Context

This dataset contains Real Estate listings in the US broken by State and zip code.

Download

kaggle API Command !kaggle datasets download -d ahmedshahriarsakib/usa-real-estate-dataset

Content

The dataset has 1 CSV file with 10 columns -

1. realtor-data.csv (2,226,382 entries)
   • brokered by (categorically encoded agency/broker)
   • status (Housing status - a. ready for sale or b. ready to build)
   • price (Housing price, it is either the current listing price or recently sold price if the house is sold recently)
   • bed (# of beds)
   • bath (# of bathrooms)
   • acre_lot (Property / Land size in acres)
   • street (categorically encoded street address)
   • city (city name)
   • state (state name)
   • zip_code (postal code of the area)
   • house_size (house area/size/living space in square feet)
   • prev_sold_date (Previously sold date)

NB: 1. brokered by and street addresses were categorically encoded due to data privacy policy 2. acre_lot means the total land area, and house_size denotes the living space/building area

Acknowledgements

Data was collected from - - https://www.realtor.com/ - A real estate listing website operated by the News Corp subsidiary Move, Inc. and based in Santa Clara, California. It is the second most visited real estate listing website in the United States as of 2024, with over 100 million monthly active users.

Cover Image

Image by Mohamed Hassan from Pixabay

Disclaimer

The data and information in the data set provided here are intended to use for educational purposes only. I do not own any data, and all rights are reserved to the respective owners.

Inspiration

• Can we predict housing prices based on the features?
• How are housing price and location attributes correlated?
• What is the overall picture of the USA housing prices w.r.t. locations?
• Do house attributes (bedroom, bathroom count) strongly correlate with the price? Are there any hidden patterns?

Overview

This residential real estate data set was created by Redfin, an online real estate brokerage. Published on January 9th, 2022, this data summarize the monthly housing market for every State, Metro, and Zip code in the US from 2012 to 2021. Redfin aggregated this data across multiple listing services and has been gracious enough to include property type in their reporting. Please properly cite and link to RedFin if you end up using this data for your research or project.

Source: RedFin Data Center

Property Type

Property type defined by RedFin

• All Residential: All properties defined as single-family, condominium, co-operative, townhouses, and multi-family (2-4 units) homes with a county record.
• Single Family Home (SFH): are homes built on a single lot, with no shared walls. Sometimes there’s a garage, attached or detached.
• Condominium (Condo): Usually a single unit within a larger building or community. Generally come with homeowners’ associations (HOAs), which require the residents to pay monthly or yearly dues.
• Cooperatives (Co-op): Usually a single unit within a larger building or community, but with a different way of holding a title to a shared building. You join a community and everyone in the community owns the building together.
• Townhouse: a hybrid between a condo and a single-family home. They are often multiple floors, with one or two shared walls, and some have a small yard space or rooftop deck. They’re generally larger than a condo, but smaller than a single-family home.
• Multifamily (2-4 units): They are essentially a home that has been turned into two or more units but the units cannot be purchased individually. There is one owner for the whole building.
• Land: Just land, no home of any type for sale.

Source: Building Types

Property Type

For more definitions, please visit RedFin Data Center Metrics

• Average sale to list: The mean ratio of each home's sale price divided by their list price covering all homes with a sale date during a given time period. Excludes properties with a sale price of 50%.
• Home sales: Total number of homes with a sale date during a given time period.
• Inventory: Total number of active listings on the last day of a given time period.
• Median active list ppsf: The median list price per square foot of all active listings.
• Median active list price: The median list price of all active listings.
• Median active listings: The median of how many listings were active on each day within a given time period.
• Median days on market: The number of days between the date the home was listed for sale and when the home went off-market/pending sale covering all homes with an off-market date during a given time period where 50% of the off-market homes sat longer on the market and 50% went off the market faster. Excludes homes that sat on the market for more than 1 year.
• Median days to close: The median number of days a home takes to go from pending to sold.
• Median list price: The most recent listing price covering all homes with a listing date during a given time period where 50% of the active listings were above this price and 50% were below this price.
• Median list price per square foot: The most recent listing price divided by the total square feet of the property (not the lot) covering all homes with a listing date during a given time period where 50% of the active listings were above this price per sqft and 50% were below this price per sqft.
• Median listing with price drops: The median of how many listings were active on each day and whose current list price is less than the original list price within a given time period.
• Median sale price: The final home sale price covering all homes with a sale date during a given time period where 50% of the sales were above this price and 50% were below this price.
• Median sale price per square foot: The final home sale price divided by the total square feet of the property (not the lot) covering all homes with a sale date during a given time period where 50% of the sales were above this price per sqft and 50% were below this price per sqft.
• Months of supply: When data are monthly, it is inventory divided by home sales. This tells you how long it would take supply to be bought up if no new homes came on the market.
• New listings: Total number of homes with a listing added date during a given time period.
• Off market in two weeks: The total number of homes that went under contract within two weeks of their listing date.
• Pending home sales: Total homes that went under contract during the period. Excludes homes that were on the market longer than 90 ...

