MapReduce is a programming model and an associated implementation for processing and generating large data sets. Users specify a map function that processes a key/value pair to generate a set of intermediate key/value pairs, and a reduce function that merges all intermediate values associated with the same intermediate key. Many real world tasks are expressible in this model, as shown in the paper.

Programs written in this functional style are automatically parallelized and executed on a large cluster of commodity machines. The run-time system takes care of the details of partitioning the input data, scheduling the program's execution across a set of machines, handling machine failures, and managing the required inter-machine communication. This allows programmers without any experience with parallel and distributed systems to easily utilize the resources of a large distributed system.

Our implementation of MapReduce runs on a large cluster of commodity machines and is highly scalable: a typical MapReduce computation processes many terabytes of data on thousands of machines. Programmers find the system easy to use: hundreds of MapReduce programs have been implemented and upwards of one thousand MapReduce jobs are executed on Google's clusters every day.

/pdf/mapreduce-simplified-data-processing-on-large-clusters-16emdgg5h5.pdf

MapReduce: simplified data processing on large clusters

MapReduce is a programming model and an associated implementation for processing and generating large datasets that is amenable to a broad variety of real-world tasks. Users specify the computation in terms of a map and a reduce function, and the underlying runtime system automatically parallelizes the computation across large-scale clusters of machines, handles machine failures, and schedules inter-machine communication to make efficient use of the network and disks. Programmers find the system easy to use: more than ten thousand distinct MapReduce programs have been implemented internally at Google over the past four years, and an average of one hundred thousand MapReduce jobs are executed on Google's clusters every day, processing a total of more than twenty petabytes of data per day.

/pdf/mapreduce-simplified-data-processing-on-large-clusters-1zth35tzla.pdf

We present a new algorithm, called marching cubes, that creates triangle models of constant density surfaces from 3D medical data. Using a divide-and-conquer approach to generate inter-slice connectivity, we create a case table that defines triangle topology. The algorithm processes the 3D medical data in scan-line order and calculates triangle vertices using linear interpolation. We find the gradient of the original data, normalize it, and use it as a basis for shading the models. The detail in images produced from the generated surface models is the result of maintaining the inter-slice connectivity, surface data, and gradient information present in the original 3D data. Results from computed tomography (CT), magnetic resonance (MR), and single-photon emission computed tomography (SPECT) illustrate the quality and functionality of marching cubes. We also discuss improvements that decrease processing time and add solid modeling capabilities.

/pdf/marching-cubes-a-high-resolution-3d-surface-construction-3ealswkmtb.pdf

Marching cubes: A high resolution 3D surface construction algorithm

Preface * Introduction * The Bayes Error * Inequalities and alternatedistance measures * Linear discrimination * Nearest neighbor rules *Consistency * Slow rates of convergence Error estimation * The regularhistogram rule * Kernel rules Consistency of the k-nearest neighborrule * Vapnik-Chervonenkis theory * Combinatorial aspects of Vapnik-Chervonenkis theory * Lower bounds for empirical classifier selection* The maximum likelihood principle * Parametric classification *Generalized linear discrimination * Complexity regularization *Condensed and edited nearest neighbor rules * Tree classifiers * Data-dependent partitioning * Splitting the data * The resubstitutionestimate * Deleted estimates of the error probability * Automatickernel rules * Automatic nearest neighbor rules * Hypercubes anddiscrete spaces * Epsilon entropy and totally bounded sets * Uniformlaws of large numbers * Neural networks * Other error estimates *Feature extraction * Appendix * Notation * References * Index

/pdf/a-probabilistic-theory-of-pattern-recognition-1p19dg485x.pdf

A Probabilistic Theory of Pattern Recognition

Data mining applications place special requirements on clustering algorithms including: the ability to find clusters embedded in subspaces of high dimensional data, scalability, end-user comprehensibility of the results, non-presumption of any canonical data distribution, and insensitivity to the order of input records. We present CLIQUE, a clustering algorithm that satisfies each of these requirements. CLIQUE identifies dense clusters in subspaces of maximum dimensionality. It generates cluster descriptions in the form of DNF expressions that are minimized for ease of comprehension. It produces identical results irrespective of the order in which input records are presented and does not presume any specific mathematical form for data distribution. Through experiments, we show that CLIQUE efficiently finds accurate cluster in large high dimensional datasets.

