Click here to open Wish.com 🎉

Define the problem 🐱‍🚀

• The dataset is the wish.com product dataset. i collected the data combined with some available data. Some nosies are added to the dataset. The goal is to predict the product ratings given the other features known for a product on Wish.com. Ratings are in categories from 1 to 5. For one product, the higher the rating is, the more the customers like the product. In this way, when you have a new product to be put on wish.com, you can estimate how likely people will like your product, without actually listing out there. Also, by doing this, it helps us to understand under what certain conditions that a product will be highly rated, as a way to understand the customer base of the wish.com

---

Goal ✔🎁

• The goal is to predict product ratings based on other information available about a product on Wish.com. The rating range from 1 to 5. The higher the rating for a product, the more satisfied for the customers. When you have a new product to list on wish.com, you may use this model to predict how probable people will like it before actually listing it. Also , by doing so, we may better determine under what circumstances a product will be highly rated, as well as the wish.com consumer base.

what is the input 🤳

the input is the features of the product to predict the rating value but the model will select this features from the data input 🐱‍💻

Column	Description
price	Price for the buyer
retail_price	Retail price, or reference price in other stores/places. Used by the seller to indicate a regular value or the price before discount.
currency_buyer	Currency of the prices
units_sold	Number of units sold. Lower bound approximation by steps
uses_ad_boosts	Whether the seller paid to boost his product within the platform (highlighting, better placement or whatever).
rating	Mean product rating
rating_count	Total number of ratings of the product
badges_count	Number of badges the product or the seller have.
badge_local_product	A badge that denotes the product is a local product. Conditions may vary (being produced locally, or something else). Some people may prefer buying local products rather than. 1 means Yes, has the badge.
badge_product_quality	Badge awarded when many buyers consistently gave good evaluations. 1 means Yes, has the badge
badge_fast_shipping	Badge awarded when this product's order is consistently shipped rapidly
tags	Tags set by the seller
product_color	Product's main color
product_variation_size_id	One of the available size variations for this product
product_variation_inventory	Inventory the seller has. Max allowed quantity is 50
shipping_option_price	Shipping price
shipping_is_express	Whether the shipping is express or not. 1 for True
countries_shipped_to	Number of countries this product is shipped to. Sellers may choose to limit where they ship a product to
inventory_total	Total inventory for all the product's variations (size/color variations for instance)
has_urgency_banner	Whether there was an urgency banner with an urgency
merchant_rating	Merchant's rating
Merchant Has Profile Picture	Indicates whether the merchant selling the product has a profile picture on the platform

⚠ _ Note: Not all the columns are present in the above description🤢

What is the output? 🎂

:the out put is the prediction rating for the input product

---

evaluation metric 🎶

The evaluation metric for this competition is Mean F1-Score. The F1 score, commonly used in information retrieval, measures accuracy using the statistics precision p and recall r. Precision is the ratio of true positives (tp) to all predicted positives (tp + fp). Recall is the ratio of true positives to all actual positives (tp + fn).

The F1 metric weights recall and precision equally, and a good retrieval algorithm will maximize both precision and recall simultaneously. Thus, moderately good performance on both will be favored over extremely good performance on one and poor performance on the other

Just want to quickly look at some notebooks, without executing any code?

• 

• Click here to open in Google Colab