Mobility Models. Larissa Marinho Eglem de Oliveira. May 26th CMPE 257 Wireless Networks. (UCSC) May / 50

Size: px
Start display at page:

Download "Mobility Models. Larissa Marinho Eglem de Oliveira. May 26th CMPE 257 Wireless Networks. (UCSC) May / 50"

Transcription

1 Mobility Models Larissa Marinho Eglem de Oliveira CMPE 257 Wireless Networks May 26th 2015 (UCSC) May / 50

2 1 Motivation 2 Mobility Models 3 Extracting a Mobility Model from Real User Traces 4 Self-similar Least Action Walk (SLAW) (UCSC) May / 50

3 Contents 1 Motivation 2 Mobility Models 3 Extracting a Mobility Model from Real User Traces 4 Self-similar Least Action Walk (SLAW) (UCSC) May / 50

4 Motivation Why Mobility Models? Wireless nodes are carried by humans and hence inherit human features in what concerns movement. Wireless Community Networks: park, home, city, campus; Hotspots: areas covered by one or more APs; Roaming: users tend to move from one hotspot to another; Predict Movement Patterns: handover optimization; network planning and dimensioning; location updating and paging; path/route computation. (UCSC) May / 50

5 Motivation Why Traces? Traces are essential to understand node mobility behavior under different conditions and hence are key to benchmark analytical models for movement of nodes, i.e, mobility models. Most mobility models are analised from traces based on Bayesian Models; Best mobility models should be able to capture: social behavior; time and space correlation. (UCSC) May / 50

6 Classification Individual Mobility Models: attempts to mimic the mobility pattern behavior of a specific node and from a temporal timeframe; Group Mobility Models: takes into account the need to access a node s perspective towards its neighbors; Synthetic analytical models: Brownian Motion, Random Walk, Levy Patterns; Trace-based: WLAN and GPS traces. (UCSC) May / 50

7 General Characteristics Position; Speed; Direction; Accelaration; Pause Time; Inter-contact time. (UCSC) May / 50

8 Contents 1 Motivation 2 Mobility Models 3 Extracting a Mobility Model from Real User Traces 4 Self-similar Least Action Walk (SLAW) (UCSC) May / 50

9 Individual Models - Synthetic Random Walk: Chooses direction and speed randomly. No pause time; Probabilistic Random Walk: Uses a probability transition matrix to determine a node s next position. Speed is variable. Node stays in a position according to transition matrix; Random Waypoint (RWP): Extension of random walk, however adds pause time between changes of direction and speed. Pause time, direction and speed are randomly chosen from a Uniform distribution; (UCSC) May / 50

10 Individual Models - Trace-based Model-T: based on traces of user s AP registration patterns on a campus. User registration is asymmetric in space distributions, which are hierarchical; A Simple Human Mobility Model: first attempt on trying to identify social behavior; based on traces of a campus indoor. Mobile nodes travel only in straight lines with pre-configured speed. Groups of mobiles nodes are divided according to their common interest. (UCSC) May / 50

11 Summary Most of individual models based on the notion that movement is random; Mobility traces show that nodes on wireless networks do not exhibit random behavior; (UCSC) May / 50

12 Group Models - Synthetic Reference Point Group Model (RPGM): Every group has a Reference Point (RP) and each individual node has a personal space where it behaves randomly. The movement of the RP characterizes the group movement: uses RWP for RP and nodes mobility; military applications; RPs are usually mobile nodes, but could be considered as APs. Nomadic Mobility Model: captures mobility of nomadic groups that move from time to time. Based on RPGM; military operations, as well as robotics applied to agriculture. (UCSC) May / 50

13 Group Models - Hybrid Community Based Mobility Model (CMM): nodes are grouped into communities and move towards another nodes or communities according to the strengh of social associations between nodes: Social associations defined based on likelihood of geographic co-location; People that share strong social ties tend to move to similar places; Communities are defined by using a clustering algorithm; Next direction chosen proportional to the attractiveness of each community Home Cell Community Mobility Model (HCMM): based on CMM, but considers not only attraction torwards other nodes, but also attraction torwards locations. A second work furthers the model taking into account traces. Self-similar Least Action Walk (SLAW): uses fractal waypoints behavior to model attraction by popular places. (UCSC) May / 50

14 Summary Trend: evolution towards social mobility models; Node movement in wireless networks is not random, instead, it exhibits specific characteristics; CMM takes into account social behavior, however does not address pause time, or accelaration. Target of movement is always a node in a cluster, never a location; HCMM adds attraction torwards a location, but does no model pause time; SLAW captures essential properties of human behavior, such as pause time, attraction to popular places and preference to walk around their own confined area. (UCSC) May / 50

15 Design Goals Time and Spatial correlation: capable of determining nodes movement with a relation in time and space; Node pattern: type of movement a node has in its individual space; Obstacle modeling: capable of capturing collisions between nodes; Social behavior: attraction areas, communities, hotspots. (UCSC) May / 50

16 Applicability Handover optimization; Resource management; Network optimization: routing, context-aware applications, opportunistic networks; Simulation. (UCSC) May / 50

17 Contents 1 Motivation 2 Mobility Models 3 Extracting a Mobility Model from Real User Traces 4 Self-similar Least Action Walk (SLAW) (UCSC) May / 50

18 Collecting User Traces Traces extracted from always-on WiFi devices from a 13-month data set (VoIP traces); Direction of movement closely reflects the direction of roads and walkways; Focus on movement among hotspots; Speed and pause time follow a log-normal distribution; Synthetic traces derived from the mobility model match real traces with a median relative error of 17%; Syslog traces parsed to obtain mobility traces: list locations of APs with which each device associated; for each user, the events were extracted for each day; working days from 8am to 6pm; workday traces divided into mobile and stationary; if the distance between any two APs that a user visits in a workday is greater or equal to 100m, the trace is considered mobile if user off for more than 30min: multiple walks mobile traces generate 3838 walks and 3876 stationary traces generates 4006 walks; (UCSC) May / 50

19 Processing Trace Algorithm From the mobility traces, extract users location or user tracks; Estimate pause duration; Users generally do not stay close to an AP; Mobile devices do not necessarily associate with the closest AP; Different mobile devices have different policies to changing associations (UCSC) May / 50

20 Estimating User Location Triangle Centroid: uses location of past three AP associations as the vertice of the triangle; location estimate at every association message Time-based centroid: every p seconds, update the user s location with associations that happened during the past q seconds. n is the number of associations within the past p seconds. (UCSC) May / 50

