has created a new paradigm of artificial intelligence. With the combination of data-driven and knowledge-driven, it has opened up the next generation of artificial intelligence and realized the collaboration between human and human, human and machine, and machine and machine. In addition, the knowledge graph has broken through the traditional artificial intelligence research field. Extracting knowledge from a wide range of multi-model data such as text, structure, vision and time series has become one of the main directions for the development of knowledge graphs. The construction of multi-modal knowledge graphs It can deeply integrate and flexibly use explicit symbolic knowledge and implicit data knowledge. Deep integration of deep learning, graph deep learning, transfer learning and meta-learning is the development trend of knowledge graphs, which can be used to construct large-scale knowledge graphs of all types and high coverage, and realize more profound knowledge reasoning. The road to artificial intelligence explained.
Dieter Fensel is one of the pioneers of semantic network research, and this book is one of the main achievements of his team in the field of sequence knowledge maps of translators. This book consists of 5 chapters, mainly discussing the entire life cycle of knowledge graphs, the concept, construction, implementation, maintenance and deployment, technical architecture and future work direction of knowledge graphs, which can be used as knowledge graphs, pattern recognition and artificial intelligence and computer vision. A reference book for scientists and engineers.
The knowledge map has created a new paradigm of artificial intelligence. With the combination of data-driven and knowledge-driven, it has opened up the next generation of artificial intelligence and realized the collaboration between human and human, human and machine, and machine and machine. In addition, the knowledge map has broken through the traditional artificial intelligence research field. Extracting knowledge from a wide range of multi-model data such as text, structure, vision, and time series has become one of the main directions for the development of knowledge maps. The construction of multi-modal knowledge maps can be Deep integration and flexible use of explicit symbolic knowledge and implicit data knowledge. Deep integration of deep learning, graph deep learning, transfer learning and meta-learning is the development trend of knowledge graphs, which can be used to construct large-scale knowledge graphs of all types and high coverage, and realize more profound knowledge reasoning. The road to artificial intelligence explained.
Excerpted from the translator's preface of "Knowledge Graph: Methods, Tools and Cases"
