Icon

02 Data Cleaning 1

02 Data Cleaning - Exercise
Task 1: Filter rows1. Read the population, life expectancy, and travel advisory data by executing the providedreader nodes2. Filter the population data to countries in Europe3. Filter the life expectancy data to countries/regions where the overall life expectancy is atleast 80 years4. Filter the travel advisory data to the safest countries (Level 1: Exercise normalprecautions) Task 2: Filter columns1. Read the life expectancy data from the SQLite database by executing the providedworkflow2. Exclude the Rank column- Manually- By including only string and double type columns Task 3: Manipulate strings1. Read the travel advisories data by executing the provided CSV Reader node2. Replace the colons in the Level column by hyphen like this: "Level 1- Exercise normalprecautions"3. Extract the risk levels as numbers into a separate column. You can use, for example, thesubstr() function.4. Complete the tasks above with the Column Expressions node Task 4: Evaluate mathematical expressions and apply rules1. Read the population data by executing the provided Excel Reader node2. Create a new column "Change" by subtracting the population values in 2018 from thepopulation values in 20193. Convert the absolute population change values into percentages: Divide them by the populationvalues in 2018. Multiply the result by 100.4. Create a categorical column with two values "increasing"/"decreasing" based on whether thechange is positive or negative Demographics.sqliteLifeExpectancydemographics.xlsx(population)demographics.xlsx(life_expectancy)demographics.xlsx(population)travel_advidories.csvtravel_advidories.csvNode 38Node 39Node 40Node 41Node 42 SQLite Connector DB Table Selector Excel Reader Excel Reader Excel Reader DB Reader CSV Reader CSV Reader Row Filter Row Filter Row Filter Column Filter String Manipulation Task 1: Filter rows1. Read the population, life expectancy, and travel advisory data by executing the providedreader nodes2. Filter the population data to countries in Europe3. Filter the life expectancy data to countries/regions where the overall life expectancy is atleast 80 years4. Filter the travel advisory data to the safest countries (Level 1: Exercise normalprecautions) Task 2: Filter columns1. Read the life expectancy data from the SQLite database by executing the providedworkflow2. Exclude the Rank column- Manually- By including only string and double type columns Task 3: Manipulate strings1. Read the travel advisories data by executing the provided CSV Reader node2. Replace the colons in the Level column by hyphen like this: "Level 1- Exercise normalprecautions"3. Extract the risk levels as numbers into a separate column. You can use, for example, thesubstr() function.4. Complete the tasks above with the Column Expressions node Task 4: Evaluate mathematical expressions and apply rules1. Read the population data by executing the provided Excel Reader node2. Create a new column "Change" by subtracting the population values in 2018 from thepopulation values in 20193. Convert the absolute population change values into percentages: Divide them by the populationvalues in 2018. Multiply the result by 100.4. Create a categorical column with two values "increasing"/"decreasing" based on whether thechange is positive or negative Demographics.sqliteLifeExpectancydemographics.xlsx(population)demographics.xlsx(life_expectancy)demographics.xlsx(population)travel_advidories.csvtravel_advidories.csvNode 38Node 39Node 40Node 41Node 42SQLite Connector DB Table Selector Excel Reader Excel Reader Excel Reader DB Reader CSV Reader CSV Reader Row Filter Row Filter Row Filter Column Filter String Manipulation

Nodes

Extensions

Links