The Geodesic Distance of Data

Total Page:16

File Type:pdf, Size:1020Kb

The Geodesic Distance of Data The Geodesic Distance of Data Paul Beinat 1 Data is normally assumed to exist as a multidimensional cloud. Many of the methods that have been used to represent data, such as principal components analysis for example, make this implicit assumption. When we fit models to data we also make this implicit assumption, fitting the outcome to the overt predictor variables in the data. If data does have an inherent structure, rather than being just a cloud, then these assumptions are unsafe and questionable at least. Any inherent structure in the data will be in the form of a manifold. A manifold is a multidimensional surface on which the data lies. Its dimensionality is less than that of the data. This makes it hard to detect, and that’s why we make assumptions about data being a cloud. A new algorithm has been developed that follows and measures the geodesic of the manifold of data. Tests are detailed to determine whether this algorithm finds false positives, a geodesic where none exists, and whether it can find a geodesic when artificially prepared data is analysed. Finally, we use the algorithm on a large real world data set. Introduction The task of analytics and data science in general, is to develop models from data. One of the key tasks involved is to reduce the dimensionality of the data to the some form of key variables for the modelling. One common way of achieving this is by reducing the number of variables based on their correlation to the desired outcome. Of course these correlations take many values, some Fig. 3. The “Swiss roll” data by geodesic distance along The two-dimensional embedding set, illustrating how Isomap the low-dimensional manifold recovered by isomap in step predictor variables are better than others, and variables exploits geodesic paths for (length of solid curve). (B) three, which best preserves the nonlinear dimensionality The neighborhood graph G shortest path distances in the can also be correlated with each other. Another often reduction (A) for two arbitary constructed in step one of neighborhood graph (overlaid). used method is to reduce the data dimensionality by points (circled) on a nonlinear Isomap (with K = 7 and N = Straight lines in the embedding manifold, their Euclidean distance 1000 data points) allows an (blue) now represents simpler using principal components analysis (PCA). PCA works in the high dimensional input approximation (red segments) and cleaner approximations to space (length of dashed line) to the true geodesic path to be the true geodesic paths than by finding the direction of the largest variance in the data may not accurately reflect their computed efficiently in step two, do the corresponding graph and fitting a line in the direction. It then finds the largest intrinsic similarity, as measured as the shortests paths in G. (C) paths (red). variance perpendicular to the first line. This process can In the case above the linear distance in the manifold be repeated many times, until the variance is exhausted. equates to the geodesic distance in the data. Manifolds Significantly PCA creates new data dimensions, which are in higher dimensions, the one above is 2-dimensional, linear combination of the overt data dimensions. cannot be reduced to two dimensions without distortion. In the age of big data, dimensionality reduction is a Therefore linear distances in higher order manifolds do not key objective. equate to geodesics. Another key approach has been nonlinear dimensionality Significantly, while the data above has 3 dimensions, it reduction. Most of the algorithms rely on inter data point actually lies on a two dimensional surface embedded in distances as the basis for structure discovery. These inter the 3 dimensional space. In more dimensions this is not point distances are represented in a matrix, which is then overtly visualisable. used as the basis of these algorithms. Inter point distances If there are N points in the data set, then the distances are calculated as Euclidean distance. One of the most matrix is N by N, and it has N2 entries. (Actually distances notable algorithms is the ISOMAP algorithm [1]. Another are not directed so we only need the triangular matrix.) If N is Local Linear Embedding [2]. Many more algorithms are is 10,000 then there are 100,000,000 entries. This is one detailed in [3]. Below are example results from ISOMAP. of the key limitations of these distance matrix approaches, and why they have not been applied to large data sets. There are new algorithms produced from time to time. A modern algorithm is [4]. 1 Finity and University of Technology Sydney finity.com.au Geodesic Distance of Data Algorithm The Geodesic Distance of Data Algorithm we In a simplistic implementation of the steps in “2.” it can implemented and will examine calculates the geodesic be readily seen that in order to find the point nearest distance between any two points, embedded in d the current point and closer to the goal point involves dimensions. It uses local Euclidean distances as the basis searching through virtually all points. If the number of of following the geodesic of the manifold. points is large then this search can be prohibitive. This is the search problem discussed below. If the data had d dimensions then the manifold has m dimensions where m < d. It can be seen that “3a” to “3d” represents a recursive algorithm, although it can be implemented as an iterative The algorithm has the following structure: one. The series of points are locally smoothed. The smoothing algorithm, a local linear regression, is a type 1. Start with 2 