Welcome to my project on the Mercari Price Suggestion Challenge! This project showcases my exploration of data and application of machine learning techniques to solve the pricing prediction problem.

In this project, I tackled the challenge of predicting product prices for Mercari, Japan’s leading community-powered shopping app. With the vast array of products listed on Mercari's marketplace, accurately determining prices is essential. Leveraging exploratory data analysis (EDA) and machine learning (ML), I aimed to develop an algorithm that suggests appropriate prices based on user-provided text descriptions, including details like product category name, brand name, and item condition.

My objective was to delve into the dataset, uncover insights through EDA, and build predictive models using ML techniques. By understanding the relationships between features and prices, I aimed to create a robust pricing prediction system.

Given user-provided text descriptions of products, including details like category name, brand name, and item condition, predict the sale price of the listing.

This competition involves predicting the sale price of listings based on user-provided information. The dataset includes train.tsv and test.tsv files, which consist of a list of product listings in tab-delimited format. The data fields include:

• train_id or test_id: the ID of the listing
• name: the title of the listing
• item_condition_id: the condition of the items provided by the seller
• category_name: category of the listing
• brand_name
• price: the price that the item was sold for (target variable)
• shipping: 1 if shipping fee is paid by the seller and 0 by the buyer
• item_description: the full description of the item

Note: The data has been preprocessed to remove text resembling prices to avoid leakage.

• train_id or test_id: ID of the listing
• name: Title of the listing
• item_condition_id: Condition of the items provided by the seller
• category_name: Category of the listing
• brand_name
• price: Price that the item was sold for (target variable)
• shipping: 1 if shipping fee is paid by the seller and 0 by the buyer
• item_description: Full description of the item

I conducted extensive EDA to explore the dataset, visualize distributions, analyze correlations, and identify factors influencing product prices. Through visualizations and statistical analysis, I gained valuable insights into the dataset's characteristics.

Pending Task Using ML algorithms, I trained predictive models to estimate product prices based on the provided dataset. I experimented with various models, performed feature engineering, and optimized model performance to achieve accurate pricing predictions.

This README serves as a showcase of my work on the Mercari Price Suggestion Challenge. It highlights my expertise in EDA and ML, showcasing the methodologies employed, insights gained, and the predictive models developed.

I will continue to refine my models, explore advanced ML techniques, and enhance the accuracy of pricing predictions. Additionally, I am open to feedback and collaboration opportunities to further improve my work.

Thank you for taking the time to explore my project on the Mercari Price Suggestion Challenge!