# Structure Data Science Projects 

A few months back, I faced the challenge of structuring and maintaining consistency in my data science projects. After extensive research, I discovered the best practices for organizing project folders to enhance productivity and collaboration. Below is a Python script that automatically sets up a well-structured folder hierarchy tailored to different types of data science projects: machine learning (ML), data engineering (DE), and business intelligence (BI). This script not only saves time but also ensures that your projects remain organized and manageable.  

## Python Script  To use the script below, go to your `C:\\Users\\{Your Username}` (in my case, it was Admin, but you can change it inside the code) and create a `new_project.py` file. Add the code below inside this file.

## Usage Samples  

### Creating Machine Learning Project  ![Creating Machine Learning project](<https://example.com/path-to-your-image/structure-1.png>)   

### Folder Image: 1 

 ![Folder image: 1](<https://example.com/path-to-your-image/structure-2.png>) 

Structure Data Science Projects 

In today's digital age, understanding user feedback is critical to the continued development of any app. We analyzed Instagram comments for the M10 Digital Wallet app and identified user issues. Using the power of machine learning and generative artificial intelligence, we transformed raw data into useful results and visualized them with attractive and informative graphs.

##A Few Words About the Data
Each row of the dataset represents an Instagram comment. The dataset consists of 3 columns: timestamp, text, and pain_type. The dataset contains 800+ rows.
The values of the pain_type column are categorized as follows:

- None
- Security concerns
- User experience issues
- Regulatory compliance
- Technical glitches
- Transaction delays
- Cost of services
- Limited functionality
- Customer support issues
- Integration challenges
- Data privacy concerns
- Cashback problems
- Transparency concerns
- Lemon and others

##Analysis of User Reviews

I used a racing bar chart to visualize the data, which allows us to see which issues the user reviews complain about the most. The graph below shows the issues that users complain about the most in their reviews. The racing bar chart also shows the dynamic change of data over time, making it convenient for our analysis.

<iframe src="/m10-user-pain-type.html" width="120%" height="600" allowfullscreen="true" style="border: 0px;"></iframe>

##Results

The results from this analysis can be summarized as follows:

- The most complained about issues in the M10 Digital Wallet application are "technical glitches" and "user experience issues". These indicate that users are experiencing problems with the application.
- Additionally, issues such as "security concerns" and "data privacy concerns" are also prominent. From time to time, issues like phishing or informal sharing of data worry users.



This blog post describes the process of identifying and analyzing user issues by analyzing Instagram comments for the M10 Digital Wallet app.