PP: Time series clustering via community detection in Networks
Improvement can be done in fulture:
1. the algorithm of constructing network from distance matrix.
2. evolution of sliding time window
3. the later processing or visual analysis of generated graphs.
Thinking:
1.What's the ground truth in load profiles?
For clustering, there's no ground truth, so how to tune the parameters or options in step2, step3 and step4? In this paper, they have the labels of time series, so they use RI to guide their selection of parameters, for example: k and \epsilon.
Suppose: similar time series tend to connect to each other and form communities.
Background and related works
shaped based distance measures; feature based distance measures; structure based distance measures. time series clustering; community detection in networks.
Methodology
- data normalization
- time series distance calculation
- network construction
- community detection
Which step influence the clustering results:
distance calculation algorithm; network construction methods. community detection methods.
2. distance matrix
calculating the distance for each pair of time series in the data set and construct a distance matrix D, where dij is the distance between series Xi and XJ . A good choice of distance measure has strong influence on the network construction and clustering result.
3. network construction
Two common method: K-NN and \epsilon-NN; EXPLORATION
Experiments
45 time series data sets.
Purpose: check the performance of each combination of step2, step3,and step4 to each data sets.
Index指标:Rand index.
Vary the parameters: the k of k-NN from 1 to n-1; the epsilon of epsilon-NN from min(D) to max(D) in 100 steps.
Step2: Manhattan, Euclidean, infinite Norm, DTW, short time series, DISSIM, Complexity-Invariant, Wavlet tranform, Pearson correlation, Intergrated periodogram.
Step3: fast greedy; multilevel; walktrap; infomap; label propagration.
Step4: vary the parameter of k and \epsilon.
Results
1. the effect of k and \epsilon to the clustering results(RI).
The k-NN construction method just allows discrete values of k while the ε-NN method accepts continuous values. When k and ε are small, vertices tend to make just few connections.
??what's the meaning of A,B,C,D in figure 5.
2. the statistical test of the effect of different distance methods. Friedman test and Nemenyi test.
多个算法在多个数据库上的对比:
- 如果样本符合ANOVA(repeated measure)的假设(如正态、等方差),优先使用ANOVA。
- 如果样本不符合ANOVA的假设,使用Friedman test配合Nemenyi test做post-hoc。
- 如果样本量不一样,或因为特定原因不能使用Friedman-Nemenyi,可以尝试Kruskal Wallis配合Dunn's test。值得注意的是,这种方法是用来处理独立测量数据,要分情况讨论。
DTW measure presents the best results for both network construction methods.
3. the statistical test of the effect of community detection algorithms. Friedman test and Nemenyi test.
4. comparison to rival methods.
i. some classic clustering algorithms: k-medoids, complete-linkage, single-linkage, average-linkage, median-linkage, centroid-linkage and diana;
ii. three up-to-date ones: Zhang’s method [41], Maharaj’s method [24] and PDC [5]
5. detect time series clusters with time-shifts
Suppose: Clustering algorithms should be capable of detecting groups of time series that have similar variations in time.
CBF dataset: 30个序列,一共三组, 全部正确分组/clustering.
6. detect shape patterns
1000 time series of length 128, four groups.
detect shape patterns (UD, DD, DU, UU);
Discussion
1. the same idea can be extended to multivariate time series clustering.
2. evaluate the simulation results using different indexes.
3. As future works, we plan to propose automatic strategies for choosing the best number of neighbors (k and ε) and speeding up the network construction method, instead of using the naive method.
4. We also plan to apply the idea to solve other kinds of problems in time series analysis, such as time series prediction. ??
Supplementary knowledge:
1. box plot
它能显示出一组数据的最大值、最小值、中位数、及上下四分位数。
以下是箱形图的具体例子:
- +-----+-+
- * o |-------| + | |---|
- +-----+-+
- +---+---+---+---+---+---+---+---+---+---+ 分数
- 0 1 2 3 4 5 6 7 8 9 10
这组数据显示出:
- 最小值(minimum)=5
- 下四分位数(Q1)=7
- 中位数(Med --也就是Q2)=8.5
- 上四分位数(Q3)=9
- 最大值(maximum )=10
- 平均值=8
- 四分位间距(interquartile range)={\displaystyle (Q3-Q1)}
=2 (即ΔQ)
2. 观念转变, experiment部分也很重要,不是可有可无的, 要细看。
3. 统计学检验
All models are wrong, but some are useful. ----------统计学家George Box.
4. univariate and multivariate time series.
Univariate time series: Only one variable is varying over time. For example, data collected from a sensor measuring the temperature of a room every second. Therefore, each second, you will only have a one-dimensional value, which is the temperature.
Multivariate time series: Multiple variables are varying over time. For example, a tri-axial accelerometer三轴加速器. There are three accelerations, one for each axis (x,y,z) and they vary simultaneously over time.