/pdf/automatic-subspace-clustering-of-high-dimensional-data-for-541pkakmb5.pdf

Automatic subspace clustering of high dimensional data for data mining applications

In many scientific and technical endeavors, a three-dimensional solid must be reconstructed from serial sections, either to aid in the comprehension of the object's structure or to facilitate its automatic manipulation and analysis. This paper presents a general solution to the problem of constructing a surface over a set of cross-sectional contours. This surface, to be composed of triangular tiles, is constructed by separately determining an optimal surface between each pair of consecutive contours. Determining such a surface is reduced to the problem of finding certain minimum cost cycles in a directed toroidal graph. A new fast algorithm for finding such cycles is utilized. Also developed is a closed-form expression, in terms of the number of contour points, for an upper bound on the number of operations required to execute the algorithm. An illustrated example which involves the construction of a minimum area surface describing a human head is included.

Optimal surface reconstruction from planar contours

This paper describes a new algorithm for solving the hidden surface (or line) problem, to more rapidly generate realistic images of 3-D scenes composed of polygons, and presents the development of theoretical foundations in the area as well as additional related algorithms. As in many applications the environment to be displayed consists of polygons many of whose relative geometric relations are static, we attempt to capitalize on this by preprocessing the environment's database so as to decrease the run-time computations required to generate a scene. This preprocessing is based on generating a “binary space partitioning” tree whose in order traversal of visibility priority at run-time will produce a linear order, dependent upon the viewing position, on (parts of) the polygons, which can then be used to easily solve the hidden surface problem. In the application where the entire environment is static with only the viewing-position changing, as is common in simulation, the results presented will be sufficient to solve completely the hidden surface problem.

/pdf/on-visible-surface-generation-by-a-priori-tree-structures-4vfk1ok2bi.pdf

On visible surface generation by a priori tree structures

Discovering frequent itemsets is a key problem in important data mining applications, such as the discovery of association rules, strong rules, episodes, and minimal keys. Typical algorithms for solving this problem operate in a bottom-up breadth-first search direction. The computation starts from frequent 1-itemsets (minimal length frequent itemsets) and continues until all maximal (length) frequent itemsets are found. During the execution, every frequent itemset is explicitly considered. Such algorithms perform reasonably well when all maximal frequent itemsets are short. However, performance drastically decreases when some of the maximal frequent itemsets are relatively long. We present a new algorithm which combines both the bottom-up and top-down searches. The primary search direction is still bottom-up, but a restricted search is also conducted in the top-down direction. This search is used only for maintaining and updating a new data structure we designed, the maximum frequent candidate set. It is used to prune candidates in the bottom-up search. A very important characteristic of the algorithm is that it does not require explicite examination of every frequent itemset. Therefore the algorithm performs well even when some maximal frequent itemsets are long. As its output, the algorithm produces the maximum frequent set, i.e., the set containing all maximal frequent itemsets, which therefore specifies immediately all frequent itemsets. We evaluate the performance of the algorithm using a well-known benchmark database. The improvements can be up to several orders of magnitude, compared to the best current algorithms.

Pincer-search : A new algorithm for discovering the maximum frequent set

Discovering frequent itemsets is a key problem in important data mining applications, such as the discovery of association rules, strong rules, episodes, and minimal keys. Typical algorithms for solving this problem operate in a bottom-up breadth-first search direction. The computation starts from frequent 1-itemsets (minimal length frequent itemsets) and continues until all maximal (length) frequent itemsets are found. During the execution, every frequent itemset is explicitly considered. Such algorithms perform reasonably well when all maximal frequent itemsets are short. However, performance drastically decreases when some of the maximal frequent itemsets are relatively long. We present a new algorithm which combines both the bottom-up and top-down directions. The main search direction is still bottom-up but a restricted search is conducted in the top-down direction. This search is used only for maintaining and updating a new data structure we designed, the maximum frequent candidate set. It is used to prune candidates in the bottom-up search. As a very important characteristic of the algorithm, it is not necessary to explicitly examine every frequent itemset. Therefore it performs well even when some maximal frequent itemsets are long. As its output, the algorithm produces the maximum frequent set, i.e., the set containing all maximal frequent itemsets, which therefore specifies immediately all frequent itemsets. We evaluate the performance of the algorithm using a well-known benchmark database. The improvements can be up to several orders of magnitude, compared to the best current algorithms.

Zvi M. Kedem

Papers

Optimal surface reconstruction from planar contours

On visible surface generation by a priori tree structures

Pincer-search : A new algorithm for discovering the maximum frequent set

Pincer-Search: A New Algorithm for Discovering the Maximum Frequent Set

Charlotte: metacomputing on the Web