GraphLab Learning Records

1. 实现原理部分解读

   1. 与Hadoop的区别：Hadoop为什么不好，而要提出GraphLab框架
      1. Hadoop的并行化原理 v.s. GraphLab的并行化原理
      2. 运行试验对比

2. GraphLab原理细化

3. code reading

   collaborative filtering/als.cpp find by search "iver"

   collaborative filtering/nmf.cpp see the description

4. 代码实现

   Affinity Propagation Clustering
   单源最短路径
   统计图的顶点的度和相同度的顶点数目

   参考文献

   Distributed GraphLab: A Framework for Machine Learning and Data Mining in the Cloud