Q2 <br>(a) <br><br>(b) k=3 is more appropriate. <br>Due to the presence of erroneous data, when k=1, the most recent data will be used for classification. If there is noise, the classification result will be seriously affected, that is, the generalization error will be large. When k=3, more training information will be used for classification to avoid misinformation interference
正在翻译中..