Index

A

• Abalone dataset
  • URL / Feature construction
• accumulators
  • about / Apache Spark
• adjusted R-square
  • about / Adjusted R-square
• AUC
  • about / ROC curve and AUC, Area-based accuracy measure
• automated feature extraction
  • about / Feature engineering

B

• backward selection
  • about / Backward selection
• bagging
  • about / Bagging
• behavioral-based segmentation
  • about / Customer segmentation
• bias-variance trade-off
  • about / Bias-variance trade-off
• bigram
  • about / n-grams
• binarization
  • about / Binarization
• binning
  • about / Binning
  • unsupervised binning / Binning
  • supervised binning / Binning
• boosting
  • about / Boosting
• broadcast variables
  • about / Apache Spark

C

• canopy clustering
  • about / Canopy clustering
  • command-line options / Canopy clustering
• categorical features
  • about / Categorical features
  • categories, merging / Merging categories
  • categories, converting to binary variables / Converting to binary variables
  • categories, converting to continuous variables / Converting to continuous variables
• centroids
  • determining / Deciding the initial centroid
  • random points, generating / Random points
  • random input points, selecting / Points from the dataset
  • partition, by range / Partition by range
  • canopy centroids, using / Canopy centroids
• churn analytics
  • about / Churn analytics
  • data, obtaining / Getting the data
  • data exploration / Data exploration
  • feature engineering / Feature engineering
  • model training and validation phase / Model training and validation
• classification
  • about / A classification example, Supervised learning, Classification
  • common workflow / A classification example
  • confusion matrix / Confusion matrix
  • AUC / ROC curve and AUC
  • ROC curve / ROC curve and AUC
• cluster analysis
  • about / Cluster analysis
  • objective / Objective
  • feature representation / Feature representation
  • clustering algorithm, using / Algorithm for clustering
  • stopping criteria / A stopping criteria
• clustering
  • about / A clustering example, Clustering
  • internal evaluation / The internal evaluation
  • external evaluation / The external evaluation
• clustering, using Java code
  • example / A Mahout Java example
  • k-means, using / k-means
  • cluster evaluation / Cluster evaluation
• clustering, using Mahout command line
  • example / A Mahout command-line example
  • data, obtaining / Getting the data
  • URL, for dataset / Getting the data
  • data, preprocessing / Preprocessing the data
  • k-means / k-means
  • canopy clustering / Canopy clustering
  • fuzzy k-means / Fuzzy k-means
  • streaming k-means / Streaming k-means
• clustering algorithm
  • k-means / k-means
  • canopy clustering / Canopy clustering
  • fuzzy k-means / Fuzzy k-means
• collaborative filtering
  • about / Collaborative filtering, Collaborative filtering
  • cold start / Cold start
  • scalability / Scalability
  • sparsity / Sparsity
  • similarity / Similarity measures
  • recommender systems / Evaluating recommender
  • preferences / Inferring preferences
• column normalization
  • about / Column normalization
  • rescaling / Rescaling
  • standardization / Standardization
• command line
  • about / Mahout command line
  • clustering example / A clustering example
  • Reuter's raw data file / Reuter's raw data file
  • example, for k-means clustering / Reuter's raw data file
  • classification example / A classification example
• command line, Mahout
  • extending / Extending the command line of Mahout
  • used, for implementing LDA / LDA using the Mahout command line
• confusion matrix
  • about / Confusion matrix
• content-based filtering
  • about / Content-based filtering
• continuous features
  • about / Continuous features
  • binning / Binning
  • binarization / Binarization
  • feature standardization / Feature standardization
  • mathematical transformations / Mathematical transformations
• cosine distance measure
  • about / Cosine distance measure
• customer segmentation
  • about / Customer segmentation
  • value-based segmentation / Customer segmentation
  • behavioral-based segmentation / Customer segmentation
  • demographic-based segmentation / Customer segmentation
  • preprocessing / Preprocessing

