每日一练 | Data Scientist & Business Analyst & Leetcode 面试题 727

Data Application Lab 自2017年6月15日起，每天和你分享讨论一道数据科学(DS)和商业分析(BA) 领域常见的面试问题。

自2017年10月4日起，每天再为大家分享一道Leetcode 算法题。

DS Interview Question

How does the KNN algorithm work?

BA Interview Question

Customers Who Never Order

Suppose that a website contains two tables, the Customers table and the Orders table. Write a SQL query to find all customers who never order anything.

Table: Customers.

+----+-------+
| Id | Name  |
+----+-------+
| 1  | Joe   |
| 2  | Henry |
| 3  | Sam   |
| 4  | Max   |
+----+-------+

Table: Orders.

+----+------------+
| Id | CustomerId |
+----+------------+
| 1  | 3          |
| 2  | 1          |
+----+------------+

Using the above tables as example, return the following:

+-----------+
| Customers |
+-----------+
| Henry     |
| Max       |
+-----------+

LeetCode Question

Add Binary

Description:

Given two binary strings, return their sum (also a binary string).

Input: a = “11” b = “1”

Output: “100”

DS Interview Question & Answer

What are methods to make a predictive model robust to outliers?

• Use a model that is resistant to outliers.  Tree-based models are not as affected by outliers as regression models.  For statistical tests, choose non-parametric test instead of parametric test

• Use a more robust error metric.  For instance, use absolute mean difference instead of mean squared error to reduce the effect of outliers

• Winsorize the data.  Cap the data at a certain threshold

• Transform the data.  If the data has a pronounced right tail, use log transform

• Remove the outliers.  If there are very few of outliers and you are certain that they are anomalies not worth predictin

Reference:https://www.quora.com/What-are-methods-to-make-a-predictive-model-more-robust-to-outliers

BA Interview Question & Answer

Duplicate Emails

Write a SQL query to find all duplicate emails in a table named Person.

+----+---------+
| Id | Email   |
+----+---------+
| 1  | [email protected] |
| 2  | [email protected] |
| 3  | [email protected] |
+----+---------+

For example, your query should return the following for the above table:

+---------+
| Email   |
+---------+
| [email protected] |
+---------+

Note: All emails are in lowercase.

Answer:

select Email
from Person
group by Email
having count(Email) > 1;

Reference:

https://leetcode.com/problems/duplicate-emails/description/s/duplicate-emails/description/