VI_7

Name: Erika Siregar

Assignment : Visualization Implementation (VI7)

Course: Information Visualization (CS725)

Semester : Spring 2016

Node-link diagram and a matrix view of the adjacency between the contiguous US states and DC
Tutorial from:
- Node-link diagram: Force-Directed Graph
- Adjacent Matrix : Les Misérables Co-occurrence
- Data : contiguous-usa.dat

Node-link Diagram

*I grouped the data based on their GMT regions. There are 4 regions being displayed: GMT+5 (blue color), GMT+6 (red color), GMT+7 (green color), and GMT+8 (orange color).
*I encode DC with bigger size of node to make it more obvious to users. So, users can easily compare or measure the distance between DC and other states.
*From the node link we can easily see the states that directly adjacent to each other. The adjacent states are connected with a link. The interesting thing to be considered here is since I grouped the nodes by region, we can directly see which states that connect to each other but have different GMT regions. For example: Indiana (IN) is located in GMT+8 region (orange color), but it is neighboring with Illinois (IL) which is located in GMT+7 (green color).
*Another interesting point is users can measure the distance between one state to another by counting the number of links that separate them. For example, there are two links between Oregon and Wyoming through Idaho. It means Oregon is located 2 states away from Wyoming, if we go through Idaho. On the other hand, Oregon and Wyoming are located 3 states away from each other if we go through Nevada and Utah, since there are 3 links that separate them.

Matrix view

*I found the matrix view is more difficult to comprehend compare to node-link diagram. I provide a dropdown list where user can choose how they like to order the states: either by name, number of connections, or time region.
Here are some insights that I could gain from the adjacent matrix:

we could identify adjacent states by looking at the colored cell. For example, Kentucky and Tennesse are adjacent because if we draw a line from the row TN to column KY, they will meet at one point (cell).
we could identify which states have the maximum connection and which states have the minimum connection by ordering them by number of connections.
We can tell the distant between two states by looking at the opacity of the cell color.

<!DOCTYPE html>
<meta charset="utf-8">
<style>
  svg {
    font: 10px sans-serif;
	}
	.node {
    stroke: #fff;
    stroke-width: 1.5px;
  }

.node-active{
    stroke: red;
    stroke-width: 1.5px;
  }
  
  .node-inactive{
    opacity: .3;
  }

.link {
    stroke: #555;
    stroke-opacity: .7;
  }

.link-active {
    stroke: red;
    stroke-width: 1.5px;
    stroke-opacity: 1;
  }
  
  .link-inactive{
    opacity: .3;
  }

.overlay {
    fill: none;
    pointer-events: all;
  }
  
  .tooltip {
    position: absolute;
    display: none;
    pointer-events: none;
    background: #fff;
    padding: 5px;
    text-align: left;
    border: solid #ccc 1px;
    color: #666;
  }
  
  .background {
    fill: #eee;
  }
  line {
    stroke: #fff;
  }
  text.active {
    fill: red;
  }
</style>
<body>
  <div style="overflow-y: scroll; height: 500px">
  	<div id="vis1"></div>
  	<div id="vis2">
    	<p>Order: <select id="order">
          <option value="name">State Name</option>
          <option value="count">Number of Connection</option>
        	<option value="group">Timezone</option>
      </select>
    </div>
  </div>
  
<script src="//d3js.org/d3.v3.min.js"></script>
<script>

var width = 860,
    height = 500;

var color = d3.scale.category10();

var force = d3.layout.force()
    .charge(-120)
    .linkDistance(30)
    .size([width, height]);
  
var tooltip = d3.select("body").append("div")
		.attr("class", "tooltip")

var svg = d3.select("#vis1").append("svg")
    .attr("width", width)
    .attr("height", height);

d3.json("contiguous-usa.json", function(error, graph) {
  if (error) throw error;
  
  var zoom = d3.behavior.zoom()
    .scaleExtent([1, 10])
    .on("zoom", zoomed);
  
  var drag = d3.behavior.drag()
    .origin(function(d) { return d; })
    .on("dragstart", dragstarted)
    .on("drag", dragged)
    .on("dragend", dragended);
  
  force
    .nodes(graph.nodes)
    .links(graph.links)
    .start();
  
  var container = svg.append("g");
  var link = container.append("g")
    .attr("class", "links")
    .selectAll(".link")
    .data(graph.links)
    .enter().append("line")
      .attr("class", "link")
      .style("stroke-width", function(d) { return Math.sqrt(d.value); });
  
  var node = container.append("g")
    .attr("class", "nodes")
    .selectAll(".node")
    .data(graph.nodes)
    .enter().append("circle") 
      .attr("class", "node")
      .attr("r", function(d) { 
        if(d.id === "DC") {
          return d.value * 2 + 20; 
        } else {
          return d.value * 2 + 12; 
        }
      })
			.style("fill", function(d) { return color(d.group); })
  		.style("stroke", function(d) { if(d.id === "DC") return "red" })
  		.style("stroke-width", function(d) { if(d.id === "DC") return 5 })
      .call(drag);
  
  var linkedByIndex = {};
  graph.links.forEach(function(d) {
    linkedByIndex[d.source.index + "," + d.target.index] = 1;
  });

link.classed("link-active", function(o) {
      return o.source === d || o.target === d ? true : false;
    });
    
    link.classed("link-inactive", function(o) {
      return o.source !== d && o.target !== d ? true : false;
    });
    
    // show tooltip
    tooltip.text(d.id)
      .style("left", (d3.event.pageX + 7) + "px")
      .style("top", (d3.event.pageY - 15) + "px")
      .style("display", "block")
      .style("opacity", 1);
  })
		
	.on("mouseout", function(d){
    node.classed("node-active", false);
    node.classed("node-inactive", false);
    link.classed("link-active", false);
    link.classed("link-inactive", false);
    
    // hide tooltip
    tooltip.style("opacity", 0)
      .style("display", "none");
  });
  
  function zoomed() {
    container.attr("transform", "translate(" + d3.event.translate + ")scale(" + d3.event.scale + ")");
  }
  
  function dragstarted(d) {
    d3.event.sourceEvent.stopPropagation();
    force.start();
  }

function dragged(d) {
    d3.select(this).attr("cx", d.x = d3.event.x).attr("cy", d.y = d3.event.y);
  }

</script>

https://d3js.org/d3.v3.min.js