Data Visualization Specialist Needed & visual analysis -- 2

ESSAY:

Read “Defining Insight for Visual Analytics” by Chang et al. and “Toward Measuring Visualization Insight” by North.

1. Summarize what the insights of visualization and visual analytics are based on the two papers.

2. The two papers were published almost 15 years ago. With the rapid growth of AI / machine learning techniques for supporting data analysis, are there any important changes on the insights of visualization and visual analytics? Make an argument for this. If there are, describe what the key changes are. If there are not, discuss your reasons. You should use examples or evidence of prior work from class or elsewhere (for which you should give the reference) to support your argument.

You can find the two papers in the attachment.

Development & Analysis:

Given the car dataset in the attachment, you need to do a visual analysis of clustering these cars.

1. You should pick at least four columns for clustering the cars. You are free to decide which columns to use and determine your rules for performing the clustering.

2. Using web-based techniques (e.g., D3.js) to visualize your clustering results.

a. Use a circle to represent each car.

b. Cars in the same cluster should be spatially placed near each other.

c. Use different colors for different clusters.

d. When mouse-hovering or selecting a car, you should show its detailed information from the given dataset.

3. You should answer the following questions, in addition to submitting your code. a. Which columns do you choose to perform the clustering?

b. What is your rule for deciding whether a car belongs to a cluster or not? c. How many cars are there in the largest cluster?

d. Do you find any car(s) falsely clustered? For example, it is more reasonable that car A should be in cluster 2, but it is displayed in cluster 1. If so, you need to explain why this happens.

Regarding to the development part, you should use good software engineering practices. Comment your code, use consistent formatting, use meaningful variable names, etc.

Machine Learning (ML) Análise estatística Research Writing Algoritmo Estatísticas

ID do Projeto: #37929165

Sobre o projeto