China's First Natural Language-based AI ChatBot Trader

James Y. Zhang, Zhi Li, Hao Fang, Jun Wu, Zhongnan Shen,Jing Zheng,Wei Chu, Weiping Duan,Peng Xu

COMPANION OF THE WORLD WIDE WEB CONFERENCE, WWW 2023(2023)

引用 1|浏览36
暂无评分
摘要
Repo (repurchase agreement) trading provides easy access to short-term financing secured by a pledge of collateral and plays an important role in the global financial system. However, repo traders face many tough challenges in their job, from managing complex financial transactions to keeping up with changing market trends and regulations in the complex financial transactions involved. Besides the difficult and tedious processes that take a lot of time and energy, repo traders need to keep up to date with various laws, regulations, and financial trends that may affect their job, worsened by the exposure to a variety of market risks. As the leader of the FinTech industry, Ant Group launched a new initiative to alleviate the affliction of the repo traders at MyBank1. By leveraging many existing platform technologies, such as AI ChatBot and forecasting platforms, and with the collective work of various engineering groups, we are able to create a ChatBot that communicates with other human traders in natural language and create electronic contracts based on the negotiated terms, equipped with proper trading strategies based on forecasting results. The fully automatic workflow not only frees our trader from tedious routines, but also reduces potential human errors. At the same time, it enables refined portfolio and risk management, while opening up the possibility to apply neural network-based trading strategies, and yielding greater returns comparing to traditional workflow reliant on human experiences. Our system has evolved beyond just providing services to our own traders, to now a fully commercialized product, covering other types of interbank trading.
更多
查看译文
关键词
chatbot,natural language,repurchase agreement,repo,interbank
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要