PatentsView is introducing our newest data download product: Sorted Download Tables. These new download tables will complement our Granted and Pre-Grant Download Tables and enhance the user experience by providing a more consistent and accurate representation of patent data across different eras. Please note that this is a Beta release, and significant future changes to the Sorted Download Tables based on community feedback are possible. Contact us at PatentsView Support.

Why the New Files?

In a previous Data in Action Spotlight post, we discussed changes in the rate of assignees in pre-grant publications. This change is often misinterpreted as a decrease in organizational involvement in patents or a data error. However, it actually reflects a shift in application rules due to the America Invents Act (AIA) and a corresponding change in where organizations appear in the data structure. A technical explanation of the problem with the source XML data and the solution implemented can also be found in this USPTO technical paper.

What are Sorted Download Tables?

The sorted files regroup records from both pre- and post-AIA eras for consistency over time. Here's how the new files were derived from the old ones:

• Organizations: Records with an organization name or where an organization was mistakenly listed as a person.
• Individuals: Records without an organization name that don't appear to be misrecorded organizations.
• Assignees: Records from the original assignee table, plus records from the original applicants table with "assignee" as the applicant’s authority.
• Inventors: Records from the original inventor table not listed as a "legal representative," plus records from the original applicants table with "inventor" as the applicant’s authority or "applicant-inventor" as the applicant’s type.
• Applicants: Remaining records from the original applicants table not marked as an inventor or assignee, plus records from the original inventor table marked as a "legal representative."

A visualization of this logic is available in the Beta Sorted Data Reference document.

Structure of the New Files

Each table starts with columns for granted patent identifier (patent_id) or pre-grant publication identifier (pgpub_id), sequence, and origin_section. The origin_section column indicates which of the original unsorted, non-disambiguated tables the record came from. Following these, you'll find individual and organization names, along with other relevant details from the original tables.  Additionally, each table contains the raw location ID, which can be matched with the raw locations table available on the main downloads page.

A detailed data dictionary Excel file is also available.

New Way of Organizing Data

These files are an additional resource and do not replace the existing files on our website. They are simply regrouped to provide a more intuitive way of navigating our extensive data. This reorganization is a test of a potential new structure that we hope will enhance your experience.

We hope to provide disambiguated data in a similar structure in the future, but at this stage, these sorted files are only available for non-disambiguated data.

We Value Your Feedback

We invite you to explore this new layout and share your thoughts with us through our service desk. Your feedback is invaluable as it will help us understand how these changes affect your ability to find and understand the data that is most relevant to you.

MoleculeSearch.ai is a cutting-edge tool designed to transform the way researchers, chemists, and patent professionals explore and analyze chemical structures and their associated patents. 

Utilizing SMILES (Simplified Molecular Input Line Entry System) notation, the platform provides a quick and precise method to search for molecular data across extensive databases. SMILES notation, akin to a zip code for molecules, distills complex molecular structures into a manageable string of characters, allowing for efficient and accurate searches.

Harnessing the Power of Open Data and Advanced Technologies

The MoleculeSearch.ai platform integrates open data from PatentsView and SureChEMBL, offering users a wealth of chemical and patent information. PatentsView provides detailed insights into patent statuses and historical data, making it a vital resource for patent strategy and research. SureChEMBL complements this by supplying patents to molecules mapping that helps researchers identify new compounds and understand existing ones.

Additionally, MoleculeSearch.ai employs RDKit, a leading cheminformatics tool, to enhance the precision of molecular searches and comparisons. This integration allows for detailed analysis of complex chemical structures, supporting efforts in drug discovery and chemical research.

The inclusion of AI capabilities through technologies like ChatGPT enhances the user experience by creating accessible summaries of the core concepts in a patent. This means that even non-experts can easily navigate complex data, making the tool accessible to a broader audience.

Practical Applications and Concrete Examples

1. Drug Discovery: Researchers can use MoleculeSearch.ai to identify similar molecules to a known drug compound. For instance, if a researcher inputs the SMILES string of Sildenafil (Viagra), the tool can help identify structurally similar compounds that might also possess therapeutic properties, accelerating the search for new treatments.
2. Patent Analysis: Patent professionals might use the platform to check the patent status of a new chemical entity before filing. By entering the SMILES code of the molecule, they can quickly find out if similar structures have already been patented, which is crucial for avoiding infringement and understanding the competitive landscape.
3. Academic Research: Academics studying chemical synthesis can use MoleculeSearch.ai to find molecules with similar properties that have been involved in previous studies. This can help in hypothesizing new synthetic pathways and understanding reaction mechanisms.
4. Quality Control in Manufacturing: In industries like pharmaceuticals, ensuring the purity of chemical compounds is critical. Using MoleculeSearch.ai, QC chemists can compare the SMILES strings of impurities with known substances to quickly identify contaminants and their possible sources.

Molecule-search.com stands out as a valuable resource for anyone involved in chemical research, drug discovery, and patent analysis. By combining open data, advanced cheminformatics technology, and AI, it offers a powerful platform that simplifies and enhances the workflow of chemical and patent professionals globally.

We Want to Hear from You!

Have you used PatentsView data in new or interesting ways? Have you published innovative research using PatentsView data? We would love to feature your work on our Data in Action Spotlight. Visit our service desk to submit an idea.

