Hi.
In your paper you said that "we filter the samples with volume values less than 10, which a common practice used in industry and
academy (Yao et al. 2018). Because in the real-world applications, cares with low traffic are of little interest."
Is these sample is abnormal data that seriously affect the prediction accuracy, so you filter them? and How do you filter them ?
Hi. In your paper you said that "we filter the samples with volume values less than 10, which a common practice used in industry and academy (Yao et al. 2018). Because in the real-world applications, cares with low traffic are of little interest."
Is these sample is abnormal data that seriously affect the prediction accuracy, so you filter them? and How do you filter them ?