{"version":"1.0","author_url":"https://blog.hatena.ne.jp/momijiame/","description":"\u4ee5\u524d\u306e\u30a8\u30f3\u30c8\u30ea\u3067\u3001Word2Vec \u306e CBOW (ContinuousBagOfWords) \u30e2\u30c7\u30eb\u3092 TensorFlow/Keras \u3067\u5b9f\u88c5\u3057\u305f\u3002 CBOW \u306f\u3001\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8 (\u5468\u8fba\u8a9e) \u304b\u3089\u30bf\u30fc\u30b2\u30c3\u30c8 (\u5165\u529b\u8a9e) \u3092\u63a8\u5b9a\u3059\u308b\u591a\u5024\u5206\u985e\u306e\u30bf\u30b9\u30af\u304c\u8003\u3048\u65b9\u306e\u30d9\u30fc\u30b9\u306b\u306a\u3063\u3066\u3044\u308b\u3002 blog.amedama.jp \u4eca\u56de\u6271\u3046\u306e\u306f\u3001CBOW \u3068\u5bfe\u3092\u6210\u3059\u30e2\u30c7\u30eb\u306e Skip Gram \u3092\u30d9\u30fc\u30b9\u306b\u3057\u305f SGNS (Skip Gram with Negative Sampling) \u306b\u306a\u308b\u3002 Skip Gram \u3067\u306f\u3001CBOW \u3068\u306f\u53cd\u5bfe\u306b\u30bf\u30fc\u30b2\u30c3\u30c8 (\u5165\u529b\u8a9e) \u304b\u3089\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8 (\u5468\u8fba\u8a9e) \u3092\u63a8\u5b9a\u3059\u308b\u591a\u5024\u5206\u985e\u2026","image_url":"https://m.media-amazon.com/images/I/512wqsxOrsL.jpg","provider_url":"https://hatena.blog","url":"https://blog.amedama.jp/entry/tf-keras-word2vec-sgns","width":"100%","height":"190","provider_name":"Hatena Blog","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Fblog.amedama.jp%2Fentry%2Ftf-keras-word2vec-sgns\" title=\"Python: TensorFlow/Keras \u3067 Word2Vec \u306e SGNS \u3092\u5b9f\u88c5\u3057\u3066\u307f\u308b - CUBE SUGAR CONTAINER\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","published":"2021-02-23 01:55:01","blog_url":"https://blog.amedama.jp/","title":"Python: TensorFlow/Keras \u3067 Word2Vec \u306e SGNS \u3092\u5b9f\u88c5\u3057\u3066\u307f\u308b","categories":["TensorFlow","NLP","\u6a5f\u68b0\u5b66\u7fd2"],"author_name":"momijiame","blog_title":"CUBE SUGAR CONTAINER","type":"rich"}