Considering the data you showed in the question, you are dealing with a multivariate time series, where value_1
, value_2
andvalue_3
are three variables changing simultaneously over time.
PP: Time series clustering via community detection in Networks的更多相关文章
- PP: Learning representations for time series clustering
Problem: time series clustering TSC - unsupervised learning/ category information is not available. ...
- 【论文阅读】A practical algorithm for distributed clustering and outlier detection
文章提出了一种分布式聚类的算法,这是第一个有理论保障的考虑离群点的分布式聚类算法(文章里自己说的).与之前的算法对比有以下四个优点: 1.耗时短O(max{k,logn}*n), 2.传递信息规模小: ...
- 论文解读(CGC)《CGC: Contrastive Graph Clustering for Community Detection and Tracking》
论文信息 论文标题:CGC: Contrastive Graph Clustering for Community Detection and Tracking论文作者:Namyong Park, R ...
- A Node Influence Based Label Propagation Algorithm for Community detection in networks 文章算法实现的疑问
这是我最近看到的一篇论文,思路还是很清晰的,就是改进的LPA算法.改进的地方在两个方面: (1)结合K-shell算法计算量了节点重重要度NI(node importance),标签更新顺序则按照NI ...
- LabelRank非重叠社区发现算法介绍及代码实现(A Stabilized Label Propagation Algorithm for Community Detection in Networks)
最近在研究基于标签传播的社区分类,LabelRank算法基于标签传播和马尔科夫随机游走思路上改装的算法,引用率较高,打算将代码实现,便于加深理解. 这个算法和Label Propagation 算法不 ...
- PP: Time series anomaly detection with variational autoencoders
Problem: unsupervised anomaly detection Model: VAE-reEncoder VAE with two encoders and one decoder. ...
- [Localization] R-CNN series for Localization and Detection
CS231n Winter 2016: Lecture 8 : Localization and Detection CS231n Winter 2017: Lecture 11: Detection ...
- PP: Toeplitz Inverse Covariance-Based Clustering of Multivariate Time Series Data
From: Stanford University; Jure Leskovec, citation 6w+; Problem: subsequence clustering. Challenging ...
- 关于目标检测(Object Detection)的文献整理
本文对CV中目标检测子方向的研究,整理了如下的相关笔记(持续更新中): 1. Cascade R-CNN: Delving into High Quality Object Detection 年份: ...
随机推荐
- .net core 开发 Windows Forms 程序
我是一名 ASP.NET 程序员,专注于 B/S 项目开发.累计文章阅读量超过一千万,我的博客主页地址:https://www.itsvse.com/blog_xzz.html 引言 .net cor ...
- 百度api识别验证码登录
import time from selenium import webdriver from aip import AipOcr def initial(): """ ...
- JS变量+作用域
基本类型-栈内存 保存基本类型的变量保存的是值本身 引用类型-堆内存 保存引用类型的变量保存的并不是对象本身,而是一个指向该对象的引用地址 引用类型判断对象是否相等 function equalObj ...
- WebApp开发-Zepto
zepto.js自己去官网下载哈. DOM操作 $(document).ready(function(){ var $cr = $("<div class='cr'>插入的div ...
- 嵊州D5T3 指令 program 神奇的位运算
指令 program [问题描述] krydom 有一个神奇的机器. 一开始,可以往机器里输入若干条指令: opt x 其中,opt 是 & | ^ 中的一种,0 ≤ x ≤ 1023 . 对 ...
- Linux网络课程学习第二天
第二天学习日志: 今天的课程主要内容: 详细介绍了如何安装红帽RHEL7的系统,并对RPM,Yum,Systemd和bash进行了简单介绍.
- 百度地图根据list经纬度算每个点到剩余点的平均距离、最远距离和最近距离
一.使用步骤 说明:给你一大串坐标list.计算每个点到其他所有点的最近最远平均距离. 1.打开百度地图api在线demo(随便一个都行) 2.替换<script></script& ...
- C语言118. 杨辉三角
给定一个非负整数 numRows,生成杨辉三角的前 numRows 行. 在杨辉三角中,每个数是它左上方和右上方的数的和. 示例: 输入: 5输出:[ [1], [1,1], [1,2,1], [1, ...
- python3练习100题——045
题目:统计 1 到 100 之和. sum(range(1,101)) 题目太容易了,我都不想用迭代浪费时间. 觉得这100道题难度设计越来越不合理.
- Windows 性能监视器的基本指标(CPU,内存,硬盘参数)
转载:http://kms.lenovots.com/kb/article.php?id=7045 Windows 性能监视器的基本指标(CPU,内存,硬盘参数) 作为一个系统工程师来说,要看懂监控的 ...