Data-Intensive Text Processing with MapReduce

【文件属性】：
文件名称：Data-Intensive Text Processing with MapReduce
文件大小：3.98MB
文件格式：PDF
更新时间：2017-03-02 00:05:12
hadoop mapreduce Data-Intensive Text Processing with MapReduce Jimmy Lin and Chris Dyer Draft of January 27, 2013 Contents Contents ii 1 Introduction 1 1.1 Computing in the Clouds . . . . . . . . . . . . . . . . . . . . . 6 1.2 Big Ideas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.3 Why Is This Dierent? . . . . . . . . . . . . . . . . . . . . . . . 14 1.4 What This Book Is Not . . . . . . . . . . . . . . . . . . . . . . 17 2 MapReduce Basics 19 2.1 Functional Programming Roots . . . . . . . . . . . . . . . . . . 21 2.2 Mappers and Reducers . . . . . . . . . . . . . . . . . . . . . . . 23 2.3 The Execution Framework . . . . . . . . . . . . . . . . . . . . . 27 2.4 Partitioners and Combiners . . . . . . . . . . . . . . . . . . . . 29 2.5 The Distributed File System . . . . . . . . . . . . . . . . . . . . 32 2.6 Hadoop Cluster Architecture . . . . . . . . . . . . . . . . . . . 36 2.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 3 Basic MapReduce Algorithm Design 39 3.1 Local Aggregation . . . . . . . . . . . . . . . . . . . . . . . . . 41 3.2 Pairs and Stripes . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.3 Computing Relative Frequencies . . . . . . . . . . . . . . . . . 56 3.4 Secondary Sorting . . . . . . . . . . . . . . . . . . . . . . . . . 60 3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 4 Inverted Indexing for Text Retrieval 65 4.1 Web Crawling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.2 Inverted Indexes . . . . . . . . . . . . . . . . . . . . . . . . . . 68 4.3 Inverted Indexing: Baseline Implementation . . . . . . . . . . . 70 4.4 Inverted Indexing: Revised Implementation . . . . . . . . . . . 72 4.5 Index Compression . . . . . . . . . . . . . . . . . . . . . . . . . 74 4.6 What About Retrieval? . . . . . . . . . . . . . . . . . . . . . . 81 4.7 Summary and Additional Readings . . . . . . . . . . . . . . . . 84 5 Graph Algorithms 87 5.1 Graph Representations . . . . . . . . . . . . . . . . . . . . . . . 89 5.2 Parallel Breadth-First Search . . . . . . . . . . . . . . . . . . . 90 5.3 PageRank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 5.4 Issues with Graph Processing . . . . . . . . . . . . . . . . . . . 103 5.5 Summary and Additional Readings . . . . . . . . . . . . . . . . 105 6 Processing Relational Data 107 6.1 Relational Joins . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 7 EM Algorithms for Text Processing 113 7.1 Expectation Maximization . . . . . . . . . . . . . . . . . . . . . 116 7.2 Hidden Markov Models . . . . . . . . . . . . . . . . . . . . . . 123 7.3 EM in MapReduce . . . . . . . . . . . . . . . . . . . . . . . . . 135 7.4 Case Study: Word Alignment for Statistical Machine Translation 138 7.5 EM-Like Algorithms . . . . . . . . . . . . . . . . . . . . . . . . 147 7.6 Summary and Additional Readings . . . . . . . . . . . . . . . . 150 8 Closing Remarks 153 8.1 Limitations of MapReduce . . . . . . . . . . . . . . . . . . . . . 153 8.2 Alternative Computing Paradigms . . . . . . . . . . . . . . . . 155 8.3 MapReduce and Beyond . . . . . . . . . . . . . . . . . . . . . . 157 Bibliography 159