Data Prep -EDA

For my project, I focused on utilizing the PatentsView API to gather data on patents related to machine learning and natural language processing. The PatentsView API offers a structured and reliable source of patent data, which I found particularly useful for extracting information on technological trends, key players, and innovation patterns within my topics of interest.

I created multiple datasets from this API, covering various years and different patent topics. By segmenting the data by year, I was able to capture the evolution of patent filings over time, providing a timeline of technological advancements. I also filtered the data by specific topics within machine learning and natural language processing, such as NLP in healthcare, machine learning in robotics, and AI-enhanced decision-making systems.

Using Python, I successfully gathered labeled and quantitative data for analysis, leveraging the API's capabilities to extract key details like patent counts, assignee information, classifications, and citation histories. These datasets will form the backbone of my analysis, helping me address key questions about technological growth, patent trends, and the competitive landscape in AI-related fields. Additionally, gathering this data sets a foundation for further visualizations and deeper qualitative insights.

How to Explore PatentView API

Using an API involves making requests to specific endpoints to retrieve data from a server. In the case of the PatentsView API, the most common method for accessing data is through an HTTP GET request. A GET request allows users to request specific data from a particular endpoint by including relevant parameters such as patent numbers, filing dates, or classification codes in the URL. These parameters help filter the data and ensure that the API returns only the information the user is interested in.

To make a GET request, users can use tools like Postman, cURL, or directly in their Python scripts using libraries such as requests. For example, to fetch patent data from the PatentsView API, a user might send a GET request to an endpoint like https://api.patentsview.org/patents/query and include parameters in the URL such as ?q={"_gte":{"patent_year":2015}} to retrieve patents filed after 2015. The API responds with the requested data, typically in a JSON format, which can then be processed or stored for further analysis.

Example link : https://api.patentsview.org/patents/query?q={"_gte":{"patent_year":2015}}

And for this project, I will go further and use f(fields) and o(outputs). Fields parameter is used to specify which fields you want the API to return in the response. It allows you to customize the response by including only the specific data fields you're interested in. And outputs parameter is used to specify additional options for the query, such as pagination or sorting.

The PatentsView API is user-friendly, and its documentation outlines available endpoints and required parameters, making it easy to explore patent data across various industries and time periods. By understanding how to structure GET requests and use the parameters effectively, users can extract valuable insights to drive their research, innovation tracking, and competitive analysis efforts.

EndPoints

Endpoints in an API are crucial access points through which users interact with the underlying data. In the context of the PatentsView API, each endpoint represents a specific set of data related to patents, including details about the patent number, filing date, assignees, inventors, classifications, and citations. These endpoints enable users to query relevant information efficiently, helping them analyze patent trends, assess technological advancements, and explore intellectual property landscapes.

By using endpoints such as "patent_number", "patent_title", and "assignee_organization", users can extract key patent metadata that reveals which companies and inventors are driving innovation in fields like machine learning and natural language processing. Additionally, endpoints like "cpc_group_id" and "patent_num_cited_by_us_patents" allow users to dive deeper into the classifications and influence of each patent, providing a comprehensive understanding of how these technologies are evolving and being utilized across industries.

These API endpoints are instrumental in offering a structured, scalable way to gather valuable patent data for research, competitive analysis, or intellectual property management. By leveraging them, users can generate datasets tailored to their needs and gain insights into the ever-changing innovation landscape.

https://patentsview.org/apis/api-endpoints/patents