Indian Real Estate - 99acres.com

Main Dataset

I scrapped data from 99acres using their (kind of) hidden API. I scrapped almost 10,000+ data using my scrapper app see here.

DESCRIPTION OF THE DATA

• Contains details of properties of Gurgaon, Hyderabad, Mumbai, Kolkata cities of India.
• All datasets of different cities contains almost 10K properties.
• In some datasets, some columns are not available. Sorry!!
• Target column: PRICE

Data Usage

This dataset can be used for various real estate-related tasks, including:

• Property price prediction.
• Market analysis to identify trends and patterns.
• Identifying popular property types and locations.
• Evaluating the impact of property attributes on price.

EXPLANATION OF EACH COLUMNS

NOTE: Not all the columns are important for you so first try to understand your problem statement and then filter this dataset accordingly.

• AGE: The age of the property in years.
• ALT_TAG: An alternative tag or description.
• AMENITIES: Describes the amenities available with the property.
• AREA: The area of the property.
• BALCONY_NUM: The number of balconies in the property.
• BATHROOM_NUM: The number of bathrooms in the property.
• BEDROOM_NUM: The number of bedrooms in the property.
• BROKERAGE: Information about the brokerage or agency associated with the property listing.
• BUILDING_ID: An integer identifier for the building.
• BUILDING_NAME: The name of the building.
• BUILTUP_SQFT: The total built-up area of the property in square feet.
• CARPET_SQFT: The total carpet area of the property in square feet.
• CITY_ID: An identifier for the city in which the property is located.
• CITY: The city where the property is located.
• CLASS_HEADING: A heading for the property class.
• CLASS_LABEL: A label representing the property class.
• CLASS: A classification label for the property.
• COMMON_FURNISHING_ATTRIBUTES: Attributes related to the furnishings and amenities commonly found in the property.
• CONTACT_COMPANY_NAME: The name of the company or agency responsible for the property listing.
• CONTACT_NAME: The name of the contact person associated with the property listing.
• DEALER_PHOTO_URL: URL to a photo or image associated with the property dealer.
• DESCRIPTION: A description of the property listing.
• EXPIRY_DATE: The date when the listing expires.
• FACING: Indicates the direction the property is facing.
• FEATURES: Describes the features of the property.
• FLOOR_NUM: The floor number of the property.
• FORMATTED_LANDMARK_DETAILS: Details of nearby landmarks.
• FORMATTED: Formatted information related to the property.
• FSL_Data: Data related to the property, possibly specific to a particular real estate agency.
• FURNISH: Indicates whether the property is furnished.
• FURNISHING_ATTRIBUTES: Attributes describing the level of furnishing in the property.
• GROUP_NAME: The name of the group or organization to which the property may belong.
• LISTING: Information about the property listing, possibly including its status and other details.
• LOCALITY_WO_CITY: The locality name without the city information.
• LOCALITY: The specific locality or neighborhood where the property is situated.
• location: Additional location information.
• MAP_DETAILS: Contains latitude and longitude information.
• MAX_AREA_SQFT: The maximum area of the property in square feet.
• MAX_PRICE: The maximum price of the property.
• MEDIUM_PHOTO_URL: URL to a medium-sized photo or image of the property.
• metadata: Additional metadata or information about the dataset.
• MIN_AREA_SQFT: The minimum area of the property in square feet.
• MIN_PRICE: The minimum price of the property.
• OWNTYPE: An integer representing the ownership type.
• PD_URL: URL to additional property details.
• PHOTO_URL: URL to photos or images associated with the property.
• POSTING_DATE: The date when the property listing was posted.
• PREFERENCE: Indicates the preference type for the property listing (e.g., "S" for sale).
• PRICE_PER_UNIT_AREA: The price per unit area of the property.
• PRICE_SQFT: The price per square foot of the property.
• PRICE: The price of the property. This is target column for ML.
• PRIMARY_TAGS: Primary tags or labels.
• PRODUCT_TYPE: The type of product listing.
• profile: Profile information related to the property or listing.
• PROJ_ID: An integer identifier for the project.
• PROP_DETAILS_URL: URL to detailed property information.
• PROP_HEADING: A heading or title for the property.
• PROP_ID: A unique identifier for each property listing, in the form of an object or string.
• PROP_NAME: The name of the property.
• PROPERTY_IMAGES: Images or photos related to the property.
• PROPERTY_NUMBER: An integer identifier for the property.
• PROPERTY_TYPE__U: An integer identifier for property type.
• PROPERTY_TYPE: The type of the property (e.g., "Residential Apartment").
• QUALITY_SCORE: A score or rating associated with the quality of the property.
• REGISTER_DATE__U: An additional registration date.
• REGISTER_DATE: The date when the property listing was registered.
• REGISTERED_DAYS: The number of days the property has been registered.
• RES_COM: Indicates if the property is residential or commercial.
• SECONDARY_AREA: Additional area information.
• SECONDARY_TAGS: Secondary tags or labels.
• SOCIETY_NAME: The name of the society or community.
• SPID: An integer identifier for each property listing.
• SUPER_SQFT: The total super area of the property in square feet.
• SUPERBUILTUP_SQFT: The total super built-up area of the property in square feet.
• THUMBNAIL_IMAGES: Thumbnail images related to the property.
• TOP_USPS: Top unique selling points of the property.
• TOTAL_FLOOR: The total number of floors in the building.
• TOTAL_LANDMARK_COUNT: The total number of landmarks near the property.
• TRANSACT_TYPE: The transaction type.
• UPDATE_DATE: An integer representing the update date.
• VALUE_LABEL: A label representing the property value.
• VERIFIED: Indicates if the property listing is verified.
• xid: An identifier or key associated with the property listing.

FACETS

I have uploaded FACETS data because they are being used to decode the encode columns present in the dataset. Columns like PREFERENCES, FACING, FURNISH, FEATURES, AMENITIES, OWNTYPE, etc. needs to be decoded to understand the dataset well enough.