Customer Behavior Prediction, Image Preprocessing, Text Preprocessing and Product Recommendation System.

1. Classification Prediction:

We do have a E-commerce dataset, we need to identify or predict who are all coming as visitors and converted as out customers. Use the given data, there is a column called “has_converted” as target variable. Do the classification to find whether user will convert as a customer or not. In the classification prediction model, we aim to analyze customer behavior using the following algorithms: Decision Tree, Logistic Regression, and Random Forest.

[DataSet:] (https://drive.google.com/drive/folders/1ATULlRKrSensZHs2SxaT7y0b68Rc1vQA)

Steps to follow:

1. Data read
2. Do Multiple EDA with Plots (Show In streamlit)
3. Preprocessing
4. Stat Analysis
5. Feature Selection
6. Model Building – Build Atleast 3 models.
7. Do predictions for the Live stream Data.
8. Display the 3 model’s precision, recall, accuracy, f1-score

Image Processing:

In this module, we process images using techniques such as EasyOCR (Optical Character Recognition) to extract text from images, and the Python Imaging Library (PIL) to identify and extract objects from images. Additionally, PIL can be used to modify images by changing formats, rotating, and manipulating pixel sizes.

Text Processing:

In this module, we provide sentiment analysis for text based on user input, utilizing text processing techniques such as NLTK (Natural Language Toolkit).

Product Recommendation System:

Build a recommendation system for product selection using NLTK techniques.

Project Demo Link: [Demo] ()

Conclusion

The "Customer Insights and Recommendation System" is a comprehensive project that employs advanced techniques in classification prediction, image processing, and text analysis to gain a deep understanding of customer behavior. By integrating Decision Tree, Logistic Regression, and Random Forest models, along with image processing tools like EasyOCR and Python Imaging Library, and sentiment analysis using NLTK, the system provides a holistic approach to customer data analysis. The product recommendation system further enhances user experience by offering personalized suggestions based on individual behavior and preferences. Make sure to install the specified dependencies before running the code to ensure seamless functionality.

Contributing

Contributions to this project are welcome! If you encounter any issues or have suggestions for improvements, please feel free to submit a pull request.

Contact:

📧 Email: kaleeswariramkumar25@gmail.com

🌐 LinkedIn: linkedin.com/in/kaleeswari-s

For any further questions or inquiries, feel free to reach out. We are happy to assist you with any queries.