Along with the 21st century industrial revolution, the technological superiority is increasing day by day. This has altered the traditional role of an IP Professional. The complex patent searching with the ever-growing patent data is now made easy by the new generation systems and patent searching and analysis software like PatSeer.

Largely, there are four types of patent searches – Patentability or Prior Art Search, Freedom-to-operate (FTO) Search, Invalidity Search, and Landscape Search. An IP Professional can go for a broader or narrower approach depending on the type of search. Taking a deep dive into the Patent Analytics, Team PatSeer has broadly crafted some of the cheat codes for your effortless patent searching.

Keyword – Classification and Number Searching

It might sound crazy to use IPCs/CPCs in your initial searching stages, but it does give better results when you start the patent searching with few keywords. Many IP Experts would agree that starting with a complicated search query with Boolean and proximity operators, dates, etc. creates a vertical trap where you go through a stack of results that were screened out or in by your first search. Doing many searches with many relevant keywords within the patent database and looking for the relevant patent or patents is always advisable. Once you find these relevant documents that are in your technological space, you can start searching by using the IPCs/CPCs in these patents. Using the patent number of an expired patent on the relevant subject matter as the search term can also give you better results because later patents in similar tech space may have cited it as a prior art.

Truncations, Boolean and Proximity Operators

Every patent analyst knows the Boolean and proximity logic for query construction. The point lies in knowing which operator to use effectively in the search query. This is because using OR in the query fetches too many results. For example, you have around 10-15 keywords including synonyms describing the technology to be searched, and you use AND operator. You’ll get a few results which might be useful, or it may also happen that your query fetches zero search results. It happens since a patent document might not contain all the keywords, especially synonyms. Also, the using proximity operators and truncations properly in a search string enhances your search result set, letting you discover more relevant patent documents easily.

Still, it’s better to join the keywords with OR, as this will fetch all the patent documents containing at least one keyword. The key aspect in this approach is to use the sorting algorithm. PatSeer’s ‘Sort by Relevance’ will display the best matched patents across all the keywords including the synonyms. And Voila! You have your relevant patent documents on top of your search results.

Let Technology Work for You – Semantic AI Searching

Using patent searching and analysis software features like PatSeer’s AI based Semantic Searching aids you in defining your priority keywords for the searching and filtering of the patent documents. Such software algorithms create a lengthy and complex query for you and allows you to modify it according to your understanding and searching needs. Working with the NextGen AI/ML technology for superior patent searching facilitates in retrieving the relevant patent data efficiently.

Similarity Search

Searching patent records that are similar to a found relevant patent utilizing multiple methods, using a patent analysis software like PatSeer, such as Co-Citations and Sub-classifications increases the relevance in the search result set. This improves the searching efficiency as you get to reach more relevant patent documents in a faster way.

Citation – Forward and Backward

The well-known and academically most studied patent data value indicator is the number of forward citations a patent has received. Patents which accrue a greater number of forward citations at a higher rate are generally considered to be important/fundamental patent documents or prior arts. After narrowing down your search, it’s advisable to sort the results by forward citations as it helps in reading and understanding the most cited patents. Similarly, the backward citations in a patent document aids in finding the prior art for the searched technology or invention.

Advanced Clustering – Keywords, Themes, and Topics

A clustering software can be greatly beneficial in creating and organizing large sets of patent documents into logical concepts. Advanced Clustering is PatSeer’s inbuilt clustering engine that enables you generate clusters based on keywords, topics, and themes. Clustering acts as a keyword extraction tool for the entire set of patent documents and enriches your analysis.

Repeat Your Searches with Different Approach

A good patent search is a hectic process. Reiterating the search makes a quite easier to approach towards the relevant search results. Zooming out with a broad search and getting the feel for the appropriate results and then zooming in with a few classification classes or assignee information or dates and again zooming out based on this data helps you understand the patent documents better. This reiteration enables you to discover all the relevant prior art or patents related to your keywords in your search.

Do a mini brainstorm for each keyword. It gives you the space to be creative and thoughtful about their use in your search. It’s sometimes excruciating to find a relevant patent but at least you know that the search strategy is right on point. Starting narrow and focused on what really you’re looking for gives you initial comfort in viewing the search results and then you can broaden your search from there. Stay focused and persistent.

Establish a Tracking System

It’s always better to keep a list of all searched relevant patents and discarding the patents that are no longer relevant. Keep the list as precise as possible, including patent number, company, and inventor information, abstract, etc. It’ll be useful throughout the length of your patent search project. Also, keeping a list of your search strings will assist you in identifying how that search has helped you in getting relevant hits. This also gives you an idea of how broad or narrow that search string has worked.