Item-Based Recommendations with Hadoop

Mahout在MapReduce上实现了Item-Based Collaborative Filtering，这里我尝试运行一下。

1. 安装Hadoop

2. 从下载Mahout并解压

3. 准备数据

   下载1 Million MovieLens Dataset，解压得到ratings.dat，用

   sed 's/:😦0-9{1,}):😦0-9{1})::0-9{1,}$/,\1,\2/' ratings.dat

   处理成需要的格式。

4. 运行

   mahout recommenditembased -s SIMILARITY_LOGLIKELIHOOD -i /path/to/input/file -o /path/to/desired/output -n 25

   参数：

   MAHOUT-JOB: /home/laxe/apple/mahout/mahout-examples-0.11.0-job.jar
   Job-Specific Options:
   --input (-i) input Path to job input directory.
   --output (-o) output The directory pathname for output.
   --numRecommendations (-n) numRecommendations Number of recommendations per user.
   --usersFile usersFile File of users to recommend for.
   --itemsFile itemsFile File of items to recommend for.
   --filterFile (-f) filterFile File containing comma-separated userID,itemID pairs. Used to exclude the item from the recommendations for that user(optional).
   --userItemFile (-uif) userItemFile File containing comma-separated userID,itemID pairs(optional). Used to include only these items into recommendations. Cannot be used together with usersFile or itemsFile.
   --booleanData (-b) booleanData Treat input as without prefvalues.
   --maxPrefsPerUser (-mxp) maxPrefsPerUser Maximum number of preferences considered per user in final recommendation phase.
   --minPrefsPerUser (-mp) minPrefsPerUser Ignore users with less preferences than this in the similarity computation (default: 1).
   --maxSimilaritiesPerItem (-m) maxSimilaritiesPerItem Maximum number of similarities considered per item.
   --maxPrefsInItemSimilarity (-mpiis) maxPrefsInItemSimilarity Max number of preferences to consider per user or item in the item similarity computation phase, users or items with more preferences will be sampled down(default: 500).
   --similarityClassname (-s) similarityClassname Name of distributed similarity measures class to instantiate,
   alternatively use one of the predefined similarities([SIMILARITY_COOCCURRENCE, SIMILARITY_LOGLIKELIHOOD, SIMILARITY_TANIMOTO_COEFFICIENT, SIMILARITY_CITY_BLOCK, SIMILARITY_COSINE, SIMILARITY_PEARSON_CORRELATION, SIMILARITY_EUCLIDEAN_DISTANCE])
   --threshold (-tr) threshold Discard item pairs with a similarity value below this.
   --outputPathForSimilarityMatrix (-opfsm) outputPathForSimilarityMatrix Write the items imilarity matrix to this path(optional).
   --randomSeed randomSeed Use this seed for sampling.
   --sequencefileOutput Write the output into a Sequence File instead of a text file.
   --help (-h) Print out help.
   --tempDir tempDir Intermediate output directory.
   --startPhase startPhase First phase to run.
   --endPhase endPhase Last phase to run specify HDFS directories while running on hadoop; else specify local file system directories.

参考
Introduction to Item-Based Recommendations with Hadoop
mahout分布式：Item-based推荐