/word2vec

word2vec的Java并行实现

Primary LanguageJava

word2vec的Java并行实现

the Parallel Implementation of word2vec in Java

参考了其它语言的并行实现方法,包括:

使用Java 7编写,读取的语料需先行分词完毕,并以空格分隔,由于低频词(<5)会被过滤,故建议使用较大的文本数据集。,例子见TestWord2Vec

write in Java 7, the word in input text should be segmented by one space。This program will filter the low frequency(<5) word, so please use large text dataset. Demo can be found in TestWord2Vec

word2vec