Special Issue on Knowledge-Graph-Enabled Artificial Intelligence (KG-enabled AI) Submission Date: 2022-01-20 Knowledge Graphs (KGs) and their underlying semantic technologies are modern implementations of symbolic Artificial Intelligence (AI). In recent years, an increasing number of KGs have been constructed and published, by both academia and industry, such as DBpedia, YAGO, Freebase, Wikidata, Google Knowledge Graph, Microsoft Satori, Facebook Entity Graph, and others. For human intelligence, it is widely recognized that thinking, learning, logical reasoning, and language comprehension are all based on knowledge stored in human brains though the details are still under research in neuroscience. However, the current AI research and development are mostly focusing on perception, recognition, and judgement by primarily using learning-based approaches and techniques, especially deep learning. In both areas of symbolic and statistical AI, an emerging trend is to take full advantage of KGs – in different forms – in order to make AI-based systems not only intelligent but also knowledgeable. One thriving research direction is representation learning for KGs, which aims to encode the rich semantics of KGs into a low-dimensional embedding space to benefit various downstream learning tasks. In multi-modal machine learning, KGs can be leveraged to bridge the gap between visual and textual artifacts. In natural language processing, KG-based Question Answering (KGQA) is devoted to more efficiently answer questions in natural languages with the help of facts from KGs. Moreover, in the field of eXplainable Artificial Intelligence (XAI), KGs offer new possibilities to address the issue of explainability for AI, which is a requirement in some critical AI applications. Therefore, researchers and practitioners in AI and related areas have been investing increasing effort into the study of KG-enabled AI.

This Special Issue aims to provide a forum for the dissemination of recent advances in research and development in areas relating to KGs and their integration with AI and ML. Challenges include KG-enabled machine learning, KG-enabled computer vision, KG representation learning, KG-enabled natural language processing, KG- enabled question answering, and KG-enabled recommendation. To address these challenges, we invite original research papers that report on state-of-the-art and recent achievements in terms of KG-enabled AI.

The Special Issue will solicit high-quality submissions from researchers world-wide that are active in the areas of knowledge engineering, machine learning, pattern recognition, natural language processing, data mining, or data management. Overall, we are interested in receiving papers on topics that include, but are not limited to:

- KG representation learning and applications

- KG-enabled neural network and deep learning

- KG-enabled computer vision and pattern recognition

- KG-enabled natural language processing and understanding

- KG-enabled question answering systems

- KG-enabled recommendation systems

- KG-enabled explainable AI

- KG data management infrastructures

- KG-enabled data mining and analysis

- KG-enabled AI systems

- KG-enabled AI applications

登录用户可以查看和发表评论, 请前往  登录 或  注册

SCHOLAT.com 学者网
免责声明 | 关于我们 | 用户反馈