21 Kalman Filter: is a recursive data processing algorithm that produces optimal estimates. Infers parametes of interest from indirect, innacurate and uncertain observations. If all noise is Gaussian, the Kalman filter minimizes the mean square error of the estimated parameters: where x k is the system model, z k is the measumerent (UCSC) May / 50

22 Validation 4 walks with GPS tracker, Voicera VoIP communicators and Cisco VoIP phones; Kalman Parameters: covariance matrix for wk and v k : assume that the variances are independent of each other; values of w k set to 1; assume movement in x and y directions are likely symmetric, therefore errors in directions x and y are the same, so only one variable for v 2. v 2 is empirically chosen as 25; (UCSC) May / 50

23 (UCSC) May / 50

24 Different types of devices have different association patterns. Vocera communicators agressively associated with many APs, whereas CISCO phones tended to stay with a single AP for a longer time. Kalman Filter offers the best common fit for both devices (UCSC) May / 50

25 (UCSC) May / 50

26 Extracting Pause Time Time between two associations divided between travel time and pause time; Pause duration is estimated based on users speed; Assuming users are pedestrians, if s i = [min, 10ms], user did not pause at l i ; where s i is the user s speed, d i is the distance from location l i+1 l i and e i = t i+1 t i (UCSC) May / 50

27 Shorter e i leads to higher speeds, however shorter e i are due to aggressive associations in search for better signal reception. Speed after pause time: (UCSC) May / 50

28 Extracting Hotspots Regions First approach: divide the area into fixed-regions or grid. The question is how to choose the size of the grid. Too small may overlook the hotspots, making it too big may create an oversized hotspot; Chosen approach: a 2-D Gaussian distribution to each pause location, weighted by duration of pause, and then sum up the distribution. At each pause location, the 2-D Gaussian distribution creates a small mountain, uniformly distributed about its vertical axis: to select the appropriate Gaussian distribution, the standard deviation should reflect the confidence of the user s location and the agressiveness in aggregation pauses. σ = 20m. (UCSC) May / 50

29 Mobility Characteristics Mobile Walks: Kalman Filter (pause time estimator) pause time; speed; direction; start and end time. Stationary Walks: Triangle Centroid (location estimator) duration of stay; start and end time. (UCSC) May / 50

30 Mobile Set Pause time and Speed: using MLE estimation, clustered pauses fit a log-normal distribution, so does speed. (UCSC) May / 50

31 Direction of Movement: manual comparison with a map of Darthmouth, shows that direction of larger peaks corresponds to direction of popular roads; symmetry around 180 degrees; people in a road move in both directions. Start/End Times of a workday: exponential distributions; Hotspots: gym, restaurant, engineering building, CS department, office of network administration and library. (UCSC) May / 50

32 Stationary Set Duration of a walk given by the difference between the time when the first and last messages of that walk were recorded. If only one message, duration is zero; (UCSC) May / 50

33 User s location computed using the triangle centroid. Gaussian distribution for each stationary location * gym and restaurant not popular Start/End Times follow a Uniform distribution. (UCSC) May / 50

34 Characterizing Hotspots hotspots defined using a Gaussian distribution: the area outside hotspots become the cold region; threshold defined to determine individual hotspots: (UCSC) May / 50

35 Hotspots and its pause time distributions: Movement between regions: nxn transition matrix where n 2 is the number of hotspots. The remaining two columns are the cold region and the off state; cold regions are treated as waypoints (l i ). the waypoint matrix is a (n 1)x(n 1) matrix that consists of the average number of waypoints when users moved between two regions. (UCSC) May / 50

36 Modeling Mobility Stationary set: a stationary user enters the network at a time from the start time distribution at the location from the initial location map. It then stays at the location for the duration chosen from the duration-of-stay distribution; Mobile set: a mobile user enters the network at a time selected from the start time distribution at a region selected from the initial region distribution. The user s next destination is then chosen based on the probabilities in the region transition matrix. The number of waypoints visited on the way to the destination is based on the waypoint matrix; Gaussian distribution with the mean based on this matrix to choose the number of waypoints, k, for each move. The speed of movement is chosen from the overall speed distribution. Once a user has reached the destination, it pauses for a period chosen from the pause time distribution for that particular region. When the pause time elapses, the next destination is chosen using the region transition matrix. (UCSC) May / 50

37 Validation Number of visitors within a given region in each hour of a workday (UCSC) May / 50

38 Distribution matches for most of the hotspots, but the first; Model does not capture the large peek during lunch time; Proposed model only takes into account variations that occur within the start and end times, but does not consider variations throughout the day; To consider variations within the day, it would be necessary to have previous knowledge of hotspots. (UCSC) May / 50

39 Contents 1 Motivation 2 Mobility Models 3 Extracting a Mobility Model from Real User Traces 4 Self-similar Least Action Walk (SLAW) (UCSC) May / 50

40 Concepts Least Action Principal: all objects (nodes) move in the direction of minimizing their discomfort, in this case, traveled distance; Self-similarity Property: the features of a given set of points hold independently of the spatial or time scale of the observation. Human destinations tend to form hotspots in the region of movement, and hotspots can clearly be observed independent of the spatial granularity of the observation. (UCSC) May / 50

41 Properties F1 - Truncated power-law flights and pause times: humans flights are straight line trips without change of direction or pause; F2 - Heterogeneously bounded mobility areas: people move mostly within their confined areas of mobility and different people may have widely different mobility areas; F3 - Truncated power-law inter-contact times (ICTs): time elapsed between two sucessive contacts of the same two persons; F4 - Fractal waypoints: people are always more attracted to more popular places. fractal: a natural phenomenon or mathematical set that exhibis a repeating pattern that displays at every scale. (UCSC) May / 50

42 SLAW produces Synthetic traces containing all the previous properties; Uses GPS traces of human walks from students in a University campus and turists in a park; Simple as RWP: only need as parameters the size of the walk area, number of walkers and the Hurst value to generate fractal waypoints; Does not depend on real traces. (UCSC) May / 50

43 Overview 1 Generates fractal waypoints (F4) using a Brownian Motion Generator over a 2-D image; 2 Generates power-law flights (F1) on top of fractal waypoints: proves analytically that fractal points induce power-law gaps (inter-spacing among neighboring fractal points) 3 Least Action Trip Planning Algorithm (LATP): people plan their trips over known destinations in a gap-preserving manner, where they visit nearby destinations first before visiting farther destinations. LATP satisfies (F4) and (F1) 4 Individual Walker Model satisfies (F2) and (F3) (UCSC) May / 50

