Dataset Card for Housing_Dataset

This typical dataset contains all the building permits issued or in progress within the city of Seattle starting from 2000 to recent, and this dataset is still updating as time flows. Information includes permit records urls, detailed address, and building costs etc., which will be presented in the housing_dataset.py file and the following description

Dataset Details

Dataset Description

This Seattle Housing permits dataset is authorized by Seattle Government and could be found in Seattle Government open data portal. The Building Permits dataset from the City of Seattle's Open Data portal provides comprehensive information about building permits issued or currently in progress within Seattle. This dataset, which dates back to 1990 and continues to be updated, includes a wide range of details such as permit numbers, types, descriptions, estimated project costs, and related contractor information could be found in the .csv table in the official website, which in total contains 25 columns. Moreover, Seattle is divided in 13 Neighborhood District. Based on the Seattle Neighborhood District GeoJson File found on Seattle government website, there will a new column created, namely NeighborhoodDistrict. With the provided GeoJson file, every housing will be assigned to the corresponding neighborhood district using the Latitude and Longitude columns in the csv for future usage.

• Curated by: Seattle Government Open data portal
• Language(s) (NLP): [English]
• License: Public Domain by Seattle Government

Dataset Sources

• Offical Website: [https://data.seattle.gov/]
• Repository for Cleaned Dataset: [https://github.com/HathawayLiu/Housing_dataset]

Uses

The Building Permits dataset from the City of Seattle is intended for use in various urban development and research applications. It can assist in understanding building trends in Seattle, aid city planning, and support academic research on urban development. The dataset is also a valuable tool for residents and businesses to stay informed about construction activities and regulations in the city. Specifically for residents, this dataset provides starting information for choosing future housing by looking at housing cost, neighborhood district, and other information in the dataset. Additionally, it supports transparency and public engagement in city planning processes.

Direct Use

The Building Permits dataset from the City of Seattle is suitable for several use cases:

• Urban Planning and Development: Planners and developers can analyze trends in building permits to inform city development strategies and infrastructure planning.
• Academic Research: Researchers in urban studies, economics, and social sciences can use the data for studies on urban growth, housing, and economic activity.
• Real Estate Analysis: Real estate professionals can assess building activities in neighborhoods for market analysis and investment decisions.
• Public Awareness: The general public can use this data to stay informed about construction activities and developmental changes in their community.
• Government and Policy Making: Local government officials can utilize this data to make informed decisions on housing policies, zoning laws, and community development projects.
• Residents housing choice: Residents could access this dataset for relevant information for their future housing choice.

Out-of-Scope Use

The Building Permits dataset from the City of Seattle should not be used for purposes that could infringe on privacy or for activities that are not in line with ethical standards. This includes any form of misuse or malicious use such as targeting individuals or businesses based on the information provided in the dataset. Additionally, the dataset may not be suitable for applications requiring highly specialized or non-public information about building structures, as it primarily contains permit-related data.

Dataset Structure

The cleaned and modified full dataset[Building_Permits_Cleaned.csv], the splited train[housing_train_dataset.csv] and test[housing_test_dataset.csv] dataset are provided in the following Github Repo: [https://github.com/HathawayLiu/Housing_dataset]. The cleaned train and test dataset are also provided in the data folder of this repo.

The cleaned dataset in total contains 26 columns:

• PermitNum(string): The tracking number used to refer to this permit in SDCI's tracking system.
• PermitClass(string): The permit class tells you the type of project.
• PermitClassMapped(string): A description of whether the permit is for a residential or non-residential project.
• PermitTypeMapped(string): The permit type by category, such as building, demolition, roofing, grading, and environmentally critical areas.
• PermitTypeDesc(string): Additional information about the type of permit. For example, whether it is an addition/alternation or a new project.
• Description(string): A brief description of the work that will be done under this permit. This description is subject to change before SDCI issues the permit. The description is generally more stable if we have issued the permit. Very long descriptions have been truncated.
• HousingUnits(int): The number of housing units included at the beginning of the project.
• HousingUnitsRemoved(int) The number of housing units removed during the project.
• HousingUnitsAdded(int): The number of housing units added during the project.
• EstProjectCost(float): The estimated project cost of the work being proposed is based on fair market value (parts plus labor). The estimated cost (if any) represents the best available information to date, and is subject to change if the project is modified. We do not collect the estimated project cost for all permit types.
• AppliedDate(string): The date SDCI accepted the application as a complete submittal.
• IssuedDate(string): The date SDCI issued the permit. If there is an Application Date but no Issue Date, this generally means the application is still under review.
• ExpiresDate(string): The date the application is due to expire. Generally, this is the date by which work is supposed to be completed (barring renewals or further extensions). If there is not an Expiration Date, this generally means the permit has not been issued.
• CompletedDate(string): The date the permit had all its inspections completed. If there is an Issue Date but not a Completed Date, this generally means the permit is still under inspection.
• RelatedMup(string): The land use permit that is related to this building permit, if there is one.
• OriginalAddress1(string): The street name and number of the project.
• OriginalCity(string): The city for the project's address.
• OriginalState(string): The state for the project's address.
• OriginalZip(string): The Zip code for the project's address.
• ContractorCompanyName(string): The contractor(s) associated with this permit.
• Link(string): A link to view full details and current status information about this permit at SDCI's website.
• Latitude(float): Latitude of the worksite where permit activity occurs. May be missing for a small number of permits considered "unaddressable."
• Longitude(float): Longitude of the worksite where permit activity occurs. May be missing for a small number of permits considered "unaddressable."
• Location1(string): The latitude and longitude location for mapping purposes.
• (New added column)NeighborhoodDistrict(string): The district that the housing belongs to according to location

Dataset Creation

Curation Rationale

The Building Permits dataset from the City of Seattle was created to foster transparency, public awareness, and engagement in the city's urban development processes. It provides residents, professionals, and researchers with detailed information about building activities, facilitating informed decision-making and community involvement in city planning and development. Regarding the importance fo 13 neighborhood districts in Seattle, the new added columns for corresponding neighborhood district gives chance for residents and government to investigate the building activities and life quality in the aspect of different neighborhood districts. The dataset supports the city's commitment to open data and the promotion of data-driven insights for improving urban infrastructure and living conditions.

