Patent searching – overview and importance
In 2015 there were 629,647 patent applications filed in the United States alone, which equates to more than 1,700 applications per day. Of these, the US Patent and Trademark Office granted 325,979, which is close to 900 grants per day. Nowadays, innovators in R&D centres do a preliminary patent search for his or her invention. However, he or she may search and find nothing – searching for prior art results requires a professional searcher. The evolution of technology through incremental inventions has made it tough for innovators – and even tougher for professional searchers – to identify appropriate prior art in a fixed timeframe.
Apple, Google and Samsung have set significant examples in recent lawsuits. In the IP industry, the slightest detail can be determinative in a million-dollar lawsuit. Therefore, a robust, efficient and quick search mechanism for identifying patents can make a huge difference to businesses.
Evolution of patent searching – approaches and tools
From searching patents manually in patent office libraries to the use of complex algorithms in patent databases, patent searching has matured considerably. A new generation of patent analysis tools has emerged that apply big-data analytics, cloud technologies, modern software development and data-improvement techniques. Further, with the advent of Symantec Search Algorithms, developers have created a new milestone in patent searching.
Patents may be broadly classified as either utility patents or design patents. Accordingly, searchers may perform a utility or a design patent search to identify relevant prior art. Since patent documents are classified (eg, International Patent Classifications and Cooperative Patent Classifications (CPCs)), searches for a given concept must not be limited to a keyword-based approach. Sometimes a concept can be expressed perfectly using a classification symbol. While utility patents are classified into CPCs and national patent classes, design patents are generally classified into design classes such as the national design classes in the United States, Canada and Japan and the Locarno Classification. For best practice, patents should be searched for using a combination of keywords and classes. Patent search databases and tools such as Questel Orbit, Thomson Innovation and the World Intellectual Property Organisation’s Global Design Database are capable of combining concepts in a user-friendly manner in order to perform prior art searches.
Generally speaking, for mechanical engineers, technology becomes simpler after looking at the mechanisms. In the field of mechanical engineering, inventions are supported and described by corresponding images. Further, in case of an incremental invention, a descriptive patent drawing can assist in the understanding and implementation of the invention.
As a patent analyst, I perform patent searches to assist R&D centres globally. In my experience, while searching for a mechanical or mechanical-allied technology, a good patent image is highly useful. Thousands of patents can be screened based on their images and the useful ones can be quickly identified for further analysis. Moreover, this methodology assists in drastically reducing the time spent on searching and allows me to spend more time on analysis.
Next-generation patent searching
Imagine if examiners had a tool for uploading images of an invention and retrieving similar patents filed worldwide. Even better, imagine if that tool could accept hand-drawn images and return visually similar patents.
Introducing reverse image searching
In lay terms, image search engines (eg, reverse image search engines) are special search engines which do not require keywords to find pictures. Instead, they find images similar to those entered by the searcher. Reverse image searching includes an option to use images along with attributes such as tags, qualifiers, domains and categories, which allows the reverse image search algorithm to narrow search process and yield better results.
When searching using an image, results may include:
- similar images; and
- sites that include that image.
Searching using images works best when the image is likely to show up in other places on the Internet. Therefore, famous landmarks are likely to yield more hits than personal images like family photos.
However, the real task is identifying the correct document based on the uploaded image. Visually similar images and documents comprising those images are easily retrieved, but establishing their relevance requires manual analysis.
Even though the first reverse image search tool from Google was introduced in 2001, these tools are yet to receive global acceptance for patent searching, which may be due to lack of awareness of, or lack of appropriate search algorithms used in these tools. Our next report will take a closer look at the types of reverse image search tool available on the Internet, their application in patent searching and their limitations and challenges.
For further information please contact:
This is a co-published article whose content has not been commissioned or written by the IAM editorial team, but which has been proofed and edited to run in accordance with the IAM style guide.