44 (UCSC) May / 50

45 LAPT Next waypoint is selected based on a weight function of 1 d α, where d is the distance between the current waypoint to an unvisited waypoint and α is the constant deciding the distance weight. If α = 0, randomly chooses next waypoint, if α =, chooses the closest waypoint. (UCSC) May / 50

46 (UCSC) May / 50

47 Individual Walk Model 1 Restricts the mobility of each walker to a predicted sub-section of the total area; 2 Selects a subset of fractal waypoint clusters and restrict the movement of each walker to its own designated set of clusters; 3 To add randomness, it allows walkers to move out of their predefined walkabout areas occasionally with some controlled probability; 4 The walker model ensures (F2) and combined with LAPT and fractal waypoints it generates truncated power-law ICTs (F3). (UCSC) May / 50

48 Performance Analysis (UCSC) May / 50

49 Performance Analysis (UCSC) May / 50

50 Performance Analysis - DTN Routing (UCSC) May / 50

MOBILITY MODELS FOR NEXT WIRELESS NETWORKS GENERATION WILEY AD HOC, VEHICULAR AND MESH NETWORKS. Paolo Santi

MOBILITY MODELS FOR NEXT WIRELESS NETWORKS GENERATION WILEY AD HOC, VEHICULAR AND MESH NETWORKS. Paolo Santi MOBILITY MODELS FOR NEXT GENERATION WIRELESS NETWORKS AD HOC, VEHICULAR AND MESH NETWORKS Paolo Santi Istituto di Informatica e Telematica del CNR, Italy WILEY A John Wiley & Sons, Ltd., Publication Contents

More information

Movement Activity Estimation and Forwarding Effects for Opportunistic Networking Based on Urban Mobility Traces

Movement Activity Estimation and Forwarding Effects for Opportunistic Networking Based on Urban Mobility Traces RESEARCH ARTICLE for Opportunistic Networking Based on Urban Mobility Traces Karin Anna Hummel 1 and Andrea Hess 2 1 ETH Zurich, Computer Engineering and Networks Laboratory, Communication Systems Group,

More information

Modeling users mobility among WiFi access points

Modeling users mobility among WiFi access points Modeling users mobility among WiFi access points Minkyong Kim minkyong@cs.dartmouth.edu Department of Computer Science Dartmouth College David Kotz dfk@cs.dartmouth.edu Department of Computer Science Dartmouth

More information

PARAMETERIZATION OF THE SWIM MOBILITY MODEL USING CONTACT TRACES

PARAMETERIZATION OF THE SWIM MOBILITY MODEL USING CONTACT TRACES 1 PARAMETERIZATION OF THE SWIM MOBILITY MODEL USING CONTACT TRACES OMNeT++ Summit 2017, Bremen Zeynep Vatandas, Manikandan Venkateswaran, Koojana Kuladinithi, Andreas Timm-Giel Hamburg University of Technology

More information

Human-Mobility Modeling for Cellular Networks

Human-Mobility Modeling for Cellular Networks Human-Mobility Modeling for Cellular Networks Farnaz Fathali and Hatairatch Charoenkulvanich Degree project in Electrical Engineering Advanced level Stockholm, Sweden 2013 XR-EE-LCN 2013:007 Human-Mobility

More information

ABSTRACT. as the performance study of mobile networks, disease outbreak control, city planning, etc. To

ABSTRACT. as the performance study of mobile networks, disease outbreak control, city planning, etc. To ABSTRACT HONG, SEONGIK. Human Movement Patterns, Mobility Models and Their Impacts on Wireless Applications. (Under the direction of Dr. Injong Rhee.) Simulating human mobility is important in most areas

More information

Mobile Wireless Networking Mobility and Location Management

Mobile Wireless Networking Mobility and Location Management Mobile Wireless Networking The University of Kansas EECS 882 Mobility and James P.G. Sterbenz Department of Electrical Engineering & Computer Science Information Technology & Telecommunications Research

More information

Electronic Theses and Dissertations UC Santa Cruz

Electronic Theses and Dissertations UC Santa Cruz Electronic Theses and Dissertations UC Santa Cruz Peer Reviewed Title: Characterizing User Mobility in Wireless Networks Author: Nunes, Bruno Astuto Arouche Acceptance Date: 01-01-2012 Series: UC Santa

More information

A Mobility Model Based on WLAN Traces and its Validation

A Mobility Model Based on WLAN Traces and its Validation A Mobility Model Based on WLAN Traces and its Validation Cristian Tuduce and Thomas Gross ETH Zurich, Switzerland Abstract The simulation of mobile networks calls for a mobility model to generate the trajectories

More information

Model T: An Empirical Model for User Registration Patterns in a Campus Wireless LAN

Model T: An Empirical Model for User Registration Patterns in a Campus Wireless LAN Model T: An Empirical Model for User Registration Patterns in a Campus Wireless LAN Ravi Jain jain@docomolabsusa.com Dan Lelescu lelescu@docomolabsusa.com DoCoMo Communications Labs USA 8 Metro Drive,

More information

Robotics. Lecture 5: Monte Carlo Localisation. See course website for up to date information.

Robotics. Lecture 5: Monte Carlo Localisation. See course website  for up to date information. Robotics Lecture 5: Monte Carlo Localisation See course website http://www.doc.ic.ac.uk/~ajd/robotics/ for up to date information. Andrew Davison Department of Computing Imperial College London Review:

More information

FMA901F: Machine Learning Lecture 3: Linear Models for Regression. Cristian Sminchisescu

FMA901F: Machine Learning Lecture 3: Linear Models for Regression. Cristian Sminchisescu FMA901F: Machine Learning Lecture 3: Linear Models for Regression Cristian Sminchisescu Machine Learning: Frequentist vs. Bayesian In the frequentist setting, we seek a fixed parameter (vector), with value(s)

More information

Demystifying Levy Walk Patterns in Human Walks

Demystifying Levy Walk Patterns in Human Walks Demystifying Levy Walk Patterns in Human Walks Kyunghan Lee, Seongik Hong, Seong Joon Kim, Injong Rhee and Song Chong School of EECS, KAIST, Daejeon, Korea, khlee@netsys.kaist.ac.kr, song@ee.kaist.ac.kr