D

• data exploration, churn analytics
  • R, installing / Installing R
• Davies-Bouldin index
  • about / The Davies–Bouldin index
• demographic-based segmentation
  • about / Customer segmentation
• dense vector
  • about / Initializing a vector inline
• development environment
  • setting up / Setting up the development environment
  • Maven, configuring / Configuring Maven
  • Mahout, configuring / Configuring Mahout
  • Eclipse, configuring / Configuring Eclipse with the Maven plugin and Mahout
• dimensionality reduction
  • about / Feature engineering, Dimensionality reduction
• distance measure
  • about / A notion of similarity and dissimilarity
  • Euclidean distance measure / Euclidean distance measure
  • squared Euclidean distance measure / Squared Euclidean distance measure
  • Manhattan distance measure / Manhattan distance measure
  • cosine distance measure / Cosine distance measure
  • Tanimoto distance measure / Tanimoto distance measure
• document indexing
  • about / Document indexing
• DRM
  • about / Basics of Mahout Scala DSL
• Dunn index
  • about / The Dunn index

E

• Eclipse
  • configuring / Configuring Eclipse with the Maven plugin and Mahout
  • Mahout source code, importing / Importing the Mahout source code into Eclipse
• embedded feature selection
  • about / Embedded feature selection
• Euclidean distance measure
  • about / Euclidean distance measure
• Euclidean distance similarity
  • about / Euclidean distance similarity
• evaluation
  • about / Evaluation
  • bias-variance trade-off / Bias-variance trade-off
  • function complexity / Function complexity and amount of training data
  • training data consideration / Function complexity and amount of training data
  • dimensionality, of input space / Dimensionality of the input space
  • noise, in data / Noise in data
• external evaluation, clustering
  • about / The external evaluation
  • Rand index / The Rand index
  • F-measure / F-measure

F

• F-measure
  • about / F-measure
• feature
  • about / Feature engineering
• feature construction
  • about / Feature construction
  • categorical features / Categorical features
  • continuous features / Continuous features
• feature engineering
  • about / Feature engineering
  • manual feature construction / Feature engineering
  • automated feature extraction / Feature engineering
  • feature selection / Feature engineering
  • dimensionality reduction / Feature engineering
• feature extraction
  • about / Feature extraction
  • techniques / Feature extraction
• feature extraction, customer segmentation
  • day calls / Day calls
  • evening calls / Evening calls
  • international calls / International calls
  • files, preprocessing / Preprocessing the files
• feature representation
  • about / Feature representation
  • feature normalization / Feature normalization
  • similarity / A notion of similarity and dissimilarity
  • dissimilarity / A notion of similarity and dissimilarity
  • distance measure / A notion of similarity and dissimilarity
• feature selection
  • about / Feature engineering, Feature selection
  • filter-based feature selection / Filter-based feature selection
  • wrapper-based feature selection / Wrapper-based feature selection
  • embedded feature selection / Embedded feature selection
• feature standardization
  • about / Feature standardization
  • rescaling / Rescaling
  • mean standardization / Mean standardization
  • scaling / Scaling to unit norm
• feature transformation
  • about / Feature transformation derived from the problem domain
  • ratios / Ratios
  • frequency / Frequency
  • aggregate transformations / Aggregate transformations
  • normalization / Normalization
• filter-based feature selection
  • about / Filter-based feature selection
• fixed size neighborhood
  • about / Fixed size neighborhood
• forward selection
  • about / Forward selection
• FP-Growth
  • about / Frequent pattern mining
• FP Tree
  • about / Frequent pattern mining
  • building / Building FP Tree
  • constructing / Constructing the tree
  • frequent patterns, identifying / Identifying frequent patterns from FP Tree
• frequent pattern mining
  • about / Frequent pattern mining
  • rules, identifying / Measures for identifying interesting rules
  • considerations / Things to consider
  • FP-Growth / Frequent pattern mining
  • FP Tree / Frequent pattern mining
  • implementing, with Mahout / Frequent pattern mining with Mahout
  • Mahout command line, extending / Extending the command line of Mahout
  • data, obtaining / Getting the data
  • data description / Data description
  • implementing, with Mahout API / Frequent pattern mining with Mahout API
• frequent pattern mining (FPM)
  • about / Frequent pattern mining with Mahout
• frequent pattern mining, considerations
  • actionable rules / Actionable rules
  • association, determining / What association to look for
• frequent pattern mining, rules
  • identifying / Measures for identifying interesting rules
  • support / Support
  • confidence / Confidence
  • lift / Lift
  • conviction / Conviction
• frequent pattern mining, with Mahout API
  • MapReduce execution / MapReduce execution
  • linear execution / Linear execution
  • results, formatting / Formatting the results and computing metrics
  • metrics, computing / Formatting the results and computing metrics
• fuzzy k-means
  • about / Fuzzy k-means
  • fuzzy factor, deciding / Deciding the fuzzy factor
  • command-line options / Fuzzy k-means