specified points from the d dimensional of kernel smoother. These types of kernel algorithms are data. Select one as the start point A and the other as well known in the literature. These smoothed points are the goal B. Add A to a list of points to remember, in L. then sent to the same smoothing algorithm. It stops when 2. Find the closest point to A that reduces the distance the new geodesic distance is not significantly smaller to B. “Reducing the distance” means that the than the previous. One complication not discussed is distance between the new point and the end point B what the smoothing does at the points closest to each is smaller than the distance of the original point A and end of the sequence, but this is also well described in the the end point B. Mark this new point as C. Add C to L. literature. a. Find the closest point to C that reduces the distance to B. Mark that point as a new C and add Implementation it to L. b. Repeat until there are no more points that are The search problem has severe implications. As the data close to C and reduce the distance to B. becomes more numerous the number of points on the path between two given points increases. The search c. Add B to L. for the nearest point also increases with the number of data points. This means that the search time increases 3. L now contains all the points from A to B, including approximately as the square of number of data points. them. Now smooth the path and subsequently the distance measure from A to B. In order to alleviate the search problem a Finity proprietary artificial neural network is used that models a. Select a subset of contiguous points from L. the joint probability distribution of data. Points close to Fit a linear regression to these points. Use the each other in the joint probability distribution are also regression to derive a new location for the central closest to each other in the data space itself. This allows point. the algorithm to exploit the neural network to limit the search problem to only searching a local region of the b. Move the points selection along by 1, selecting whole data. It can then connect from one local region the same number of points. Repeat the linear to another local region which is on the path to the goal regression and central point estimation. point. This makes the search problem approximately c. Repeat b. until it reaches the end of L. Calculate linear with the number of data points. It allows the the distance as the sum of the distances between algorithm to use large data sets. subsequent points. This is the current geodesic In all the experiments described below the recursive distance. local linear kernel smoother is set to terminate when d. Repeat a. to c. using the new points as L. it can no longer reduce the summed distance, using Calculate the new sum of the distances. If the the current smoothed points, by 1%, over the previous new distance is not smaller than the previous smoothed distance. It takes very few iterations to reach distance by some small margin then repeat d. the termination condition. 4. The points in L are now the smooth locally derived distance between A and B. It is the estimate of the geodesic distance. 2 of 4 Examples The first test of the algorithm is whether it can correctly The results were unsurprising. On many occasions it determine geodesic distances for a known manifold. was impossible to smooth the path because not enough points exist on path from the start to the end point. A 2 dimensional circular manifold is generated using This is expected given the random nature of the data and random points from x2 + y2 = 1. 10,000 points are the dimensionality – it’s a sparse data set. On the few generated using a uniform random distribution for x, occasions when enough points could be found on the using this function to derive y.
Recommended publications
  • Glossary Physics (I-Introduction)
    1 Glossary Physics (I-introduction) - Efficiency: The percent of the work put into a machine that is converted into useful work output; = work done / energy used [-]. = eta In machines: The work output of any machine cannot exceed the work input (<=100%); in an ideal machine, where no energy is transformed into heat: work(input) = work(output), =100%. Energy: The property of a system that enables it to do work. Conservation o. E.: Energy cannot be created or destroyed; it may be transformed from one form into another, but the total amount of energy never changes. Equilibrium: The state of an object when not acted upon by a net force or net torque; an object in equilibrium may be at rest or moving at uniform velocity - not accelerating. Mechanical E.: The state of an object or system of objects for which any impressed forces cancels to zero and no acceleration occurs. Dynamic E.: Object is moving without experiencing acceleration. Static E.: Object is at rest.F Force: The influence that can cause an object to be accelerated or retarded; is always in the direction of the net force, hence a vector quantity; the four elementary forces are: Electromagnetic F.: Is an attraction or repulsion G, gravit. const.6.672E-11[Nm2/kg2] between electric charges: d, distance [m] 2 2 2 2 F = 1/(40) (q1q2/d ) [(CC/m )(Nm /C )] = [N] m,M, mass [kg] Gravitational F.: Is a mutual attraction between all masses: q, charge [As] [C] 2 2 2 2 F = GmM/d [Nm /kg kg 1/m ] = [N] 0, dielectric constant Strong F.: (nuclear force) Acts within the nuclei of atoms: 8.854E-12 [C2/Nm2] [F/m] 2 2 2 2 2 F = 1/(40) (e /d ) [(CC/m )(Nm /C )] = [N] , 3.14 [-] Weak F.: Manifests itself in special reactions among elementary e, 1.60210 E-19 [As] [C] particles, such as the reaction that occur in radioactive decay.