More information

Analyzing Mobility Models

Analyzing Mobility Models Wireless and Mobile Communication Group Faculty of Electrical Engineering, Mathematics and Computer Science Delft University of Technology Delft, The Netherlands Analyzing Mobility Models by Yukun Zhao

More information

CLUSTERED MOBILITY MODEL & EFFECT OF NODE CLUSTERING IN MULTI-HOP WIRELESS NETWORKS

CLUSTERED MOBILITY MODEL & EFFECT OF NODE CLUSTERING IN MULTI-HOP WIRELESS NETWORKS CLUSTERED MOBILITY MODEL & EFFECT OF NODE CLUSTERING IN MULTI-HOP WIRELESS NETWORKS VIKRAM SINGH RAWAT Bachelor of Engineering in Electronics & Telecommunications University of Pune, India May, 2000 submitted

More information

Constructing Popular Routes from Uncertain Trajectories

Constructing Popular Routes from Uncertain Trajectories Constructing Popular Routes from Uncertain Trajectories Ling-Yin Wei, Yu Zheng, Wen-Chih Peng presented by Slawek Goryczka Scenarios A trajectory is a sequence of data points recording location information

More information

Sensor Tasking and Control

Sensor Tasking and Control Sensor Tasking and Control Outline Task-Driven Sensing Roles of Sensor Nodes and Utilities Information-Based Sensor Tasking Joint Routing and Information Aggregation Summary Introduction To efficiently

More information

On Exploiting Transient Contact Patterns for Data Forwarding in Delay Tolerant Networks

On Exploiting Transient Contact Patterns for Data Forwarding in Delay Tolerant Networks On Exploiting Transient Contact Patterns for Data Forwarding in Delay Tolerant Networks Wei Gao and Guohong Cao Dept. of Computer Science and Engineering Pennsylvania State University Outline Introduction

More information

A Survey on Group Mobility Model and Handover in Mobile WiMAX Network

A Survey on Group Mobility Model and Handover in Mobile WiMAX Network A Survey on Group Mobility Model and Handover in Mobile WiMAX Network 1 Miss. Ishani R. Desai, 2 Prof. Krunal N. Patel 1 Departement of Electronics & Communication. Parul Institute Of Engineering & Technology,

More information

Modeling Spatial and Temporal Dependencies of User Mobility in Wireless Mobile Networks

Modeling Spatial and Temporal Dependencies of User Mobility in Wireless Mobile Networks 1564 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 17, NO. 5, OCTOBER 2009 Modeling Spatial and Temporal Dependencies of User Mobility in Wireless Mobile Networks Wei-Jen Hsu, Member, IEEE, Thrasyvoulos Spyropoulos,

More information

Study on Indoor and Outdoor environment for Mobile Ad Hoc Network: Random Way point Mobility Model and Manhattan Mobility Model

Study on Indoor and Outdoor environment for Mobile Ad Hoc Network: Random Way point Mobility Model and Manhattan Mobility Model Study on and Outdoor for Mobile Ad Hoc Network: Random Way point Mobility Model and Manhattan Mobility Model Ibrahim khider,prof.wangfurong.prof.yinweihua,sacko Ibrahim khider, Communication Software and

More information

Navigation methods and systems

Navigation methods and systems Navigation methods and systems Navigare necesse est Content: Navigation of mobile robots a short overview Maps Motion Planning SLAM (Simultaneous Localization and Mapping) Navigation of mobile robots a

More information

Network Traffic Measurements and Analysis

Network Traffic Measurements and Analysis DEIB - Politecnico di Milano Fall, 2017 Introduction Often, we have only a set of features x = x 1, x 2,, x n, but no associated response y. Therefore we are not interested in prediction nor classification,

More information

Intelligent Robotics

Intelligent Robotics 64-424 Intelligent Robotics 64-424 Intelligent Robotics http://tams.informatik.uni-hamburg.de/ lectures/2013ws/vorlesung/ir Jianwei Zhang / Eugen Richter Faculty of Mathematics, Informatics and Natural

More information

INDIAN INSTITUTE OF MANAGEMENT CALCUTTA WORKING PAPER SERIES. WPS No. 644/ August A Markov-based Diurnal Mobility Model for 3G Cellular Networks

INDIAN INSTITUTE OF MANAGEMENT CALCUTTA WORKING PAPER SERIES. WPS No. 644/ August A Markov-based Diurnal Mobility Model for 3G Cellular Networks INDIAN INSTITUTE OF MANAGEMENT CALCUTTA WORKING PAPER SERIES WPS No. 644/ August 2009 A Markov-based Diurnal Mobility Model for 3G Cellular Networks by Samir K Sadhukhan SSA, IIM Calcutta, Diamond Harbour

More information

Event Coverage in Theme Parks Using Wireless Sensor Networks with Mobile Sinks

Event Coverage in Theme Parks Using Wireless Sensor Networks with Mobile Sinks Event Coverage in Theme Parks Using Wireless Sensor Networks with Mobile Sinks Gürkan Solmaz and Damla Turgut Department of Electrical Engineering and Computer Science University of Central Florida - Orlando,

More information

SCALING A DISTRIBUTED SPATIAL CACHE OVERLAY. Alexander Gessler Simon Hanna Ashley Marie Smith

SCALING A DISTRIBUTED SPATIAL CACHE OVERLAY. Alexander Gessler Simon Hanna Ashley Marie Smith SCALING A DISTRIBUTED SPATIAL CACHE OVERLAY Alexander Gessler Simon Hanna Ashley Marie Smith MOTIVATION Location-based services utilize time and geographic behavior of user geotagging photos recommendations

More information

08 An Introduction to Dense Continuous Robotic Mapping

08 An Introduction to Dense Continuous Robotic Mapping NAVARCH/EECS 568, ROB 530 - Winter 2018 08 An Introduction to Dense Continuous Robotic Mapping Maani Ghaffari March 14, 2018 Previously: Occupancy Grid Maps Pose SLAM graph and its associated dense occupancy

More information

University of Florida CISE department Gator Engineering. Clustering Part 4

University of Florida CISE department Gator Engineering. Clustering Part 4 Clustering Part 4 Dr. Sanjay Ranka Professor Computer and Information Science and Engineering University of Florida, Gainesville DBSCAN DBSCAN is a density based clustering algorithm Density = number of

More information

Exploring Social Interactions via Multi-Modal Learning

Exploring Social Interactions via Multi-Modal Learning Eploring Social Interactions via Multi-Modal Learning Aarti Munjal Dept. of Electrical Engineering and Computer Science Colorado School of Mines Golden, Colorado 80401 amunjal@mines.edu Thyago Mota Dept.

