Skip to content

Commit

Permalink
word2vec补充
Browse files Browse the repository at this point in the history
  • Loading branch information
shataowei committed Nov 27, 2019
1 parent 5d21306 commit 7be5f1d
Show file tree
Hide file tree
Showing 3 changed files with 43 additions and 28 deletions.
67 changes: 40 additions & 27 deletions .idea/workspace.xml

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

1 change: 1 addition & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -341,6 +341,7 @@
# 推荐
- DIN
- DeepFM
-
- YoutubeNet
- Wide&Deep
- MLR
Expand Down
3 changes: 2 additions & 1 deletion 自然语言处理/Word2Vec.md
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,8 @@
- 负采样的核心思想是:利用负采样后的输出分布来模拟真实的输出分布

# word2vec两种方法各自的优势?
**Mikolov 的原论文,Skip-gram 在处理少量数据时效果很好,可以很好地表示低频单词。而 CBOW 的学习速度更快,对高频单词有更好的表示**
- **Mikolov 的原论文,Skip-gram 在处理少量数据时效果很好,可以很好地表示低频单词。而 CBOW 的学习速度更快,对高频单词有更好的表示**
- Skip-gram的时间复杂度是o(kv),CBOW的时间复杂度o(v)

# 怎么衡量学到的embedding的好坏?
- 从item2vec得到的词向量中随机抽出一部分进行人工判别可靠性。即人工判断各维度item与标签item的相关程度,判断是否合理,序列是否相关
Expand Down

0 comments on commit 7be5f1d

Please sign in to comment.