知识卡片
双向编码需看到两侧文本
双向循环网络分别沿正向和反向处理输入,再组合表示,让某位置同时利用前后文本。对完整评论可以这样编码;在线逐字预测时未来文本尚不存在,不能借双向结构声称能预知未来。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 电影评论情感分析
原书出处 · 1 处
- · 查看原书依据
data versatility, requiring their data to be fixed. Moreover, Standard Recurrent Neural Networks (RNNs) have constraints as future data information can’t emerge out of the current state. Notwithstanding what may be normal, Bi-directional recurrent neural networks don’t need their data to be fixed. Furthermore, their future data information is reachable from the current state. The Bi-Directional Long Short-Term Memory has been shown in Fig. (6).
参考来源
- 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第07章《电影评论情感分析》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications/_epub-src/OEBPS/Text/9789815136746-C7.html)
- 结论依据:摘录中关于双向编码需看到两侧文本的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:data versatility, requiring their data to be fixed. Moreover, Standard Recurrent Neural Networks (RNNs) have constraints as future data information can’t emerge out of the current state. Notwithstanding what may be normal, Bi-directional recurrent neural networks don’t need their data to be fixed. Furthermore, their future data information is reachable from the current state. The Bi-Directional Long Short-Term Memory has been shown in Fig. (6).