More information

Clustering Part 4 DBSCAN

Clustering Part 4 DBSCAN Clustering Part 4 Dr. Sanjay Ranka Professor Computer and Information Science and Engineering University of Florida, Gainesville DBSCAN DBSCAN is a density based clustering algorithm Density = number of

More information

CS Introduction to Data Mining Instructor: Abdullah Mueen

CS Introduction to Data Mining Instructor: Abdullah Mueen CS 591.03 Introduction to Data Mining Instructor: Abdullah Mueen LECTURE 8: ADVANCED CLUSTERING (FUZZY AND CO -CLUSTERING) Review: Basic Cluster Analysis Methods (Chap. 10) Cluster Analysis: Basic Concepts

More information

Modeling Movement for Mobile IP Networks

Modeling Movement for Mobile IP Networks Modeling Movement for Mobile IP Networks Andres Rojas anrojas@swin.edu.au Center for Advanced Internet Architectures Swinburne University of Technology 30th March 2005 http://caia.swin.edu.au - anrojas@swin.edu.au

More information

10-701/15-781, Fall 2006, Final

10-701/15-781, Fall 2006, Final -7/-78, Fall 6, Final Dec, :pm-8:pm There are 9 questions in this exam ( pages including this cover sheet). If you need more room to work out your answer to a question, use the back of the page and clearly

More information

Inferring Coarse Views of Connectivity in Very Large Graphs

Inferring Coarse Views of Connectivity in Very Large Graphs Inferring Coarse Views of Connectivity in Very Large Graphs Reza Motamedi, Reza Rejaie, Walter Willinger, Daniel Lowd, Roberto Gonzalez http://onrg.cs.uoregon.edu/walkabout 10/8/14 1 Introduction! Large-scale

More information

INTERNET TRAFFIC MEASUREMENT (PART II) Gaia Maselli

INTERNET TRAFFIC MEASUREMENT (PART II) Gaia Maselli INTERNET TRAFFIC MEASUREMENT (PART II) Gaia Maselli maselli@di.uniroma1.it Prestazioni dei sistemi di rete 2 Overview Basic concepts Characterization of traffic properties that are important to measure

More information

Scalable Selective Traffic Congestion Notification

Scalable Selective Traffic Congestion Notification Scalable Selective Traffic Congestion Notification Győző Gidófalvi Division of Geoinformatics Deptartment of Urban Planning and Environment KTH Royal Institution of Technology, Sweden gyozo@kth.se Outline

More information

Challenges in Ubiquitous Data Mining

Challenges in Ubiquitous Data Mining LIAAD-INESC Porto, University of Porto, Portugal jgama@fep.up.pt 1 2 Very-short-term Forecasting in Photovoltaic Systems 3 4 Problem Formulation: Network Data Model Querying Model Query = Q( n i=0 S i)

More information

Evaluation of Mobility Models with AODV & OLSR Protocol by Varying Node Speed in MANET

Evaluation of Mobility Models with AODV & OLSR Protocol by Varying Node Speed in MANET Evaluation of Mobility Models with AODV & OLSR Protocol by Varying Node Speed in MANET Smt. Rekha Shahapurkar 1, Dr. Umesh Kumar Singh 2, Sh. Yogesh Mishra 3 1 Reasearch Scholar, 2 Reader 3 Assistant ptofessor

More information

Modeling Spatial and Temporal Dependencies of User Mobility in Wireless Mobile Networks

Modeling Spatial and Temporal Dependencies of User Mobility in Wireless Mobile Networks 1 Modeling Spatial and Temporal Dependencies of User Mobility in Wireless Mobile Networks Wei-jen Hsu, Thrasyvoulos Spyropoulos, Konstantinos Psounis and Ahmed Helmy Dept. of Computer and Information Science

More information

Clustering: Classic Methods and Modern Views

Clustering: Classic Methods and Modern Views Clustering: Classic Methods and Modern Views Marina Meilă University of Washington mmp@stat.washington.edu June 22, 2015 Lorentz Center Workshop on Clusters, Games and Axioms Outline Paradigms for clustering

More information

Machine Learning A W 1sst KU. b) [1 P] Give an example for a probability distributions P (A, B, C) that disproves

Machine Learning A W 1sst KU. b) [1 P] Give an example for a probability distributions P (A, B, C) that disproves Machine Learning A 708.064 11W 1sst KU Exercises Problems marked with * are optional. 1 Conditional Independence I [2 P] a) [1 P] Give an example for a probability distribution P (A, B, C) that disproves

More information

Artificial Intelligence for Robotics: A Brief Summary

Artificial Intelligence for Robotics: A Brief Summary Artificial Intelligence for Robotics: A Brief Summary This document provides a summary of the course, Artificial Intelligence for Robotics, and highlights main concepts. Lesson 1: Localization (using Histogram

More information

Supplementary Figure 1. Decoding results broken down for different ROIs

Supplementary Figure 1. Decoding results broken down for different ROIs Supplementary Figure 1 Decoding results broken down for different ROIs Decoding results for areas V1, V2, V3, and V1 V3 combined. (a) Decoded and presented orientations are strongly correlated in areas

More information

CMX Analytics Documentation

CMX Analytics Documentation CMX Analytics Documentation Version 10.3.1 CMX Analytics Definitions Detected Signals Signals sent from devices to WIFI Access Points (AP) are the basis of CMX Analytics. There are two types of signals

More information

Experimental Evaluation of Gateway Based Interdomain Routing Scheme for DTN

Experimental Evaluation of Gateway Based Interdomain Routing Scheme for DTN Lehigh University Lehigh Preserve Theses and Dissertations 2011 Experimental Evaluation of Gateway Based Interdomain Routing Scheme for DTN Dongli Li Lehigh University Follow this and additional works

More information

Mobility Model for User s Realistic Behavior in Mobile Ad Hoc Network

Mobility Model for User s Realistic Behavior in Mobile Ad Hoc Network Mobility Model for User s Realistic Behavior in Mobile Ad Hoc Network Shiddhartha Raj Bhandari, Gyu Myoung Lee, and Noel Crespi Department of Wireless Networks and Multimedia Services Institut Telecom,

