About our Visualisations

We have chosen 4 different techniques of visualisations where you can select each two of your liking as we aimed to present 16 different combinations in total to present maximum data information for you.

In a force-directed graph, every node represents a certain person of the company, and every link is an email which has been sent from one person to another person.
Repelling forces are applied to the nodes. Attractive forces are applied to attract the end points of the links. In order to keep the graph at the center, a center force is applied.
The colors of the nodes are randomly generated and correspond to the job title of a particular person. The color of the links is related to the sentiment value of a mail.
Since most of the time, there is a negligible difference in the sentiment values of different links, the color of the links is nearly the same, even though there is a range for the colors, going from green (lowest sentiment) to orange (highest sentiment).

Contrary to Force Graph, the nodes in the Bubble Chart attract each other. Plus, the Bubble Chart is clustered, nodes, representing people, appear close to each other if they belong to the same job title and are associated with the same coloring.
The size of the nodes, corresponds to the number of emails, which have been sent from a particular person, the bigger the node radius, the more emails sent from that person.
The interactivity supports the graph to be dynamic, by pulling a node out of its equilibrium state, one can observe it going back to its initial position. Furthermore, smaller/bigger radius nodes can be inspected by zooming in or out.

The Heat Map is implemented as a yearly calendar where months are shown on the x-axis and years on the y-axis.
Days are represented as thin colored rectangles, aligned with the other days in the same year horizontally and in the same month vertically. That orientation gives us a clear layout when comparing months of different years or days of the different months but the same year.
If there are no emails sent on a particular day, there is a blank space shown on the map. Every rectangle, aka. cell, has properties of date, sentiment, senders ID and email.
Moreover, the cell is colored regarding the sentiment of the email. Since sentiments of the emails are not equally distributed but rather concentrated around 0, we selected the main domain [-0.05,0.05] by analyzing the mode of sentiments in the data set.

The Chord Diagram has its main use which is to show flow of information (emails) between several entities, by depicting them in a radial format: the entities, in our case the people who work at the company, are situated around the circle. In order not to overwhelm the visualization, the particular emails of the people are omitted, and are generalized to job titles. Each job title is represented by a fragment, whose color is generated randomly. Then the arcs between different fragments show the exchange of emails from one fragment to another, where the color of the arc is the same as the color of the fragment, from where the email is sent. The thickness of the arc is proportional to the amount of emails sent from one fragment to another, in other words, from one job title to another.

Our visualisations interact between them using linking & brushing techniques. Any combination of two visualisations features a linking and brushing technique between them. For example, it is possible for a node to be selected in one visualisation and the corresponding elements will be highlighted in the other.