-
Notifications
You must be signed in to change notification settings - Fork 1
Expand file tree
/
Copy paththinkful2.1
More file actions
24 lines (18 loc) · 1.53 KB
/
Copy paththinkful2.1
File metadata and controls
24 lines (18 loc) · 1.53 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Assignment
1. What is the goal of EDA (exploratory data analysis)?
The goal of EDA is to explore relationships within the data and find meaningful value of the relationships.
2. Suppose that you are given a dataset of customer product reviews for an e-commerce company.
Each review is scored as a Likert-style survey item where 1 indicates a negative sentiment about the product and a 5 is
positive. These reviews are collected on the company's website.
a. What problems do you expect to find in the raw data?
From any raw data, I'd expect to find miscellaneous characters, missing values, NaN values. I'd expect some values might be in
incorrect date formats or formats that are not readable.
b. If your task is to build features that give information about customer sentiments,
how would you approach this task and what kind of methods would you apply to accomplish it?
I'd request for feedback after sale/purchase. I'd most likely ask questions such as relating to customer service satisfaction and
ease of shopping/use of website. We can use a likert scale but it would aslo be ideal to get opinions, so for the survey, I'd leave a space
for comments/questions/concerns.
c. Try to identify some potentially useful features that you might derive from the raw data.
How would you derive them and how would you assess the usefulness of those features?
Raw data can give us a lot of useful information. We can use data visualizations and stastical analysis to determine the relationships
of the variables and from there come up with a conclusion.