    [Show full text]
  • Forces Different Types of Forces
    Forces and motion are a part of your everyday life for example pushing a trolley, a horse pulling a rope, speed and acceleration. Force and motion causes objects to move but also to stay still. Motion is simply a movement but needs a force to move. There are 2 types of forces, contact forces and act at a distance force. Forces Every day you are using forces. Force is basically push and pull. When you push and pull you are applying a force to an object. If you are Appling force to an object you are changing the objects motion. For an example when a ball is coming your way and then you push it away. The motion of the ball is changed because you applied a force. Different Types of Forces There are more forces than push or pull. Scientists group all these forces into two groups. The first group is contact forces, contact forces are forces when 2 objects are physically interacting with each other by touching. The second group is act at a distance force, act at a distance force is when 2 objects that are interacting with each other but not physically touching. Contact Forces There are different types of contact forces like normal Force, spring force, applied force and tension force. Normal force is when nothing is happening like a book lying on a table because gravity is pulling it down. Another contact force is spring force, spring force is created by a compressed or stretched spring that could push or pull. Applied force is when someone is applying a force to an object, for example a horse pulling a rope or a boy throwing a snow ball.
    [Show full text]
  • Geodesic Distance Descriptors
    Geodesic Distance Descriptors Gil Shamai and Ron Kimmel Technion - Israel Institute of Technologies [email protected] [email protected] Abstract efficiency of state of the art shape matching procedures. The Gromov-Hausdorff (GH) distance is traditionally used for measuring distances between metric spaces. It 1. Introduction was adapted for non-rigid shape comparison and match- One line of thought in shape analysis considers an ob- ing of isometric surfaces, and is defined as the minimal ject as a metric space, and object matching, classification, distortion of embedding one surface into the other, while and comparison as the operation of measuring the discrep- the optimal correspondence can be described as the map ancies and similarities between such metric spaces, see, for that minimizes this distortion. Solving such a minimiza- example, [13, 33, 27, 23, 8, 3, 24]. tion is a hard combinatorial problem that requires pre- Although theoretically appealing, the computation of computation and storing of all pairwise geodesic distances distances between metric spaces poses complexity chal- for the matched surfaces. A popular way for compact repre- lenges as far as direct computation and memory require- sentation of functions on surfaces is by projecting them into ments are involved. As a remedy, alternative representa- the leading eigenfunctions of the Laplace-Beltrami Opera- tion spaces were proposed [26, 22, 15, 10, 31, 30, 19, 20]. tor (LBO). When truncated, the basis of the LBO is known The question of which representation to use in order to best to be the optimal for representing functions with bounded represent the metric space that define each form we deal gradient in a min-max sense.
    [Show full text]
  • Robust Topological Inference: Distance to a Measure and Kernel Distance
    Journal of Machine Learning Research 18 (2018) 1-40 Submitted 9/15; Revised 2/17; Published 4/18 Robust Topological Inference: Distance To a Measure and Kernel Distance Fr´ed´ericChazal [email protected] Inria Saclay - Ile-de-France Alan Turing Bldg, Office 2043 1 rue Honor´ed'Estienne d'Orves 91120 Palaiseau, FRANCE Brittany Fasy [email protected] Computer Science Department Montana State University 357 EPS Building Montana State University Bozeman, MT 59717 Fabrizio Lecci [email protected] New York, NY Bertrand Michel [email protected] Ecole Centrale de Nantes Laboratoire de math´ematiquesJean Leray 1 Rue de La Noe 44300 Nantes FRANCE Alessandro Rinaldo [email protected] Department of Statistics Carnegie Mellon University Pittsburgh, PA 15213 Larry Wasserman [email protected] Department of Statistics Carnegie Mellon University Pittsburgh, PA 15213 Editor: Mikhail Belkin Abstract Let P be a distribution with support S. The salient features of S can be quantified with persistent homology, which summarizes topological features of the sublevel sets of the dis- tance function (the distance of any point x to S). Given a sample from P we can infer the persistent homology using an empirical version of the distance function. However, the empirical distance function is highly non-robust to noise and outliers. Even one outlier is deadly. The distance-to-a-measure (DTM), introduced by Chazal et al. (2011), and the ker- nel distance, introduced by Phillips et al. (2014), are smooth functions that provide useful topological information but are robust to noise and outliers. Chazal et al. (2015) derived concentration bounds for DTM.
