| **You can download the initial exercise file from [link](https://quera.org/contest/assignments/103149/download_problem_initial_project/356096/).** | | ------------------------------------------------------------------------------ | Welcome to "The Krabby Real Estate"! Mr. Krabs, the famous restaurant owner, has recently realized that the Krabby Patty formula is no longer the only way to get rich in the city of "Bikini Bottom". He has figured out that the real money lies in the volatile real estate market! But a massive crisis has struck the company! Plankton, using his computer wife's AI, is pre-purchasing the city's best houses with highly precise and engineered prices. Meanwhile, Mr. Krabs discovered that Squidward (who until yesterday was responsible for pricing and appraising the houses) was setting the prices entirely by chance, based on his mood, and just by looking at the exterior of the houses! Mr. Krabs fired Squidward out of pure anger and now senses the danger of his new empire going bankrupt. Now, Mr. Krabs wants one thing from you: build a system that predicts the exact price of houses in Bikini Bottom before Plankton gets his hands on them. SpongeBob's team (who now collects data across the city instead of flipping burgers) has extracted a dataset of sold houses in various neighborhoods and provided it to you. Your task is to design this model. The data collection team has provided the extracted information in the form of two files (train and test). Each row of this dataset represents the specifications of a house. The details of the columns in this dataset are as follows: | **Column Name** | **Description** | | --------------------------- | ---------------------------------------------------------------- | | **lat** | House latitude (map coordinates) | | **long** | House longitude (map coordinates) | | **area** | House area (in square meters) | | **age** | Building age (years) | | **total_floors** | Total number of building floors | | **floor_number** | Floor number where the unit is located | | **rooms** | Number of bedrooms | | **document_type** | Property deed type (Private, Endowment, Unregistered Deed, etc.) | | **cooling_type** | Cooling system (Split, Chiller, Water Cooler, etc.) | | **monthly_levy** | Monthly building maintenance fee | | **has_master_bedroom** | Does the house have a master bedroom? (1: Yes, 0: No) | | **parking_type** | Parking status and type (Covered, Courtyard, etc.) | | **neighbor_noise_level** | Neighbors' noise level (Low, Average, High) | | **distance_to_hospital_km** | Distance to the nearest hospital (kilometers) | | **has_smart_home** | Equipped with a smart home system (1: Yes, 0: No) | | **water_pressure** | Building water pressure status | | **exterior_style** | Building exterior style (Roman, Modern, Classic) | | **fiber_internet** | Access to fiber optic internet (1: Yes, 0: No) | | **manager_present** | Presence of a building manager or lobbyman (1: Yes, 0: No) | | **window_type** | Type of windows (Double-glazed, Single-glazed, etc.) | | **price** | Final sale price of the house | ### **Output** To evaluate your program, you must predict the prices of the houses in `test.csv`. Your output should be a text or CSV file containing only **one column** named `price`. Each row in this column must exactly represent the predicted price for the corresponding row in the input data. Code snippet ``` price 2450000 1860500 3400000 985000 ... ``` ### **Evaluation Method** The following formula is used to evaluate your system: $$Score = \max\left(0, 100 \times \left(1 - \frac{1 - R^2 Score}{0.04}\right)\right)$$