Our extensive database contains approximately 800,000 active rental property listings from across the United States. Updated daily, this comprehensive collection provides real estate professionals, investors, and property managers with valuable market intelligence and business opportunities. Database Contents

Property Addresses: Complete location data including street address, city, state, ZIP code Listing Dates: Original listing date and most recent update date Availability Status: Currently available, pending, or recently rented properties Geographic Coverage: Properties spanning all 50 states and major metropolitan areas

Applications & Uses

Market Analysis: Track rental pricing trends across different regions and property types Investment Research: Identify high-opportunity markets with favorable rental conditions Lead Generation: Connect with property owners potentially needing management services Competitive Intelligence: Monitor listing volumes, vacancy rates, and market saturation Business Development: Target specific neighborhoods or property categories for expansion

File Format & Delivery

Organized in easy-to-use CSV format for seamless integration with data analysis tools Accessible through secure download portal or API connection Daily updates ensure you're working with the most current market information Custom filtering options available to narrow results by location, date range, or other criteria

Data Quality

Rigorous validation processes to ensure address accuracy Duplicate listing detection and removal Regular verification of active status Standardized format for consistent analysis

Subscription Benefits

Access to historical listing archives for trend analysis Advanced search capabilities to target specific property characteristics Regular market reports summarizing key trends and opportunities Custom data exports tailored to your specific business needs

AK ~ 1,342 listings AL ~ 6,636 listings AR ~ 4,024 listings AZ ~ 25,782 listings CA ~ 102,833 listings CO ~ 14,333 listings CT ~ 10,515 listings DC ~ 1,988 listings DE ~ 1,528 listings FL ~ 152,258 listings GA ~ 28,248 listings HI ~ 3,447 listings IA ~ 4,557 listings ID ~ 3,426 listings IL ~ 42,642 listings IN ~ 8,634 listings KS ~ 3,263 listings KY ~ 5,166 listings LA ~ 11,522 listings MA ~ 53,624 listings MD ~ 12,124 listings ME ~ 1,754 listings MI ~ 12,040 listings MN ~ 7,242 listings MO ~ 10,766 listings MS ~ 2,633 listings MT ~ 1,953 listings NC ~ 22,708 listings ND ~ 1,268 listings NE ~ 1,847 listings NH ~ 2,672 listings NJ ~ 31,286 listings NM ~ 2,084 listings NV ~ 13,111 listings NY ~ 94,790 listings OH ~ 15,843 listings OK ~ 5,676 listings OR ~ 8,086 listings PA ~ 37,701 listings RI ~ 4,345 listings SC ~ 8,018 listings SD ~ 1,018 listings TN ~ 15,983 listings TX ~ 132,620 listings UT ~ 3,798 listings VA ~ 14,087 listings VT ~ 946 listings WA ~ 15,039 listings WI ~ 7,393 listings WV ~ 1,681 listings WY ~ 730 listings

Grand Total ~ 977,010 listings

Description This dataset contains detailed real estate listings from the Chittagong Division, Bangladesh, collected in August 2025. It includes key property attributes such as location, price, size, number of bedrooms, number of bathrooms, and additional features. The data is clean, structured, and ready for analysis, making it ideal for machine learning, market trend analysis, and investment research.

Dataset Highlights 1. Region: Chittagong Division, Bangladesh 2. Date Range: August 2025 3. Data Type: Tabular (XLSX format)

Fields Included:

• Property Location (city, area)
• Price (in BDT)
• Area (sq. ft.)
• Property Type (apartment, house, commercial)
• Bedrooms & Bathrooms count
• Additional property features

Possible Use Cases

1. Price Prediction Models: Build regression or machine learning models to forecast property values.
2. Market Trend Analysis: Identify emerging real estate trends in Chittagong.
3. Geospatial Insights: Map property distribution and pricing by location. 4.Comparative Studies: Compare Chittagong’s market with other regions in Bangladesh.

Why This Dataset is Valuable The Bangladeshi real estate market is rapidly growing, and the Chittagong Division is one of its most active hubs. Having structured, up-to-date, and region-specific property data enables analysts, developers, and researchers to make data-driven decisions with confidence.

Column Descriptions sku – Unique identifier for each property listing in the dataset.

price_value – Total listed price of the property (in Bangladeshi Taka).

category – Main property category, such as residential, commercial, or land.

subcategories – Specific property type within the main category (e.g., apartment, house, shop, plot).

floor_area_sqft – Floor area of the property in square feet (sq. ft.).

bedrooms – Number of bedrooms in the property (blank if not applicable, e.g., commercial plots).

bathrooms – Number of bathrooms in the property (blank if not applicable).

occupancy_status – Current occupancy state of the property, such as vacant or occupied.

geo_point – Combined latitude and longitude coordinates in the format longitude,latitude.

link_url – Direct link to the property listing on the source platform.

title – Short headline or title from the property listing.

address – Street name, neighborhood, or locality of the property.

description – Detailed property description as provided by the listing source.

longitude – Longitude coordinate of the property’s location.

