/Hadoop-MapReduce

implementation of algorithms with Hadoop MapReduce

Primary LanguageJava

Hadoop-MapReduce

implementation of algorithms with Hadoop MapReduce

PageRank

Top 10 PageRank

Step 1： Sum contributions of links and then multiply β to solve 'spider traps'。

Step 1

Step 2： Renormalize r to re-insert the leaked PageRank casued by 'dead-ends'。

Step 2

LSH

K-means

Comparison Comparison_1

Figures show the cost of two initialization strategies and two metrics

c1：randomly initialized centroids

c2：centroids which are as far apart as possible

K-means_algo

Apriori

result

We set the support threshold as 60 and get four 4-frequent itemsets. Based on these itemsets, associasion rules can be generated.