Logo Lanfrica

Jen-Ashishali/SCAMP-C6-Assessment

Créateur:
Jen
Hôte:
A technical assessment to meet the application requirements for She Code Africa mentoring program. # SCAMP-C6-Assessment # ASK ## Task Study the details of the dataset and find out which industry has the highest number of fake job postings ### Deliverables - Graph representing industry with highest number of fake job postings - Summary of Analysis - Details of Data Cleaning # PREPARE Dataset is made available by She Code Africa and has been downloaded and stored in Excel files for Analysis. # PROCESS ## Clean the dataset 1. Remove Duplicates - Select the unique identifier column “job_id” - Click Home > Conditional Formatting > Highlight Cells Rules > Duplicate Values - Select “job_id” - Click Home > Sort & Filter > Filter - Click the filter slice on “job_id” to filter by color Result: There are no highlighted cells. That means no duplicate entries. 2. Remove Irrelevant Data Fake job postings = Fraudulent job postings - Add a filter to “fraudulent” - Assuming 0 is False and 1 is True, select 1 - Copy and paste filtered data to new worksheet. 3. Structural Errors - Add a filter to “title”. Notice the inconsistent capitalization and job titles attached with salary details. - In “title” column, convert observations to proper case using =PROPER() function - Extract salary details found in title and fill in corresponding “salary_range” column - Create a column “country”, extract the country name from each ISO code in “location” - Replace values in “location” with values in “country” - Format “salary_range” to currency as Dollar - In “company_profile”, fix all imported special characters using Find & Replace - Use =TRIM() function to remove trailing spaces from “company_profile” - Add a filter to “industry”. Notice “industry” has similar observations with different naming conventions - Update observations to singular naming conventions and correct typos - Industry should be based on “company_profile”, but some industries aren’t matching company profiles - Review each industry and ensure they are matching information on “company_profile” 4. Missing Data - Check f …