[英]With Bluemix Retrieve&Rank, How do we implement a system to continuously learn?
With reference to the Web page below, using the Retrieve & Rank service of IBM Bluemix, we are creating a bot that can respond to inquiries. 参考以下Web页面,使用IBM Bluemix的Retrieve&Rank服务,我们正在创建一个可以响应查询的机器人。
Question: After learning the ranker once, based on the user's response to the inquiry, how can we construct a mechanism to continuously learn and improve response accuracy? 问:一次学习了排名后,根据用户对查询的响应,我们如何构建一种持续学习并提高响应准确性的机制?
Assumption: Because there was no API of R&R service to continuously learn from the inquiry response result of the user, tuning the GroundTruth file, I suppose that it is necessary to periodically perform such a process as training the ranker again. 假设:由于没有R&R服务的API可以从用户的查询响应结果中不断学习,因此需要调整GroundTruth文件,因此我认为有必要定期执行这样的过程,即再次训练排名者。
Tuning contents of assumed GT file: 调优假定的GT文件的内容:
In order to continuously learn, you will want to do the following: 为了不断学习,您需要执行以下操作:
NOTE: Be sure to validate that new updates to the ranker data improves the overall system performance. 注意:请确保验证对ranker数据的新更新可以提高整体系统性能。 k-fold validation is a great way to measure this.
k倍验证是衡量这一点的好方法。
All in all, learning is a continuous process that should be repeated indefinitely or until system performance is deemed suffcient. 总而言之,学习是一个连续的过程,应该无限期地重复学习,或者直到系统性能被认为足够为止。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.