H

• Hadoop
  • URL, for configuring / Setting up the development environment
• Hadoop Distributed File System (HDFS) / Reuter's raw data file
• holdout-set validation
  • about / Holdout-set validation

I

• in-core types
  • about / In-core types
  • vector / Vector
  • matrix / Matrix
• in-memory execution
  • about / Parallel versus in-memory execution mode
  • versus parallel execution / Parallel versus in-memory execution mode
• installation, R
  • about / Installing R
• inter-cluster distance
  • about / The inter-cluster distance
• internal evaluation, clustering
  • about / The internal evaluation
  • intra-cluster distance / The intra-cluster distance
  • inter-cluster distance / The inter-cluster distance
  • Davies-Bouldin index / The Davies–Bouldin index
  • Dunn index / The Dunn index
• intra-cluster distance
  • about / The intra-cluster distance
• item-based recommender system
  • about / Item-based recommender system
  • example / Mahout code example
  • recommender, building / Building the recommender
  • recommender, evaluating / Evaluating the recommender

K

• K-fold cross validation
  • about / K-fold cross validation
• k-means
  • about / k-means
  • number of clusters, determining / Deciding the number of clusters
  • initial centroid, determining / Deciding the initial centroid
  • advantages and disadvantages / Advantages and disadvantages
  • command-line options / k-means

L

• LDA
  • used, for topic modeling / Topic modeling using LDA
  • about / Topic modeling using LDA
  • implementing, Mahout command line used / LDA using the Mahout command line
• linear regression
  • with Mahout Spark / Linear regression with Mahout Spark
• log-likelihood similarity
  • about / Log-likelihood similarity
• log-likelihood test
  • about / n-grams

M

• machine learning
  • supervised learning / Supervised learning
  • unsupervised learning / Unsupervised learning
  • recommender system / Recommender system
  • model efficacy / Model efficacy
• Mahout
  • advantages / Why Mahout
  • use case / When Mahout
  • development environment, setting up / Setting up the development environment
  • configuring / Configuring Mahout
  • URL / Configuring Mahout, The classification job
  • source code, importing into Eclipse / Importing the Mahout source code into Eclipse
  • frequent pattern mining, implementing / Frequent pattern mining with Mahout
  • Spark, configuring / Configuring Spark with Mahout
• Mahout, advantages
  • simple techniques / Simple techniques and more data is better
  • better data collection / Simple techniques and more data is better
  • sampling / Sampling is difficult
  • license / Community and license
  • community / Community and license
• Mahout, use case
  • data too large for single machine / Data too large for single machine
  • data already on Hadoop / Data already on Hadoop
  • algorithms implemented in Mahout / Algorithms implemented in Mahout
• Mahout API
  • about / Mahout API – a Java program example
  • dataset / The dataset
  • frequent pattern mining, implementing / Frequent pattern mining with Mahout API
• Mahout Scala DSL
  • about / Basics of Mahout Scala DSL
  • imports / Imports
• Mahout Spark
  • DRM / Spark Mahout basics
  • linear regression / Linear regression with Mahout Spark
• Mahout Spark, DRM
  • Spark context, initializing / Initializing the Spark context
  • optimizer actions, performing / Optimizer actions
  • computational actions / Computational actions
  • caching, in Spark's block manager / Caching in Spark's block manager
• Mahout trunk
  • URL, for latest version / Configuring Spark with Mahout
• Manhattan distance measure
  • about / Manhattan distance measure
• manual feature construction
  • about / Feature engineering
• MapReduce
  • limitations / Moving beyond MapReduce
• mathematical transformations
  • about / Mathematical transformations
• matrix
  • about / Matrix
  • initializing / Initializing the matrix
  • elements, accessing / Accessing elements of a matrix
  • column, setting / Setting the matrix column
  • copy by reference / Copy by reference
• Maven
  • configuring / Configuring Maven
  • URL / Configuring Maven
• mean absolute error
  • about / Mean absolute error
• model, training
  • bagging / Bagging
  • boosting / Boosting
• model efficacy
  • about / Model efficacy
  • classification / Classification
  • regression / Regression
  • recommendation system / Recommendation system
  • clustering / Clustering
• model training and validation phase, churn analytics
  • logistic regression / Logistic regression
  • adaptive logistic regression / Adaptive logistic regression
  • random forest / Random forest