PatentsView has always been a leader in providing high-quality patents data to help drive insights into invention and innovation. The platform offers tools to help researchers better understand intellectual property (IP), inventors, and innovation. Users can also explore trends and connections between various topics to gain a deeper understanding of the IP landscape.

Our team has been working diligently behind the scenes to not only uphold our reputation for high-quality data and disambiguation, but to make PatentsView better and more functional. Here are a few ways we are making PatentsView better for you.

Service Desk

We recently launched a new service desk to help users request an API key, get technical support, report a bug, or suggest improvements. The service desk also helps the PatentsView team better track requests and use your suggestions for continuous quality improvements.

PatentSearch API

The PatentSearch API’s full-text endpoints have been updated. For clarity and efficiency, granted and pre-grant text endpoints have been separated.

• Granted text data can be retrieved at /api/v1/g_brf_sum_text/, etc. 
• Pre-grant text data can be retrieved at /api/v1/pg_brf_sum_text/, etc.
• The json keys for these endpoints' responses have been updated correspondingly to g_brf_sum_texts, etc.

Full update notes can be found at https://search.patentsview.org/docs/2024/11/06/2.2-release.

The new PatentSearch API is more advanced and efficient than the legacy API, which will be phased out in February 2025. Learn more about PatentSearch API in our PatentSearch API Reference page and Swagger interface.

Ready to switch? Request a PatentSearch API key through our service desk.

Ground Truth and Data Quality Checks

To ensure the highest level of data accuracy, the PatentsView team has implemented several Ground Truth initiatives. These efforts involve cross-referencing patent data with verified sources to validate the information and correct any discrepancies. 

By establishing a reliable ground truth, users can access more trustworthy data, which is crucial for conducting accurate analyses and making informed decisions. This commitment to data quality is what makes PatentsView the best in its class for patent data.
 

PatentsView is phasing out our legacy API, making way for the more advanced and efficient PatentSearch API. In September 2024, we began the process of retiring the old API, with full discontinuation set for February 2025. 

We encourage all users to transition to the new PatentSearch API to ensure uninterrupted access to our services and to take advantage of the enhanced features and speed it offers.

Ready to switch? Request a PatentSearch API key through our service desk.

About PatentSearch API

The new PatentSearch API is intended to inspire exploration and enhanced understanding of US intellectual property (IP) and innovation systems. The database for the PatentSearch API is updated regularly and features the best available tools for inventor disambiguation and data quality control.

Researchers and developers can use the API to uncover information about people and companies, and to visualize trends and patterns in the US innovation landscape.

The API offers seven unique endpoints that allow users to explore various questions, such as:

• Which companies hold patents in 3D printing? Discover their locations and the technologies they were innovating in before and after receiving 3D printing patents.
• What technology has been most commonly patented in the US in the last five years? Identify the top US and non-US cities producing these patents.
• Which US inventors earned the most patents in the last 30 years? Track their patenting activity, including the number and types of patents and their co-inventors.

Learn More About PatentSearch API

For detailed documentation on the new PatentSearch API, visit our PatentSearch API Reference page. Additionally, you can explore the Swagger interface.

Why Are We Switching APIs?

PatentsView has been offering an API since 2015, which has been widely used and valued by thousands of users. 

However, based on years of feedback and the evolving nature of patent data, PatentsView released the PatentSearch API in early 2024. This new API enhances the functionality and speed of the previous version.

Consolidating Names

Over time, the legacy API has been known by various names, including “PatentsView API,” “Swagger-based API,” and “MySQL API.” The new PatentSearch API has also been referred to as the “Elasticsearch API,” “Beta API,” and “Search API.” 

Moving forward, we will consolidate the naming to its official title – PatentSearch API.

In the world of software startups, patents are often seen as a golden ticket to securing investment. But do they really hold that much sway? A recent study by James Hicks looks into this question, focusing on the impact of a landmark Supreme Court decision, Alice Corp. v. CLS Bank International, on software patents and investments.

The study found that patents on certain types of software did not necessarily increase investment and other positive business outcomes. The findings encourage us to look more closely at when patents are needed to protect intellectual property.

What is the Alice decision?

The Alice decision changed the criteria for software inventions to be eligible for patents. The case involved an electronic service that implemented financial trading systems. The Alice ruling stated that the service was based on an abstract idea, and therefore was not patentable. In practice, businesses applied this principle to any software involving business methods, which led to a dramatic drop in investment in this type of patent.

Despite this, Hicks' study found no evidence that this drop in the number of patents led to less early-stage venture capital investment or affected outcomes such as acquisitions and initial public offerings (IPOs).

The results indicated that patents may not be necessary for investment in the software industry. While patents are needed to protect certain types of intellectual property, these findings suggest that we may need to take another look at the value of patents in other areas.

PatentsView Data: Unraveling the Relationship

The study used data from PatentsView to identify companies that applied for patents and compare the number of patents with investment in the same company. This helped researchers understand the impact of the Alice decision on patent approval and subsequent investment in business-methods software firms. The results indicate that patents did not have a discernible effect on funding or successful outcomes in the software industry.

Reevaluating the Role of Patents in Software Innovation

The relationship between patents and investment in innovation is complex. This study suggests that the costs of obtaining patents may not always justify the benefits and invites more research into understanding when patents add value to a business. PatentsView data can help future researchers highlight connections and provide insight that will help inventors and businesses make better decisions about their intellectual property.

Disclaimer: This post was written with the assistance of artificial intelligence.