使用 Pellet Reasoner 监控推理进度
我目前正在构建一个 OWL 本体，直到最近才使用 Pellet 推理机快速分类。然而，自从引入了几个新类后，推理性能已经慢得像爬行一样。尽管推理器完成并且本体不包含任何不可满足的概念等，但是推理所花费的时间是不可接受的。我目前正在尝试找出可能导致速度放缓的违规班级/班级。
我的问题是：是否可以记录 Pellet 的推理进度？也就是说，是否有可能产生一些输出来记录pellet在某些推理任务上花费了多长时间/跟踪对任何给定类和公理进行推理需要多长时间？如果是的话，有人可以发布一些java代码吗？
预先感谢您的回答！
I am currently constructing an OWL ontology, which - until very recently classified rapidly using the Pellet reasoner. However, since the introduction of several new classes, the reasoning performance has slowed to a crawl. Although the reasoner completes and the ontology does not contain any unsatisfiable concepts etc, the time the reasoning takes is unacceptable. I am currently trying to track down the offending classes/class that may have led to the slowdown.
Here's my question: is it possible to log the reasoning progreess of Pellet? I.e. is it possible to produce some output that will document how long pellet has spent on certain reasoning tasks/traces how long reasoning over any given class and axiom takes? If so, does anyone have some java code they could post up?
Thanks in advance for your answers!
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
Bill Barnhill 为这个问题提供了非常有用的答案关于语义溢出。请过来！
Bill Barnhill has provided an incredibly helpful answer to this question over on Semantic Overflow. Please head on over!
您是否尝试过使用 Pellet 的 lint 工具来诊断推理性能下降的情况？您还可以通过将问题发布到pellet-users 邮件列表上来获得更好的答案，包括导致问题的本体或可用于重现缓慢情况的本体的某些子集。您也没有说明您正在使用哪个版本等。
Did you try using Pellet's lint tool to diagnose the slowndown in reasoning performance? You will also have better luck getting an answer to this question by posting it on the pellet-users mailing list, including the ontology that's causing problems or some subset of it which can be used to reproduce the slowness. You also don't say which version you're using, etc.