Featured Post

How to Check Column Nulls and Replace: Pandas

Image
Here is a post that shows how to count Nulls and replace them with the value you want in the Pandas Dataframe. We have explained the process in two steps - Counting and Replacing the Null values. Count null values (column-wise) in Pandas ## count null values column-wise null_counts = df.isnull(). sum() print(null_counts) ``` Output: ``` Column1    1 Column2    1 Column3    5 dtype: int64 ``` In the above code, we first create a sample Pandas DataFrame `df` with some null values. Then, we use the `isnull()` function to create a DataFrame of the same shape as `df`, where each element is a boolean value indicating whether that element is null or not. Finally, we use the `sum()` function to count the number of null values in each column of the resulting DataFrame. The output shows the count of null values column-wise. to count null values column-wise: ``` df.isnull().sum() ``` ##Code snippet to count null values row-wise: ``` df.isnull().sum(axis=1) ``` In the above code, `df` is the Panda

How to Use Chaid Useful for Data Science Developers

CHAID

The Chaid is one of the most asked skills for Data Science engineers. The CHAID Analysis (Chi-Square Automatic Interaction Detection) is a form of analysis that determines how variables best combine to explain the outcome in a given dependent variable.

Chaid Model


  • The model can be used in cases of market penetration, predicting and interpreting responses, or a multitude of other research problems.

  • CHAID analysis is especially useful for data expressing categorized values instead of continuous values.

  • For this kind of data, some common statistical tools such as regression are not applicable and CHAID analysis is a perfect tool to discover the relationship between variables. 

  • One of the outstanding advantages of CHAID analysis is that it can visualize the relationship between the target (dependent) variable and the related factors with a tree


1. CHAID Analysis for Surveys


Analysis

  • Most survey answers have categorized values instead of continuous values. 

  • Finding out the statistical relationship in this kind of data is a challenge. 



2. CHAID Analysis for Customer Profiling


Profiling


  • Based on historical customer data, CHAID Analysis can be used to analyze all characteristics within the file. 

  • For example, product/service purchased, the dollar amount spent, major demographics, and demography of the customers, and so on. 

  • A blueprint can be produced to provide an understanding of the customer profile: strong or weak sales of products/services; active or inactive customers; factors affecting customers’ decisions or preferences, and so on. 

  • Such a customer profile will give the Sales & Marketing Team a clear picture of which type of person is most likely to buy the products and services based on factual purchase history, geo-demographics, and lifestyle attributes.


3. CHAID Analysis for Customer Targeting


Customer Targetting


  • Recruiting new customers via direct contact (phone or mail) is a time-consuming and costly effort.

  • For most products or services, the hit rate is less than 1%. That means, in order to get a new customer, over one hundred contacts are required.

  • By mapping the current customer list to a general population database (e.g., SMR Residential Database that contains 12 million listed households), CHAID Analysis can find the household clusters that have much higher incidence rates than the average.

  • By concentrating on these household clusters, the actual hit rate can be dramatically raised. The result is “Fewer phone calls or mail pieces with higher sales returns!”.

Comments

Popular posts from this blog

Explained Ideal Structure of Python Class

How to Check Kafka Available Brokers

6 Python file Methods Real Usage