关于数据处理的问题

[Savant-HO]

          学长你好！

我是大三的学生，最近也在研究关于DDI的问题.
我想问一下您在论文中提到说将小于10个DDI的罕见事件删除了，但是我在您的event.db中的发现有的event是低于10的。比如如下：(这是我整理你的event.db中收集到的）
('the vasopressor activities increase', 9), ('a decrease in the absorption resulting in a reduced serum concentration and potentially a decrease in efficacy cause', 9), ('the hyponatremic activities increase', 9), ('the excretion rate which could result in a lower serum level and potentially a reduction in efficacy increase', 7), ('The risk or severity of myopathy and rhabdomyolysis increase', 7), ('the neuromuscular blocking activities decrease', 6), ('The risk of a hypersensitivity reaction increase', 5), ('the hypocalcemic activities increase', 5), ('the vasodilatory activities increase', 5), ('the myelosuppressive activities increase', 5), ('the hyperglycemic activities increase', 5)
是我理解错了还是怎么回事呢?

Originally posted by @Savant-HO in #14 (comment)

[Savant-HO]

学长！还有一个问题：您的论文中提到“remove DDIs associated
with more than one event” 这个意思是将两个药物节点中出现的多个关联的线删到只剩一条线关联吗？如果是这个意思的话，您是出于什么考虑这样处理呢？是因为您论文中下游任务是多分类任务，所以就这样处理数据集吗？那如果我把DDI任务看作是多标签任务，是不是就不需要做这样的处理呢？
非常希望您的指导和回答！感谢感谢！