Difference between revisions of "Group05 Proposal"
Ivygoh.2017 (talk | contribs) |
|||
| Line 31: | Line 31: | ||
==Background on Time Series Clustering== | ==Background on Time Series Clustering== | ||
<b>What is Time Series Clustering?</b> | <b>What is Time Series Clustering?</b> | ||
| − | + | Clustering is a data analysis technique for organizing observed data (e.g. people, things, events, brands, companies) into meaningful taxonomies, groups or clusters without advanced knowledge of the groups’ definition. Clusters are formed based on combinations of input variable, which maximizes the similarity of cases within each cluster while maximizing the dissimilarity between groups that are initially unknows. | |
| − | + | A special type of clustering is time-series clustering, which is essentially dynamic data as its feature values changes as a function of time. | |
<br> | <br> | ||
| − | + | <b>Key Parameters of Time-Series Clustering: </b> | |
| − | <b>Key | + | {| class="wikitable" |
| − | + | |- | |
| − | + | ! Parameters !! Algorithm | |
| − | + | |- | |
| − | + | | Type|| | |
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
* Hierarchical Clustering | * Hierarchical Clustering | ||
* Partitional Clustering | * Partitional Clustering | ||
| − | + | |- | |
| − | + | | Distances|| | |
| − | + | * Dynamic Time Wrapping (DTW) | |
| − | + | * Global Alignment Kernels (GAK) | |
| − | + | * Shape-Based Distance (SBD) | |
| − | + | |- | |
| − | + | | Centroid || | |
| − | + | * DTW Barycenter Averaging (DBA) | |
| − | * | + | * Partitioning Around Medoids (PAM) |
| − | * | + | * Shape Averaging (Shape) |
| − | + | |} | |
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | * | ||
| − | * | ||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
<br> | <br> | ||
| Line 86: | Line 66: | ||
<b>To build an application for Time Series Clustering</b> | <b>To build an application for Time Series Clustering</b> | ||
<br> | <br> | ||
| − | This dashboard aims to allow user to do time series clustering on time series related data to uncover patterns which have potential use case in the respective | + | Time-series data are of interest due to their ubiquity in various areas ranging from science, engineering, business, economics, healthcare, to government. This dashboard aims to allow user to do time series clustering on time series related data to uncover patterns which have potential use case in the respective domain. |
| − | + | <br> | |
| − | |||
| − | |||
Revision as of 17:15, 20 November 2018
Visual Application for Time Series Clustering
|
|
|
|
|
Abstract
XXXXX
Background on Time Series Clustering
What is Time Series Clustering?
Clustering is a data analysis technique for organizing observed data (e.g. people, things, events, brands, companies) into meaningful taxonomies, groups or clusters without advanced knowledge of the groups’ definition. Clusters are formed based on combinations of input variable, which maximizes the similarity of cases within each cluster while maximizing the dissimilarity between groups that are initially unknows.
A special type of clustering is time-series clustering, which is essentially dynamic data as its feature values changes as a function of time.
Key Parameters of Time-Series Clustering:
| Parameters | Algorithm |
|---|---|
| Type |
|
| Distances |
|
| Centroid |
|
Packages Used
This dashboard mainly uses dtwclust package from R.
dtwclust:
XXXX
Objective
To build an application for Time Series Clustering
Time-series data are of interest due to their ubiquity in various areas ranging from science, engineering, business, economics, healthcare, to government. This dashboard aims to allow user to do time series clustering on time series related data to uncover patterns which have potential use case in the respective domain.