Data Collection and Processing

The Building Permits dataset is collected by Seattle Government where it contains all of the recent information about housing permits in Seattle. The dataset is published on Seattle Government Open Data Portal and it's keep updating along with time. You can download the raw data from Seattle Government Website in different formats. For my own purpose I downloaded the CSV version that updated until the modified time of this repo and you can find it in the following Github Repo:[https://github.com/HathawayLiu/Housing_dataset] (File name: Building_Permits_20240213.csv). To process and clean the dataset, I did the following steps:

1. Pre-process the data to make sure that they are in the correct types.
2. Use the provided latitude and longitude columns in the dataset along with Google GeoCoding API to fill in the blanks for the OriginalZip(Zip code) column.
3. Use the provided latitude and longitude columns and the GeoJSon file of Seattle Neighborhood District to assign building permits to their corresponding neighborhood districts.
4. (The GeoJSon file of Seattle Neighborhood District could be found under this GitHub Repo:[https://github.com/HathawayLiu/Housing_dataset]. You could also download it through Seattle GeoData Portal:https://data-seattlecitygis.opendata.arcgis.com/datasets/SeattleCityGIS::neighborhood-map-atlas-districts/about)
5. Fill in the blanks left in the dataset with N/A for easier future use
6. Split the dataset into train and test set for future use.

For more details about data cleaning and processing, you could refer to the data_cleaning.py file under this repo. Notice that to be able to use the function to get zipcode, you need to use your own API Key. Applying for a Google GeoCoding API is free. You could simply follow this link to apply it: https://developers.google.com/maps/documentation/geocoding/get-api-key You are more than welcome to download the raw data and process the dataset yourself.

To load the dataset, you could use the following command:

!pip install datasets
from datasets import load_dataset
dataset = load_dataset("HathawayLiu/housing_dataset", trust_remote_code=True)

To generate the exmaple from train/test set, use:

next(iter(dataset['train']))
## next(iter(dataset['test']))

You can see the example from dataset like the following:

{'PermitNum': '6075593-CN',
 'PermitClass': 'Single Family/Duplex',
 'PermitClassMapped': 'Residential',
 'PermitTypeMapped': 'Building',
 'PermitTypeDesc': 'Addition/Alteration',
 'Description': 'Replace existing windows; Upgrade new windows and framing for existing single family residence subject to field inspection',
 'HousingUnits': 0,
 'HousingUnitsRemoved': 0,
 'HousingUnitsAdded': 0,
 'EstProjectCost': 43014.0,
 'AppliedDate': '10/12/05',
 'IssuedDate': '10/12/05',
 'ExpiresDate': '4/12/07',
 'CompletedDate': '2/1/06',
 'StatusCurrent': 'Completed',
 'RelatedMup': 'nan',
 'OriginalAddress1': '624 NW 88TH ST',
 'OriginalCity': 'SEATTLE',
 'OriginalState': 'WA',
 'OriginalZip': 98117,
 'ContractorCompanyName': 'STATEWIDE INC',
 'Link': 'https://cosaccela.seattle.gov/portal/customize/LinkToRecord.aspx?altId=6075593-CN',
 'Latitude': 47.692996978759766,
 'Longitude': -122.36441040039062,
 'Location1': '47.69299754, -122.3644121',
 'NeighborDistrict': 'Northwest'}

Who are the source data producers?

The Building Permits dataset is originally created and maintained by the City of Seattle, specifically by its Department of Construction and Inspections. This department is responsible for overseeing building and land use in Seattle, ensuring safety and compliance with city codes. The dataset reflects the department's ongoing work in managing and documenting building permits issued in the city. For detailed information, visit the Seattle Department of Construction & Inspections.

Bias, Risks, and Limitations

The Building Permits dataset from the City of Seattle has both technical and sociotechnical limitations:

1. Technical Limitations:

   • Data Completeness: Not all building permits may be captured, especially older records. Data for specific columns like IssuedDate, CompletedDate, AppliedDate, RelatedMup, and etc. contains lots of missing values.
   • Data Accuracy: There may be errors or inconsistencies in the data, especially in historical records.
   • Timeliness: The dataset might not be updated in real-time, causing delays in reflecting the most current information.

2. Sociotechnical Limitations:

   • Privacy Concerns: Detailed permit data could potentially be used to infer private information about property owners or residents.
   • Bias in Planning Decisions: The data might be used to reinforce existing biases in urban planning, affecting marginalized communities.
   • Dependence on Technical Proficiency: The dataset's utility is limited by the user's ability to interpret and analyze the data effectively.

3. Bias: The dataset reflects only permitted construction, not all building activities. This can bias analyses towards formal, recorded developments, overlooking informal or unpermitted construction.

4. Risk: Misuse can occur if data is used to unfairly target specific neighborhoods or communities for enforcement or political reasons. These limitations should be considered when using this dataset for research, policy-making, or urban planning.

Recommendations

To address the bias and limitations above, users should intake the following recommendations:

• Cross-Verification: Use supplementary data sources for a more comprehensive view.
• Privacy and Ethical Use: Handle data responsibly, respecting privacy and avoiding discriminatory practices.
• Data Cleaning and Validation: Regularly update and clean the dataset to maintain accuracy and reliability.