N

• n-grams
  • about / n-grams
• normalization
  • about / Normalization
• normalization, feature
  • about / Feature normalization
  • row normalization / Row normalization
  • column normalization / Column normalization

O

• ordinary least square (OLS)
  • about / Linear regression with Mahout Spark

P

• p-norm
  • about / Normalization
• parallel execution
  • about / Parallel versus in-memory execution mode
  • versus in-memory execution / Parallel versus in-memory execution mode
• patsy library
  • about / Converting to binary variables
• Pearson correlation similarity
  • about / Pearson correlation similarity
• precision
  • about / Precision and recall
• preferences
  • about / Inferring preferences
• preprocessing, customer segmentation
  • feature extraction / Feature extraction
  • clusters, creating with Fuzzy k-means / Creating the clusters using fuzzy k-means
  • clustering, with k-means / Clustering using k-means
  • evaluation / Evaluation

R

• R
  • installing / Installing R
  • summary statistics, viewing / Summary statistics
  • correlation, calculating / Correlation
• R-square
  • about / R-square
• Rand index
  • about / The Rand index
• recommendation system
  • about / Recommendation system
  • score difference / Score difference
  • precision and recall / Precision and recall
• recommender system
  • about / Recommender system, Evaluating recommender
  • collaborative filtering / Collaborative filtering
  • content-based filtering / Content-based filtering
  • evaluating / Evaluating recommender
  • user-based recommender system / User-based recommender system
  • item-based recommender system / Item-based recommender system
• recursive feature elimination
  • about / Recursive feature elimination
• regression
  • about / Supervised learning, Regression
  • mean absolute error / Mean absolute error
  • root mean squared error (RMSE) / Root mean squared error
  • R-square / R-square
  • adjusted R-square / Adjusted R-square
• relative squared error (RSE)
  • about / Root mean squared error
• rescaling, feature
  • about / Rescaling
• resilient distributed dataset (RDD)
  • about / Apache Spark
• ROC curve
  • about / ROC curve and AUC
  • used, for evaluating classifier / Evaluating classifier using the ROC curve
  • area-based accuracy measure / Area-based accuracy measure
  • Euclidian distance comparison / Euclidian distance comparison
  • example / Example
• ROC graphs
  • features / Features of ROC graphs
• root mean squared error (RSME)
  • about / Root mean squared error
• row normalization
  • about / Row normalization