    [Show full text]
  • Network Science
    This is a preprint of Katy Börner, Soma Sanyal and Alessandro Vespignani (2007) Network Science. In Blaise Cronin (Ed) Annual Review of Information Science & Technology, Volume 41. Medford, NJ: Information Today, Inc./American Society for Information Science and Technology, chapter 12, pp. 537-607. Network Science Katy Börner School of Library and Information Science, Indiana University, Bloomington, IN 47405, USA [email protected] Soma Sanyal School of Library and Information Science, Indiana University, Bloomington, IN 47405, USA [email protected] Alessandro Vespignani School of Informatics, Indiana University, Bloomington, IN 47406, USA [email protected] 1. Introduction.............................................................................................................................................2 2. Notions and Notations.............................................................................................................................4 2.1 Graphs and Subgraphs .........................................................................................................................5 2.2 Graph Connectivity..............................................................................................................................7 3. Network Sampling ..................................................................................................................................9 4. Network Measurements........................................................................................................................11
    [Show full text]
  • Latent Distance Estimation for Random Geometric Graphs ∗
    Latent Distance Estimation for Random Geometric Graphs ∗ Ernesto Araya Valdivia Laboratoire de Math´ematiquesd'Orsay (LMO) Universit´eParis-Sud 91405 Orsay Cedex France Yohann De Castro Ecole des Ponts ParisTech-CERMICS 6 et 8 avenue Blaise Pascal, Cit´eDescartes Champs sur Marne, 77455 Marne la Vall´ee,Cedex 2 France Abstract: Random geometric graphs are a popular choice for a latent points generative model for networks. Their definition is based on a sample of n points X1;X2; ··· ;Xn on d−1 the Euclidean sphere S which represents the latent positions of nodes of the network. The connection probabilities between the nodes are determined by an unknown function (referred to as the \link" function) evaluated at the distance between the latent points. We introduce a spectral estimator of the pairwise distance between latent points and we prove that its rate of convergence is the same as the nonparametric estimation of a d−1 function on S , up to a logarithmic factor. In addition, we provide an efficient spectral algorithm to compute this estimator without any knowledge on the nonparametric link function. As a byproduct, our method can also consistently estimate the dimension d of the latent space. MSC 2010 subject classifications: Primary 68Q32; secondary 60F99, 68T01. Keywords and phrases: Graphon model, Random Geometric Graph, Latent distances estimation, Latent position graph, Spectral methods. 1. Introduction Random geometric graph (RGG) models have received attention lately as alternative to some simpler yet unrealistic models as the ubiquitous Erd¨os-R´enyi model [11]. They are generative latent point models for graphs, where it is assumed that each node has associated a latent d point in a metric space (usually the Euclidean unit sphere or the unit cube in R ) and the connection probability between two nodes depends on the position of their associated latent points.
    [Show full text]
  • Average D-Distance Between Vertices of a Graph
    italian journal of pure and applied mathematics { n. 33¡2014 (293¡298) 293 AVERAGE D-DISTANCE BETWEEN VERTICES OF A GRAPH D. Reddy Babu Department of Mathematics Koneru Lakshmaiah Education Foundation (K.L. University) Vaddeswaram Guntur 522 502 India e-mail: [email protected], [email protected] P.L.N. Varma Department of Science & Humanities V.F.S.T.R. University Vadlamudi Guntur 522 237 India e-mail: varma [email protected] Abstract. The D-distance between vertices of a graph G is obtained by considering the path lengths and as well as the degrees of vertices present on the path. The average D-distance between the vertices of a connected graph is the average of the D-distances between all pairs of vertices of the graph. In this article we study the average D-distance between the vertices of a graph. Keywords: D-distance, average D-distance, diameter. 2000 Mathematics Subject Classi¯cation: 05C12. 1. Introduction The concept of distance is one of the important concepts in study of graphs. It is used in isomorphism testing, graph operations, hamiltonicity problems, extremal problems on connectivity and diameter, convexity in graphs etc. Distance is the basis of many concepts of symmetry in graphs. In addition to the usual distance, d(u; v) between two vertices u; v 2 V (G) we have detour distance (introduced by Chartrand et al, see [2]), superior distance (introduced by Kathiresan and Marimuthu, see [6]), signal distance (introduced by Kathiresan and Sumathi, see [7]), degree distance etc. 294 d. reddy babu, p.l.n. varma In an earlier article [9], the authors introduced the concept of D-distance be- tween vertices of a graph G by considering not only path length between vertices, but also the degrees of all vertices present in a path while de¯ning the D-distance.