More information

Detecting and Identifying Moving Objects in Real-Time

Detecting and Identifying Moving Objects in Real-Time Chapter 9 Detecting and Identifying Moving Objects in Real-Time For surveillance applications or for human-computer interaction, the automated real-time tracking of moving objects in images from a stationary

More information

Chapter-4. Simulation Design and Implementation

Chapter-4. Simulation Design and Implementation Chapter-4 Simulation Design and Implementation In this chapter, the design parameters of system and the various metrics measured for performance evaluation of the routing protocols are presented. An overview

More information

Efficient Orienteering-Route Search over Uncertain Spatial Datasets

Efficient Orienteering-Route Search over Uncertain Spatial Datasets Efficient Orienteering-Route Search over Uncertain Spatial Datasets Mr. Nir DOLEV, Israel Dr. Yaron KANZA, Israel Prof. Yerach DOYTSHER, Israel 1 Route Search A standard search engine on the WWW returns

More information

Accelerometer Gesture Recognition

Accelerometer Gesture Recognition Accelerometer Gesture Recognition Michael Xie xie@cs.stanford.edu David Pan napdivad@stanford.edu December 12, 2014 Abstract Our goal is to make gesture-based input for smartphones and smartwatches accurate

More information

Unsupervised: no target value to predict

Unsupervised: no target value to predict Clustering Unsupervised: no target value to predict Differences between models/algorithms: Exclusive vs. overlapping Deterministic vs. probabilistic Hierarchical vs. flat Incremental vs. batch learning

More information

Empirical Evaluation of Hybrid Opportunistic Networks

Empirical Evaluation of Hybrid Opportunistic Networks Empirical Evaluation of Hybrid Opportunistic Networks Pan Hui Joint work with Anders Lindgren and Jon Crowcroft (University of Cambridge) 1 Introduction Two trends observed Lots of work done on opportunistic

More information

IDENTIFYING OPTICAL TRAP

IDENTIFYING OPTICAL TRAP IDENTIFYING OPTICAL TRAP Yulwon Cho, Yuxin Zheng 12/16/2011 1. BACKGROUND AND MOTIVATION Optical trapping (also called optical tweezer) is widely used in studying a variety of biological systems in recent

More information

Hidden Markov Models. Slides adapted from Joyce Ho, David Sontag, Geoffrey Hinton, Eric Xing, and Nicholas Ruozzi

Hidden Markov Models. Slides adapted from Joyce Ho, David Sontag, Geoffrey Hinton, Eric Xing, and Nicholas Ruozzi Hidden Markov Models Slides adapted from Joyce Ho, David Sontag, Geoffrey Hinton, Eric Xing, and Nicholas Ruozzi Sequential Data Time-series: Stock market, weather, speech, video Ordered: Text, genes Sequential

More information

COMPUTER AND ROBOT VISION

COMPUTER AND ROBOT VISION VOLUME COMPUTER AND ROBOT VISION Robert M. Haralick University of Washington Linda G. Shapiro University of Washington A^ ADDISON-WESLEY PUBLISHING COMPANY Reading, Massachusetts Menlo Park, California

More information

Effect of Group Mobility Models on Routing Protocols in MANET

Effect of Group Mobility Models on Routing Protocols in MANET Effect of Group Mobility Models on Routing Protocols in MANET Smt. Rekha Shahapurkar 1, Dr. Umesh Kumar Singh 2, Leena Chowdhary 3 1 Researcher, 2 Reader 3 ICS Vikram University, Ujjain Abstract: Mobile

More information

Chapter 2 Mobility Model Characteristics

Chapter 2 Mobility Model Characteristics Chapter 2 Mobility Model Characteristics Abstract The salient characteristics of mobility models of mobile nodes in mobile ad hoc networks are described. We have described how the different mobility models

More information

Data gathering using mobile agents for reducing traffic in dense mobile wireless sensor networks

Data gathering using mobile agents for reducing traffic in dense mobile wireless sensor networks Mobile Information Systems 9 (23) 295 34 295 DOI.3233/MIS-364 IOS Press Data gathering using mobile agents for reducing traffic in dense mobile wireless sensor networks Keisuke Goto, Yuya Sasaki, Takahiro

More information

Modeling Spatial and Temporal Dependencies of User Mobility in Wireless Mobile Networks

Modeling Spatial and Temporal Dependencies of User Mobility in Wireless Mobile Networks 1 Modeling Spatial and Temporal Dependencies of User Mobility in Wireless Mobile Networks Wei-Jen Hsu, Thrasyvoulos Spyropoulos, Konstantinos Psounis and Ahmed Helmy Dept. of Computer and Information Science

More information

Social-Aware Routing in Delay Tolerant Networks

Social-Aware Routing in Delay Tolerant Networks Social-Aware Routing in Delay Tolerant Networks Jie Wu Dept. of Computer and Info. Sciences Temple University Challenged Networks Assumptions in the TCP/IP model are violated DTNs Delay-Tolerant Networks

More information

Range Imaging Through Triangulation. Range Imaging Through Triangulation. Range Imaging Through Triangulation. Range Imaging Through Triangulation

Range Imaging Through Triangulation. Range Imaging Through Triangulation. Range Imaging Through Triangulation. Range Imaging Through Triangulation Obviously, this is a very slow process and not suitable for dynamic scenes. To speed things up, we can use a laser that projects a vertical line of light onto the scene. This laser rotates around its vertical

More information

Novel Sampling Algorithm for Levy-Walk Based Mobile Phone Sensing

Novel Sampling Algorithm for Levy-Walk Based Mobile Phone Sensing Novel Sampling Algorithm for Levy-Walk Based Mobile Phone Sensing Thejaswini M, P. Rajalakshmi, U. B. Desai Department of Computer Science and Engineering Department of Electrical Engineering Indian Institute

More information

Sensor Modalities. Sensor modality: Different modalities:

Sensor Modalities. Sensor modality: Different modalities: Sensor Modalities Sensor modality: Sensors which measure same form of energy and process it in similar ways Modality refers to the raw input used by the sensors Different modalities: Sound Pressure Temperature

More information

MCMC Methods for data modeling

MCMC Methods for data modeling MCMC Methods for data modeling Kenneth Scerri Department of Automatic Control and Systems Engineering Introduction 1. Symposium on Data Modelling 2. Outline: a. Definition and uses of MCMC b. MCMC algorithms