latitude – Latitude coordinate of the property’s location.

price_per_sqft – Price of the property per square foot, calculated as price_value / floor_area_sqft.

invalid_data_flag – Data quality indicator:

0 = Valid entry

1 = Potentially invalid or incomplete entry

area_zone – Classified zone or region within Chittagong Division where the property is located.

nearest_hospital – Name of the closest hospital to the property.

dist_to_hospital_km – Distance from the property to the nearest hospital, in kilometers.

nearest_school – Name of the closest school to the property.

dist_to_school_km – Distance from the property to the nearest school, in kilometers.

nearest_shopping – Name of the closest shopping center, plaza, or market.

dist_to_shopping_km – Distance from the property to the nearest shopping area, in kilometers.

nearest_station – Name of the nearest public transportation hub (bus terminal or train station).

dist_to_station_km – Distance from the property to the nearest station, in kilometers.

Special Scoring Fields walkability_score – Measures how pedestrian-friendly the property location is (0–1 scale):

0 = Poor walkability (very few amenities within walking distance)

0.5 = Moderate walkability (some amenities nearby)

1 = Excellent walkability (most amenities within walking distance)

population_density_band – Classification of the surrounding area’s population density:

Low = Sparse population, more open space

Medium = Balanced population density

High = Densely populated, urbanized area

competitive_price_score – Indicates how competitive the property’s price is compared to similar listings:

0 = Above market average (less competitive)

1 = At or below market average (more competitive)

popularity_score – Reflects public interest in the property based on engagement signals (0–1 scale):

0 = Low interest

0.5 = Moderate interest

1 = High interest

lead_hotness_score – Predicts the likelihood of generating buyer leads (0–1 scale):

Values closer to 0 = Low chance of generating leads

Values closer to 1 = High chance of generating leads

AI-Assisted Real Estate Valuation Market Outlook

According to our latest research, the AI-Assisted Real Estate Valuation market size reached USD 1.68 billion in 2024, with a robust compound annual growth rate (CAGR) of 14.7% projected through the forecast period. By 2033, the market is anticipated to achieve a value of approximately USD 5.22 billion, driven by escalating digital transformation across the real estate sector and increasing adoption of advanced analytics for property valuation. The marketÂ’s expansion is underpinned by the growing need for accurate, transparent, and efficient valuation processes, which are critical for decision-making among real estate professionals, investors, and regulatory bodies.

One of the primary growth factors fueling the AI-Assisted Real Estate Valuation market is the rapid digitalization of the real estate industry. As property markets become increasingly dynamic and complex, traditional methods of valuation are often unable to keep pace with the volume and diversity of data generated. AI-powered valuation tools leverage machine learning algorithms and big data analytics to process vast datasets, including historical sales, location-based trends, and market fluctuations, thereby delivering more precise and timely property valuations. This technological advancement not only enhances the accuracy of appraisals but also reduces the time and operational costs associated with manual processes, making AI solutions highly attractive for real estate agencies and financial institutions.

Another significant driver is the rising demand for transparency and compliance in property transactions. Regulatory bodies and financial institutions are placing greater emphasis on standardized and auditable valuation methodologies to mitigate risks associated with property investments and lending. AI-assisted platforms offer traceable, data-driven insights that align with regulatory requirements and foster trust among stakeholders. The ability of AI systems to continuously learn and adapt to changing market conditions further strengthens their value proposition, ensuring that valuations remain relevant and reliable even in volatile market environments. This shift towards data-driven decision-making is expected to accelerate the adoption of AI-assisted valuation tools globally.

The integration of AI with emerging technologies such as Geographic Information Systems (GIS), Internet of Things (IoT), and blockchain is also propelling market growth. These integrations enable real-time data collection and analysis, automate property inspections, and secure transaction records, thereby streamlining the entire valuation process. In addition, the proliferation of cloud-based platforms has democratized access to sophisticated AI tools, enabling small and medium-sized enterprises (SMEs) and individual appraisers to leverage advanced analytics without significant upfront investments in infrastructure. As a result, the AI-Assisted Real Estate Valuation market is witnessing increased participation from diverse end-user segments, further amplifying its growth trajectory.

Regionally, North America leads the market, owing to the early adoption of AI technologies, a mature real estate ecosystem, and supportive regulatory frameworks. Europe follows closely, driven by stringent compliance standards and a high degree of digital literacy among market participants. The Asia Pacific region is emerging as a high-growth market, fueled by rapid urbanization, expanding property markets, and increasing investments in PropTech. Latin America and the Middle East & Africa are also exhibiting steady growth, albeit from a smaller base, as digital transformation initiatives gain momentum in these regions. Overall, the global landscape is characterized by a strong emphasis on innovation, data security, and scalability, which are expected to shape market dynamics through 2033.

In recent years, the advent of Real Estate Investment Analytics AI has revolutionized the way investors approach the property market. By harnessing the power of artificial intelligence, this technology enables investors to analyze vast amounts of data, including historical trends, market dynamics, and economic indicators, to make more informed investment decisions. The ability to predict market movements and identify lu