Group 3 Augmented Data Preparation by Amy Goh https://padlet.com/arielmist/pei8k6w834hw3j8u Data Preparation Steps en-us 2020-12-01 11:54:03 UTC 2020-12-09 05:36:44 UTC hello@padlet.com Project Description arielmist https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184957 Problem: Customer churn is one of the challenges in the banking industry. A bank is facing churn from its credit card customers and the bank manager would like to identify potential customers that are likely to leave the bank. ]]> 2020-12-01 11:54:03 UTC https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184957 Access arielmist https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184960 We extracted the dataset (BankChurners.csv) from Kaggle (source URL: https://www.kaggle.com/sakshigoyal7/credit-card-customers).

There are in total 23 columns and 10128 rows. We will be analysing the following columns:
1. Customer_Age
2. Income_Category
3. Months_on_book
4. Months_Inactive_12_mon
5. Contacts_Count_12_mon
6. Credit_Limit
7. Total_Revolving_Bal
8. Total_Amt_Chng_Q4_Q1
9. Total_Trans_Ct
10. Total_Trans_Amt
11. Card_Category]]> 2020-12-01 11:54:03 UTC https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184960 Data Profiling arielmist https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184962 Single column profiling was used to analyze the values in each column and identify the appropriate data type. 
 
- identify data type
- identify missing values
- summary statistics (mean, median, interquartile range, minimum and maximum)

]]> 2020-12-01 11:54:03 UTC https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184962 Clean/Transform arielmist https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184963 We have performed the following data cleaning or transformation tasks:

- delete and\or input missing data
- normalization
- pivot
- encoding (e.g. one hot encoding)
- identify outliers
- identify inconsistent data]]> 2020-12-01 11:54:03 UTC https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184963 Publish arielmist https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184965 File format *.CSV (Comma-separated values) are accepted into many different types of databases and applications like TIBCO and Microsoft Excel. Therefore, we had decided to export the data in *.CSV format. 

Refer to steps detailed in the document attached. ]]> 2020-12-01 11:54:03 UTC https://padlet.com/arielmist/pei8k6w834hw3j8u/wish/975184965