More information

Introduction to Machine Learning CMU-10701

Introduction to Machine Learning CMU-10701 Introduction to Machine Learning CMU-10701 Clustering and EM Barnabás Póczos & Aarti Singh Contents Clustering K-means Mixture of Gaussians Expectation Maximization Variational Methods 2 Clustering 3 K-

More information

Workload Characterization Techniques

Workload Characterization Techniques Workload Characterization Techniques Raj Jain Washington University in Saint Louis Saint Louis, MO 63130 Jain@cse.wustl.edu These slides are available on-line at: http://www.cse.wustl.edu/~jain/cse567-08/

More information

Table of Contents (As covered from textbook)

Table of Contents (As covered from textbook) Table of Contents (As covered from textbook) Ch 1 Data and Decisions Ch 2 Displaying and Describing Categorical Data Ch 3 Displaying and Describing Quantitative Data Ch 4 Correlation and Linear Regression

More information

The Challenges of Measuring Wireless Networks. David Kotz Dartmouth College August 2005

The Challenges of Measuring Wireless Networks. David Kotz Dartmouth College August 2005 The Challenges of Measuring Wireless Networks David Kotz Dartmouth College August 2005 Why measure? For better management Large WLANs need scalable management To manage you must monitor To monitor you

More information

Approximation Algorithms for Clustering Uncertain Data

Approximation Algorithms for Clustering Uncertain Data Approximation Algorithms for Clustering Uncertain Data Graham Cormode AT&T Labs - Research graham@research.att.com Andrew McGregor UCSD / MSR / UMass Amherst andrewm@ucsd.edu Introduction Many applications

More information

PARALLEL AND DISTRIBUTED PLATFORM FOR PLUG-AND-PLAY AGENT-BASED SIMULATIONS. Wentong CAI

PARALLEL AND DISTRIBUTED PLATFORM FOR PLUG-AND-PLAY AGENT-BASED SIMULATIONS. Wentong CAI PARALLEL AND DISTRIBUTED PLATFORM FOR PLUG-AND-PLAY AGENT-BASED SIMULATIONS Wentong CAI Parallel & Distributed Computing Centre School of Computer Engineering Nanyang Technological University Singapore

More information

Solution Sketches Midterm Exam COSC 6342 Machine Learning March 20, 2013

Solution Sketches Midterm Exam COSC 6342 Machine Learning March 20, 2013 Your Name: Your student id: Solution Sketches Midterm Exam COSC 6342 Machine Learning March 20, 2013 Problem 1 [5+?]: Hypothesis Classes Problem 2 [8]: Losses and Risks Problem 3 [11]: Model Generation

More information

Overview. Monte Carlo Methods. Statistics & Bayesian Inference Lecture 3. Situation At End Of Last Week

Overview. Monte Carlo Methods. Statistics & Bayesian Inference Lecture 3. Situation At End Of Last Week Statistics & Bayesian Inference Lecture 3 Joe Zuntz Overview Overview & Motivation Metropolis Hastings Monte Carlo Methods Importance sampling Direct sampling Gibbs sampling Monte-Carlo Markov Chains Emcee

More information

Spatio-Temporal Stereo Disparity Integration

Spatio-Temporal Stereo Disparity Integration Spatio-Temporal Stereo Disparity Integration Sandino Morales and Reinhard Klette The.enpeda.. Project, The University of Auckland Tamaki Innovation Campus, Auckland, New Zealand pmor085@aucklanduni.ac.nz

More information

Random Graph Model; parameterization 2

Random Graph Model; parameterization 2 Agenda Random Graphs Recap giant component and small world statistics problems: degree distribution and triangles Recall that a graph G = (V, E) consists of a set of vertices V and a set of edges E V V.

More information

Introduction to Data Mining

Introduction to Data Mining Introduction to JULY 2011 Afsaneh Yazdani What motivated? Wide availability of huge amounts of data and the imminent need for turning such data into useful information and knowledge What motivated? Data

More information

Small World in Motion (SWIM): Modeling Communities in Ad-Hoc Mobile Networking

Small World in Motion (SWIM): Modeling Communities in Ad-Hoc Mobile Networking Small World in Motion (SWIM): Modeling Communities in Ad-Hoc Mobile Networking Sokol Kosta, Alessandro Mei, and Julinda Stefa Department of Computer Science Sapienza University of Rome, Italy Email: {kosta,

More information

DS595/CS525: Urban Network Analysis --Urban Mobility Prof. Yanhua Li

DS595/CS525: Urban Network Analysis --Urban Mobility Prof. Yanhua Li Welcome to DS595/CS525: Urban Network Analysis --Urban Mobility Prof. Yanhua Li Time: 6:00pm 8:50pm Wednesday Location: Fuller 320 Spring 2017 2 Team assignment Finalized. (Great!) Guest Speaker 2/22 A

More information

ECE 176 Digital Image Processing Handout #14 Pamela Cosman 4/29/05 TEXTURE ANALYSIS

ECE 176 Digital Image Processing Handout #14 Pamela Cosman 4/29/05 TEXTURE ANALYSIS ECE 176 Digital Image Processing Handout #14 Pamela Cosman 4/29/ TEXTURE ANALYSIS Texture analysis is covered very briefly in Gonzalez and Woods, pages 66 671. This handout is intended to supplement that

More information

Clustering CS 550: Machine Learning

Clustering CS 550: Machine Learning Clustering CS 550: Machine Learning This slide set mainly uses the slides given in the following links: http://www-users.cs.umn.edu/~kumar/dmbook/ch8.pdf http://www-users.cs.umn.edu/~kumar/dmbook/dmslides/chap8_basic_cluster_analysis.pdf

More information

Part I. Hierarchical clustering. Hierarchical Clustering. Hierarchical clustering. Produces a set of nested clusters organized as a

Part I. Hierarchical clustering. Hierarchical Clustering. Hierarchical clustering. Produces a set of nested clusters organized as a Week 9 Based in part on slides from textbook, slides of Susan Holmes Part I December 2, 2012 Hierarchical Clustering 1 / 1 Produces a set of nested clusters organized as a Hierarchical hierarchical clustering

More information

Mixture Models and the EM Algorithm

Mixture Models and the EM Algorithm Mixture Models and the EM Algorithm Padhraic Smyth, Department of Computer Science University of California, Irvine c 2017 1 Finite Mixture Models Say we have a data set D = {x 1,..., x N } where x i is

More information

IBL and clustering. Relationship of IBL with CBR

IBL and clustering. Relationship of IBL with CBR IBL and clustering Distance based methods IBL and knn Clustering Distance based and hierarchical Probability-based Expectation Maximization (EM) Relationship of IBL with CBR + uses previously processed

More information

OPS An Opportunistic Networking Protocol Simulator for OMNeT++

OPS An Opportunistic Networking Protocol Simulator for OMNeT++ OPS An Opportunistic Networking Protocol Simulator for OMNeT++ Asanga Udugama, Anna Förster, Jens Dede, Vishnupriya Kuppusamy and Anas bin Muslim University of Bremen, Germany OMNeT++ Community Summit

More information

Classification of MANET: A Review

Classification of MANET: A Review Classification of MANET: A Review Smita Das Assistant Professor. Dasaratha Deb Memorial College. Khowai,Tripura,India. Anuja Sarkar Informatics Research Officer. Forest Department. Agartala, Tripura, India.

