Scribd Logo
Upload

Welcome to Scribd!

  • 35 views

    0.9/ Mahout-Distribution-0.9.zip Ml-100k.zip: Download Latest

    Uploaded by

    Sankeerthana Likhitha
    This document provides instructions for setting up and running an item-based recommender system using the Apache Mahout library. It includes steps to download necessary libraries and datasets, convert the dataset to CSV format, and build the recommender model using different similarity metrics. The recommender then provides the top 10 recommendations for specific items using each of the Pearson correlation, Euclidean distance, Tanimoto coefficient, and log likelihood similarity metrics.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    0.9/ Mahout-Distribution-0.9.zip Ml-100k.zip: Download Latest

    Uploaded by

    Sankeerthana Likhitha
    35 views5 pages
    This document provides instructions for setting up and running an item-based recommender system using the Apache Mahout library. It includes steps to download necessary libraries and datasets, convert the dataset to CSV format, and build the recommender model using different similarity metrics. The recommender then provides the top 10 recommendations for specific items using each of the Pearson correlation, Euclidean distance, Tanimoto coefficient, and log likelihood similarity metrics.

    Original Description:

    item based recommender doc

    Original Title

    Documentaion on ITR

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides instructions for setting up and running an item-based recommender system using the Apache Mahout library. It includes steps to download necessary libraries and datasets, convert the dataset to CSV format, and build the recommender model using different similarity metrics. The recommender then provides the top 10 recommendations for specific items using each of the Pearson correlation, Euclidean distance, Tanimoto coefficient, and log likelihood similarity metrics.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    35 views5 pages

    0.9/ Mahout-Distribution-0.9.zip Ml-100k.zip: Download Latest

    Uploaded by

    Sankeerthana Likhitha
    This document provides instructions for setting up and running an item-based recommender system using the Apache Mahout library. It includes steps to download necessary libraries and datasets, convert the dataset to CSV format, and build the recommender model using different similarity metrics. The recommender then provides the top 10 recommendations for specific items using each of the Pearson correlation, Euclidean distance, Tanimoto coefficient, and log likelihood similarity metrics.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    of 5

    Documentation on

    Mahout-Itembased recommender

    Open Eclipse.
    Open File NewJavaProject.
    Give Project name as ItemRecommender.
    Now create a Library folder to store our jar files
    and library files, and create a new folder to add our
    Data.
    All the Libraries we need is Apache Mahout
    Libraries -https://mahout.apache.org/general/downloads.html
    Download Latest --0.9/-- mahout-distribution-0.9.zip
    -download this Zipfile.
    From http://grouplens.org/datasets/movielens/--ml100k.zip download this Zipfile.
    From http://slf4j.org/download.html -- slf4j1.7.6.zip download this Zipfile.
    From https://code.google.com/p/guava-libraries/ -- guava16.0.1.jar download this Zipfile.
    From http://commons.apache.org/proper/commonsmath/download_math.cgi -- commons-math3-3.2-bin.zip

    download this Zipfile.


    Now add below files to library folder.
    1. Commons-math3-3.2.jar
    2. guava 15.0.jar
    3. mahout-core-0.9.jar, mahout-integration-0.9.jar,
    mahout-math-0.9.jar
    4. slf4j-api-1.7.6.jar, slf4j-nop-1.7.6.jar
    We need to build the path for all library folder jar
    files

    under the refrences libraries.


    Now add DATA sets from Grouplens site to Data
    folder which we created in our JavaProject.
    Part A:
    Converting Tab separated file into csv File: The Dataset is u.data file which is in tab
    separated format .We are now converting it to
    Comma(,) separated i.e., to .csv and then we are
    running Itembased recommender engine for a
    specific Items or for all Items which we desire.
    Now create a New Package as
    com.convertmovielens and New Class as
    MovieDataConvert for Itemrecommender.
    Code :

    Outputs:

    Part B:
    1.Using PearsonCorrelationSimilarity Metric:
    For PearsonCorrelationSimilarity Metric use
    ItemSimilarity sim = new PearsonCorrelationSimilarity(dm);

    2.Using EuclidianDistanceSimilarity Metric:


    For EuclidianDistanceSimilarity Metric use

    ItemSimilarity sim = new EuclidianDistanceSimilarity(dm); .


    3.Using TanimotoCoefficientSimilarity Metric:
    For TanimotoCoefficientSimilarity Metric use

    ItemSimilarity sim = new TanimotoCoefficientSimilarity(dm); .


    4.Using LogLikelihoodSimilarity Metric:
    For EuclidianDistanceSimilarity Metric use

    ItemSimilarity sim = new LogLikelihoodSimilarity(dm);

    Outputs:
    Pearson
    Correlation Similarity
    Metric

    Euclidian
    Distance
    Similarity Metric

    Tanimoto
    Coefficient
    Similarity Metric

    LogLikelihoodSimila
    rity Metric

    1,757,1.0
    1,341,1.0
    2,75,1.0
    2,74,1.0
    3,344,1.0
    3,115,1.0
    4,643,1.0
    4,360,1.0
    5,34,1.0
    5,6,1.0

    1,757,1.0
    1,711,1.0
    2,359,1.0
    2,308,1.0
    3,377,1.0
    3,344,1.0
    4,814,1.0
    4,711,1.0
    5,533,1.0
    5,314,1.0

    1,50,0.5825688
    1,121,0.5537919
    2,233,0.53614455
    2,231,0.4918033
    3,17,0.2638889
    3,123,0.2576687
    4,385,0.46830985
    4,202,0.46407187
    5,218,0.35978836
    5,672,0.31304348

    1,117,0.9953521
    1,151,0.9953065
    2,403,0.9964998
    2,233,0.9964557
    3,405,0.99037176
    3,235,0.9893157
    4,56,0.99627966
    4,174,0.99601305
    5,218,0.99432045
    5,98,0.9922024

    You might also like