Cluster Power Activities across Multiple Subjects

This module does xxx. Objective? Use case?

Work flow Add a use case (for example, if I want h-cluster, how to set params)

1. Select data….
2. …how to set input
3. What to expect from output and how to diagnose it

Detailed Document on Inputs and Outputs

Inputs

• Data files: xxx

Outputs

• Cluster visualization: Cluster center shape xxx

---

Inputs :

Id	label	Class	Description
analysis_data	Data files located in this project's RAVE directory
check_scale	Z-score data	checkbox (logic)	if selected, z-scoring would be applied on the signals within the selected time window across electrodes
input_groups	Condition Group	list	The condition groups that we would like to study, with elements as selected conditions and given list element names as condition group name
input_method	Clustering Method	character	name of cluster algorithms (ex. 'H-Clust', 'PAM')
time_window	Time Window	number	a 2-element vector to indicate the time window range, the analysis would only apply on the power within the time window
distance_method	Clustering Distance Measurement	character	methods to calculate the distance (dissimilarity) for the clustering (ex. 'euclidean', 'maximum',"manhattan", "canberra", "minkowski")
mds_distance_method	MDS Distance Measurement		methods to calculate the distance between points to generate the mds diagnosis (ex. 'euclidean', 'maximum',"manhattan","canberra")
op_run	Optimal Number of Clusters Analysis	checkbox	if selected, generate the plots of silhouette and SSE for different clustering methods to estimate the optimal number of clusters
do_run	Run Analysis		run the clustering analysis to generate all the plots and tables

Outputs :

Id	Label	Class	Description
cluster_plot	Cluster Visualization	graph	for each clusters, visualize the mean and point-wise standard deviation of the power against different conditions
mds_plot	MDS Diagnosis	graph	visualize the original high dimension data in two dimension space with MDS
cluster_membership_table	Clustering Membership	chart	tables shown which electrode from which subject is in which cluster
dendrogram_plot	Dendrogram	diagram	hierarchical clustering only; shows how the hierarchical arrangement happens among objects
optimal_cluster_number_plot	Optimal number of clusters	graph	visualize the silhouette and SSE for the different numbers of clustering, help determine the optimal numbers of clusters (a brief introduction of silhouette and SSE(elbow) method can be found here)