More information

Probabilistic Modeling of Leach Protocol and Computing Sensor Energy Consumption Rate in Sensor Networks

Probabilistic Modeling of Leach Protocol and Computing Sensor Energy Consumption Rate in Sensor Networks Probabilistic Modeling of Leach Protocol and Computing Sensor Energy Consumption Rate in Sensor Networks Dezhen Song CS Department, Texas A&M University Technical Report: TR 2005-2-2 Email: dzsong@cs.tamu.edu

More information

Crowd Scene Understanding with Coherent Recurrent Neural Networks

Crowd Scene Understanding with Coherent Recurrent Neural Networks Crowd Scene Understanding with Coherent Recurrent Neural Networks Hang Su, Yinpeng Dong, Jun Zhu May 22, 2016 Hang Su, Yinpeng Dong, Jun Zhu IJCAI 2016 May 22, 2016 1 / 26 Outline 1 Introduction 2 LSTM

More information

Localization and Map Building

Localization and Map Building Localization and Map Building Noise and aliasing; odometric position estimation To localize or not to localize Belief representation Map representation Probabilistic map-based localization Other examples

More information

Opportunistic Routing Algorithms in Delay Tolerant Networks

Opportunistic Routing Algorithms in Delay Tolerant Networks Opportunistic Routing Algorithms in Delay Tolerant Networks Eyuphan Bulut Rensselaer Polytechnic Institute Department of Computer Science and Network Science and Technology (NeST) Center PhD Thesis Defense

More information

The exam is closed book, closed notes except your one-page cheat sheet.

The exam is closed book, closed notes except your one-page cheat sheet. CS 189 Fall 2015 Introduction to Machine Learning Final Please do not turn over the page before you are instructed to do so. You have 2 hours and 50 minutes. Please write your initials on the top-right

More information

SWIM: A Simple Model to Generate Small Mobile Worlds

SWIM: A Simple Model to Generate Small Mobile Worlds : A Simple Model to Generate Small Mobile Worlds Alessandro Mei and Julinda Stefa Department of Computer Science Sapienza University of Rome, Italy Email: {mei, stefa}@di.uniroma.it arxiv:0809.2730v2 [cs.dc]

More information

Mobile based Text Image Translation System for Smart Tourism. Saw Zay Maung Maung UCSY, Myanmar. 23 November 2017, Brunei

Mobile based Text Image Translation System for Smart Tourism. Saw Zay Maung Maung UCSY, Myanmar. 23 November 2017, Brunei Mobile based Text Image Translation System for Smart Tourism Saw Zay Maung Maung UCSY, Myanmar. 23 November 2017, Brunei 1 Smart Tourism Tourism is cultural and economic phenomenon which entails the movement

More information

Humanoid Robotics. Monte Carlo Localization. Maren Bennewitz

Humanoid Robotics. Monte Carlo Localization. Maren Bennewitz Humanoid Robotics Monte Carlo Localization Maren Bennewitz 1 Basis Probability Rules (1) If x and y are independent: Bayes rule: Often written as: The denominator is a normalizing constant that ensures

More information

International Journal of Scientific & Engineering Research, Volume 7, Issue 12, December ISSN

International Journal of Scientific & Engineering Research, Volume 7, Issue 12, December ISSN International Journal of Scientific & Engineering Research, Volume 7, Issue 12, December-2016 286 A SURVEY ON FACTORS EFFECTING ENERGY CONSUMPTION IN MANET Pooja Ph.D Scholar, Department of Computer Science

More information

Publishing CitiSense Data: Privacy Concerns and Remedies

Publishing CitiSense Data: Privacy Concerns and Remedies Publishing CitiSense Data: Privacy Concerns and Remedies Kapil Gupta Advisor : Prof. Bill Griswold 1 Location Based Services Great utility of location based services data traffic control, mobility management,

More information

Fosca Giannotti et al,.

Fosca Giannotti et al,. Trajectory Pattern Mining Fosca Giannotti et al,. - Presented by Shuo Miao Conference on Knowledge discovery and data mining, 2007 OUTLINE 1. Motivation 2. T-Patterns: definition 3. T-Patterns: the approach(es)

More information

TELCOM2125: Network Science and Analysis

TELCOM2125: Network Science and Analysis School of Information Sciences University of Pittsburgh TELCOM2125: Network Science and Analysis Konstantinos Pelechrinis Spring 2015 Figures are taken from: M.E.J. Newman, Networks: An Introduction 2

More information

CHAPTER 5 FUZZY LOGIC CONTROL

CHAPTER 5 FUZZY LOGIC CONTROL 64 CHAPTER 5 FUZZY LOGIC CONTROL 5.1 Introduction Fuzzy logic is a soft computing tool for embedding structured human knowledge into workable algorithms. The idea of fuzzy logic was introduced by Dr. Lofti

More information

CS4495 Fall 2014 Computer Vision Problem Set 6: Particle Tracking

CS4495 Fall 2014 Computer Vision Problem Set 6: Particle Tracking CS4495 Fall 2014 Computer Vision Problem Set 6: Particle Tracking DUE Tues, November 25-11:55pm Here you are going to experiment with a particle filter tracker such as was described in class. Recall that

More information

Non-Bayesian Classifiers Part I: k-nearest Neighbor Classifier and Distance Functions

Non-Bayesian Classifiers Part I: k-nearest Neighbor Classifier and Distance Functions Non-Bayesian Classifiers Part I: k-nearest Neighbor Classifier and Distance Functions Selim Aksoy Department of Computer Engineering Bilkent University saksoy@cs.bilkent.edu.tr CS 551, Fall 2017 CS 551,

More information