字袋模型

字袋模型（英文：bag-of-words model，BoW model）係自然語言處理同資訊提取入面嘅一種做法，指嘅係將一段文字當做由啲字組成嘅多重集，忽略文法甚至啲字嘅次序。

例如以下呢句嘢：

John likes to watch movies. Mary likes movies too.

用 BoW 方法表示嘅話會變成噉：

"John","likes","to","watch","movies","Mary","likes","movies","too"

睇埋

• N-gram
• 字嵌入

呢篇同語言學有關嘅文章係楔位文。 歡迎幫維基百科擴寫佢。