Commit bb4bdad4 authored by Maciej Dobrzynski's avatar Maciej Dobrzynski Committed by GitHub

Added a link to screencast

parent c29dbeee
......@@ -17,6 +17,8 @@
## What is TCI?
Time Course Inspector (TCI) is a software for visualization, analysis and clustering of time-series. The driving philosophy is to provide a simple, yet flexible GUI to perform various time-series analyses without any programming knowledge. TCI is written as an R shiny web app which provides a reactive, fast and customizable framework to produce elegant visualizations.
The **screencast** with an overview of the app's functionality and processing of the [test-case-1](https://github.com/dmattek/shiny-timecourse-inspector/tree/master/example-data/test-case-1 "Link to data") dataset can be found [here](https://youtu.be/pwRqUzf5HMs "Link to YouTube").
The starting point of the app is a plain spreadsheet in comma-separated values format (CSV) that contains the dataset to analyze. TCI embeds a module for simple preprocessing of the data (normalization, missing values handling...), various visualizations, common statistics report, spectral decomposition, a flexible module for hierarchical clustering and cluster validation. All modules are documented with tooltips and *Learn More* sections to guide users through the UI and assist them with the analysis.
TCI puts the accent on spotting the unique dynamics that underly the time-series in the datasets. Its full potential is revealed when working with groups of time-series from which one wishes to understand their respective specificity and differences. To this effect, the clustering module works on pooled data, meaning that all groups of series will be considered at once when running the clustering procedure. The resulting clusters can then be inspected to visualize which types of series, patterns and frequencies were captured by each cluster. In addition, the distribution of clusters acroos the groups can be visualized to understand which dynamics or trends were over/under-represented in each group.
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment