Network Mapping

Analysis is the principal tool for mapping the structure and the dynamics of your corpus.

Earth Negociation Bulletins have been providing a consistent report of discussions held at the different conferences of the parties for nearly 20 years. The semantic map resulting from the analysis of such a corpus feature the different issue at stake in climate change negociations.

This script produces several types of analysis and visualizations which are illustrated above. The maps feature homogeneous or heterogeneous nodes (see nodes selection page) which can be linked according to different types of proximity measures (see edges definition page) . Different advanced option (see Network Analysis & Layout page) are also proposed for tagging clusters, producing historical maps, generating “heatmaps” or contingency matrices. Last tubes provide a dynamical perception of maps transformation in time (see Dynamical Options section for more information). A web interface (see illustration below) also allows to browse maps and edit node and cluster information. It is based on TinaJS explorer developed at ISCPIF. Note that only homogeneous maps are visible with the web interface. Cluster tags and heatmaps are also only visible in the pdf  version of the maps.

Hashtag cooccurrence network in a Russian collection of Instagram photos
Hashtag cooccurrence network in a Russian collection of Instagram photos

The video below illustrates the use of the script –  network mapping, adding tags, and creating a river network (tubes) :

Node selection

Node selection

This page describes the main options available under the node panel of the mapping script Heterogeneous fields The underlying rationale behind this analysis is to use a very systematic and symmetric perspective that allow users to produce heterogeneous networks featuring any couple of nodes types. One can mix any two fields: for example from an ...
Edges Definition

Edges Definition

This page describes the parameters regarding the edges of the maps in the mapping script Proximity measure Once you have defined the fields of inquiry along with the correct time periods, co-occurrence networks should be transformed according to a given proximity measure. One can choose co-occurrences based direct (chi2, mutual information, raw, cramer) or indirect measure (cosine, distributional). ...
Edges: metrics definitions

Edges: metrics definitions

The Distributional l-l r proximity measure is similar than the classic Distributional, with a small distinction: Distributional: as shown above, the cooccurrences matrix is based on the Mutual Information between nodes. Distributional l-l r: the cooccurrences matrix is based on the Log-Likelihood Ratio between nodes. The two share the same characteristics: the nodes (e.g. keywords) ...
Dynamical Settings

Dynamical Settings

This page describes the principal options under the dynamics panel in the mapping script Defining time periods Heterogeneous mapping script also enables users to divide a corpus according to their need. One can work with custom initial time ranges (as defined by period slicer script) or with standard periods (typically a yearly defined corpus). You ...
Network Analysis & Layout

Network Analysis & Layout

Describe options under the Network Analysis and Layout panel in the mapping script   The spatialization performed by CorText follows the classical Fruchterman Reingold layout. However the traditional force directed algorithm is tuned in such a way that a gravity force attracts every node toward the centroid of the cluster they  belong to. Clusters Detection ...
Gallery

Gallery

Mapping election night voices Hashtag cooccurrence network of tweets about Trump (collected thanks to Twitter search API and using the simple query “trump”) during the election night. A second map shows hashtags that have been converted from their original casings into lower-cased featuring. Colored sets of hashtags illustrate the various kinds of concerns raised by Twitter users coming from ...