Airbnb_price_category_prediction

Problem Definition

  • Design a model to determine an appropriate listing price for a new listing on Airbnb. Instead of predicting the actual listing price using regression, the objective is to predict which pricing bin or range is appropriate for a new listing

  • Input: summary and image (7627) for a training dataset.

  • Output: price (7627).

Function

text preprocessing -> tokenization and vectorization each text ->training the model -> classification and prediction.

Challenges

We have a Uncleaned and untranslated Unbalanced data set so maybe will take some time to preprocess it and train it.

Model impact

Predicting the actual price class beginner, plus, premium and Type

The ideal solution

Using Dropout with Drop rate of 0.3