    [Show full text]
  • Hydraulics Manual Glossary G - 3
    Glossary G - 1 GLOSSARY OF HIGHWAY-RELATED DRAINAGE TERMS (Reprinted from the 1999 edition of the American Association of State Highway and Transportation Officials Model Drainage Manual) G.1 Introduction This Glossary is divided into three parts: · Introduction, · Glossary, and · References. It is not intended that all the terms in this Glossary be rigorously accurate or complete. Realistically, this is impossible. Depending on the circumstance, a particular term may have several meanings; this can never change. The primary purpose of this Glossary is to define the terms found in the Highway Drainage Guidelines and Model Drainage Manual in a manner that makes them easier to interpret and understand. A lesser purpose is to provide a compendium of terms that will be useful for both the novice as well as the more experienced hydraulics engineer. This Glossary may also help those who are unfamiliar with highway drainage design to become more understanding and appreciative of this complex science as well as facilitate communication between the highway hydraulics engineer and others. Where readily available, the source of a definition has been referenced. For clarity or format purposes, cited definitions may have some additional verbiage contained in double brackets [ ]. Conversely, three “dots” (...) are used to indicate where some parts of a cited definition were eliminated. Also, as might be expected, different sources were found to use different hyphenation and terminology practices for the same words. Insignificant changes in this regard were made to some cited references and elsewhere to gain uniformity for the terms contained in this Glossary: as an example, “groundwater” vice “ground-water” or “ground water,” and “cross section area” vice “cross-sectional area.” Cited definitions were taken primarily from two sources: W.B.
    [Show full text]
  • Mathematics of Distances and Applications
    Michel Deza, Michel Petitjean, Krassimir Markov (eds.) Mathematics of Distances and Applications I T H E A SOFIA 2012 Michel Deza, Michel Petitjean, Krassimir Markov (eds.) Mathematics of Distances and Applications ITHEA® ISBN 978-954-16-0063-4 (printed); ISBN 978-954-16-0064-1 (online) ITHEA IBS ISC No.: 25 Sofia, Bulgaria, 2012 First edition Printed in Bulgaria Recommended for publication by The Scientific Council of the Institute of Information Theories and Applications FOI ITHEA The papers accepted and presented at the International Conference “Mathematics of Distances and Applications”, July 02-05, 2012, Varna, Bulgaria, are published in this book. The concept of distance is basic to human experience. In everyday life it usually means some degree of closeness of two physical objects or ideas, i.e., length, time interval, gap, rank difference, coolness or remoteness, while the term metric is often used as a standard for a measurement. Except for the last two topics, the mathematical meaning of those terms, which is an abstraction of measurement, is considered. Distance metrics and distances have now become an essential tool in many areas of Mathematics and its applications including Geometry, Probability, Statistics, Coding/Graph Theory, Clustering, Data Analysis, Pattern Recognition, Networks, Engineering, Computer Graphics/Vision, Astronomy, Cosmology, Molecular Biology, and many other areas of science. Devising the most suitable distance metrics and similarities, to quantify the proximity between objects, has become a standard task for many researchers. Especially intense ongoing search for such distances occurs, for example, in Computational Biology, Image Analysis, Speech Recognition, and Information Retrieval. For experts in the field of mathematics, information technologies as well as for practical users.
