This project is connected to the Data Visualization course, but depending on your background knowledge of data visualization, dimple.js, and d3.js you may not need to take the whole course to complete this project.

After completing Lesson 2 and Problem Set 2 of the course, you will be able to complete this project since you will have learned about dimple.js.

If you want to become more technical and expand your skill set, you can continue to Lesson 3 and Lesson 4, in which you will learn more about narrative structures and how to create graphics using d3.js. The d3.js library has a steeper learning curve, and we encourage you to take on the challenge if you desire.

The project decide to use :

Because it can make the audience easily compare the value between the domain (x-axis) and quickly understand the trend of data.

Because it can make the audience quickly explore the interest location/state with highest/lowest value and understand the data in term of geometry (high/low value zone).

Because it can make the audience quickly compare the value between the categories and see how the data shape (distribution) so it can covert to how it significant different or it likely to be occurred by chance.

After I got the data I try to mark down what is the interesting story behide this and I found out that the most interesting variables are amount of loan and yield; I try to draft the visualization by using tableau public and found the co-related interesting variables such as Loan Original Date, Brrower Rating , Credit Grade, Prosper Rating and Borrower State.

draft file : draft.pdf (http://static.augimon.com/EDV-Laon/draft.pdf)

The data is too large so I need to process the data to optimize the size of resource that be used to create the visualization by create python script using numpy and pandas;

1st View Loan Original Date vs Sum(Loan Amount) vs ListingCategory - Show trend 2nd State vs vs Sum(LoanAmount) On date - Show Loan density 3nd View CreditGrade vs Borrower Rate - Show relation between Grade and Borrower Rate 4nd View ProsperRating vs Borrower Rate - Show relation between Grade and Borrower Rate

For US map view, it should have the state name and amount of loan when hover pointer on each state area.

For boxplot graph, it should have highlight number of borrower rate for each credit grade/ prosper rating. For example, grade A have borrowing rate as follow. 0.08: maximum 0.06 :q1 0.07: mediam 0.06:q3 0.05 :minimum

I was confused about what this data visualization want to introduce. The creator should include some brief to get audience understand his core concept. Lack of understanding greatly effects how audiences interpret each graph.

• While website loading data by each Q, the graph also keeps moving, make it confusing of what happens. Did creator want to make a point or just loading.

• Top left graph was easy to digest, loan amout by Q. However, I still don't get what creator want to say.

• I likes geography graph map. Easy to see different between each state. After saw different in color, I want to see their numbers too. Mouse hover to see exact number should improve this graph.

• Both right graphs were too hard for me to get any message.

1. When the page first loaded, it should display a graph with the final data (2013-Q4). It should not to start animation from the first data to the final data. It will take a long time to wait.

2. The display should be divided as follows:

a. Added the option to view each Quarter by selecting the drop-down (into the upper right corner of the page).

b. Segment graph in the display below (1) a graph showing data from 2005-Q3 data Quarter to the Quarter is selected and (2) a graph showing the data specific Quarter.

3. Add definition of Credit Grade (AA, A, B, C, D, E, HR and NC) and Prosper Rating (AA, A, B, C, D, E, HR and N / A) may be displayed in a small square on the graph (label description).

4. The map graph, it should show the state name and amount of loan when hover on each state.

5. The scale rate (colors) of the map graph should be fixed in every Quarter

-Graph title, to make user can clearly understand what it is and what duration that each graph include the data.

-US map view should improve by add tooltip when the user hover mouse on it.

-Add more option for quicker navigate though quater.

1.When the page first loaded, it should display a graph with the final data (2013-Q4). It should not to start animation from the first data to the final data. It will take a long time to wait..

I think it has already make sense to tell the story from the beginning of the data.

2. The scale rate (colors) of the map graph should be fixed in every Quarter.

I think this can make user can not distinguish visualization in case it has very low loan amount in some years.

1. There is a huge loan amount decrease which starts at 2008-Q3 from 20m to 3m and become 0 at 2008-Q4 to 2009-Q1,it is the same time of Subprime mortgage crisis ; After that it take 2 year to reach the same loan amount in 2011-Q2 and create new high in 2011-Q4 and start decrease at 2012-Q3 (On 6 September 2012, the ECB calmed financial markets by announcing free unlimited support for all eurozone countries ); After that 2 quarter the loan amount make new high and keep increasing till 2013-Q2.

2. California and Texas states have big loan amount through all period; From 2009-Q2 to 2012-Q1 New York's loan amount is larger than Texas's in some quarters.

3. There is no data for Credit Grade since 2009-Q2 but there is Prosper Rating instead since there.

4. Credit grade and Prosper rating has relation with borrower rate but Prosper rating can give you more precisely borrower rate (size of area in Borrower Rate V Prosper Rating 's boxplot much less than Borrower Rate V Credit Grade).