apheris-preprocessing

Your task is to develop a simple yet powerful data preprocessing tool using Python, specifically leveraging the 'apheris-preprocessing' package. This tool will serve as an essential part of any data scientist's toolkit, enabling them to quickly clean and prepare their datasets for analysis or machine learning tasks. The application should be able to handle various types of data inputs, including CSV files, and perform common preprocessing steps such as handling missing values, scaling numerical features, encoding categorical variables, and applying basic feature selection techniques.

### Core Requirements:
1. **Data Importation**: Implement functionality to import datasets from CSV files. Users should be able to specify the file path and delimiter if necessary.
2. **Handling Missing Values**: Provide options to either fill missing values with mean/median/mode or drop rows/columns containing missing data.
3. **Scaling Numerical Features**: Offer methods to scale numerical features using standardization or normalization techniques.
4. **Encoding Categorical Variables**: Allow users to choose between one-hot encoding, label encoding, or other suitable methods for categorical data preprocessing.
5. **Feature Selection**: Integrate basic feature selection methods like correlation-based selection or mutual information based on user preference.
6. **Output**: Enable users to save the preprocessed dataset back into a CSV file or another format of their choice.

### Additional Features (Optional):
- **Visualization**: Include plots showing distributions before and after preprocessing steps.
- **User Interface**: Develop a simple GUI using Tkinter or similar library to make the tool more accessible.
- **Custom Scripts**: Allow users to write custom scripts for additional preprocessing steps.
- **Documentation**: Ensure thorough documentation explaining each step and how to use the tool effectively.

### Utilizing 'apheris-preprocessing':
- Explore the 'apheris-preprocessing' package to identify which functions align with the above requirements. For instance, you might find functions for handling missing values, scaling, encoding, and feature selection that can streamline your implementation process.
- Consider how you can integrate these functions into your application flow, ensuring they enhance usability without compromising performance.
- Document how each function from 'apheris-preprocessing' is utilized within your code, providing examples where possible.