    [Show full text]
  • Topology Distance: a Topology-Based Approach for Evaluating Generative Adversarial Networks
    Topology Distance: A Topology-Based Approach For Evaluating Generative Adversarial Networks Danijela Horak 1 Simiao Yu 1 Gholamreza Salimi-Khorshidi 1 Abstract resemble real images Xr, while D discriminates between Xg Automatic evaluation of the goodness of Gener- and Xr. G and D are trained by playing a two-player mini- ative Adversarial Networks (GANs) has been a max game in a competing manner. Such novel adversarial challenge for the field of machine learning. In training process is a key factor in GANs’ success: It im- this work, we propose a distance complementary plicitly defines a learnable objective that is flexibly adaptive to existing measures: Topology Distance (TD), to various complicated image-generation tasks, in which it the main idea behind which is to compare the would be difficult or impossible to explicitly define such an geometric and topological features of the latent objective. manifold of real data with those of generated data. One of the biggest challenges in the field of generative More specifically, we build Vietoris-Rips com- models – including for GANs – is the automatic evaluation plex on image features, and define TD based on of the goodness of such models (e.g., whether or not the the differences in persistent-homology groups of data generated by such models are similar to the data they the two manifolds. We compare TD with the were trained on). Unlike supervised learning, where the most commonly-used and relevant measures in goodness of the models can be assessed by comparing their the field, including Inception Score (IS), Frechet´ predictions with the actual labels, or in some other deep- Inception Distance (FID), Kernel Inception Dis- learning models where the goodness of the model can be tance (KID) and Geometry Score (GS), in a range assessed using the likelihood of the validation data under the of experiments on various datasets.
    [Show full text]
  • "Distance Measures for Graph Theory"
    Distance measures for graph theory : Comparisons and analyzes of different methods Dissertation presented by Maxime DUYCK for obtaining the Master’s degree in Mathematical Engineering Supervisor(s) Marco SAERENS Reader(s) Guillaume GUEX, Bertrand LEBICHOT Academic year 2016-2017 Acknowledgments First, I would like to thank my supervisor Pr. Marco Saerens for his presence, his advice and his precious help throughout the realization of this thesis. Second, I would also like to thank Bertrand Lebichot and Guillaume Guex for agreeing to read this work. Next, I would like to thank my parents, all my family and my friends to have accompanied and encouraged me during all my studies. Finally, I would thank Malian De Ron for creating this template [65] and making it available to me. This helped me a lot during “le jour et la nuit”. Contents 1. Introduction 1 1.1. Context presentation .................................. 1 1.2. Contents .......................................... 2 2. Theoretical part 3 2.1. Preliminaries ....................................... 4 2.1.1. Networks and graphs .............................. 4 2.1.2. Useful matrices and tools ........................... 4 2.2. Distances and kernels on a graph ........................... 7 2.2.1. Notion of (dis)similarity measures ...................... 7 2.2.2. Kernel on a graph ................................ 8 2.2.3. The shortest-path distance .......................... 9 2.3. Kernels from distances ................................. 9 2.3.1. Multidimensional scaling ............................ 9 2.3.2. Gaussian mapping ............................... 9 2.4. Similarity measures between nodes .......................... 9 2.4.1. Katz index and its Leicht’s extension .................... 10 2.4.2. Commute-time distance and Euclidean commute-time distance .... 10 2.4.3. SimRank similarity measure .........................
    [Show full text]
  • Lesson 3 – Understanding Distance in Space (Optional)
    Lesson 3 – Understanding Distance in Space (optional) Background The distance between objects in space is vast and very difficult for most children to grasp. The values for these distances are cumbersome for astronomers and scientists to manipulate. Therefore, scientists use a unit of measurement called an astronomical unit. An astronomical unit is the average distance between the Earth and the Sun – 150 million kilometers. Using a scale model to portray the distances of the planets will help the students begin to understand the vastness of space. Astronomers soon found that even the astronomical unit was not large enough to measure the distance of objects outside of our solar system. For these distances, the parsec is used. A parsec is a unit of light-years. A parsec equals approximately 3.3 light-years. (Keep in mind that a light-year is the distance light travels in one year – 9.5 trillion km – and that light travels 300,000 km per second!) Teacher Notes and Hints Prior to lesson • Discuss and review measuring length or distance. How far is one meter? • Discuss the use of prefixes that add or subtract to “basic” units of length. Example: meter is the basic metric unit for length “kilo” means 1000 – kilometer is 1000 meters “centi” means 1/100 – centimeter is 1/100 of a meter Adapt the discussion for grade level. For example, fifth graders can learn the metric prefixes for more “lengths” and may add to decimal work in math class: Kilo – 1000 Hecto – 100 Deka – 10 Deci – 1/10 (0.1) Centi – 1/100 (0.01) Milli – 1/1000 (0.001) • When discussing kilometers, it is helpful to know the number of km’s to local places (for example, kilometers to the Mall).
    [Show full text]