S

• score difference
  • about / Score difference
• shared variables
  • about / Apache Spark
  • broadcast variables / Apache Spark
  • accumulators / Apache Spark
• similarity
  • about / Similarity measures
  • Pearson correlation similarity / Pearson correlation similarity
  • Euclidean distance similarity / Euclidean distance similarity
  • computing, without preference value / Computing similarity without a preference value
  • Tanimoto coefficient similarity / Tanimoto coefficient similarity
  • log-likelihood similarity / Log-likelihood similarity
• source code, Mahout
  • importing, into Eclipse / Importing the Mahout source code into Eclipse
• Spark
  • about / Apache Spark
  • configuring, with Mahout / Configuring Spark with Mahout
  • Mahout Scala DSL / Basics of Mahout Scala DSL
• sparse vector
  • about / Initializing a vector inline
• Squared Euclidean distance measure
  • about / Squared Euclidean distance measure
• standard generalized markup language (SGML) / Reuter's raw data file
• standardization, feature
  • about / Standardization
• stemming
  • about / Stemming
• stop words
  • removing / Stop word removal
• streaming k-means
  • command-line options / Streaming k-means
• subversion (svn)
  • about / Configuring Spark with Mahout
• supervised binning
  • about / Binning
• supervised learning
  • about / Supervised learning
  • regression / Supervised learning
  • classification / Supervised learning
  • objective, determining / Determine the objective
  • training data, determining / Decide the training data
  • training set, creating / Create and clean the training set
  • training set, cleaning / Create and clean the training set
  • feature extraction / Feature extraction
  • model, training / Train the models
  • validation / Validation
  • evaluation / Evaluation

T

• Tanimoto coefficient similarity
  • about / Tanimoto coefficient similarity
• Tanimoto distance measure
  • about / Tanimoto distance measure
• term frequency (TF)
  • about / Document indexing
• text, categorizing
  • about / Categorizing text
  • dataset / The dataset
  • dataset, URL / The dataset
  • feature extraction / Feature extraction
  • example / The classification job
• text, clustering
  • about / Clustering text
  • dataset / The dataset
  • feature extraction / Feature extraction
  • example / The clustering job
• text, preprocessing
  • tokenization / Tokenization
  • stop word removal / Stop word removal
  • stemming / Stemming
  • example / Preprocessing example
• text analytics
  • about / Text analytics
  • VSM / Vector space model
• TF-IDF weighting
  • about / TF-IDF weighting
• threshold-based neighborhood
  • about / Threshold-based neighborhood
• tokenization
  • about / Tokenization
• topic modeling
  • LDA, using / Topic modeling using LDA
• trigrams
  • about / n-grams

U

• unigram
  • about / n-grams
• unsupervised binning
  • about / Binning
• unsupervised learning
  • about / Unsupervised learning
  • cluster analysis / Cluster analysis
  • frequent pattern mining / Frequent pattern mining
• user-based recommender system
  • about / User-based recommender system
  • user neighborhood / User neighborhood
  • dataset / The dataset
  • URL, for dataset / The dataset
  • example / Mahout code example
  • recommender, building / Building the recommender
  • recommender, evaluating / Evaluating the recommender
• user neighborhood
  • about / User neighborhood
  • fixed size neighborhood / Fixed size neighborhood
  • threshold-based neighborhood / Threshold-based neighborhood

V

• validation
  • about / Validation
  • holdout-set validation / Holdout-set validation
  • K-fold cross validation / K-fold cross validation
• value-based segmentation
  • about / Customer segmentation
• vector
  • about / Vector
  • dense vector / Initializing a vector inline
  • initializing / Initializing a vector inline
  • sparse vector / Initializing a vector inline
  • elements, accessing / Accessing elements of a vector
  • element values, setting / Setting values of an element
  • arithmetic operations, performing / Vector arithmetic
  • arithmetic operations, performing with scalar / Vector operations with a scalar
• VSM
  • text, preprocessing / Preprocessing
  • document indexing / Document indexing
  • TF-IDF weighting / TF-IDF weighting
  • n-grams / n-grams
  • normalization / Normalization

W

• wrapper-based feature selection
  • about / Wrapper-based feature selection
  • backward selection / Backward selection
  • forward selection / Forward selection
  • recursive feature elimination / Recursive feature elimination