This guide covers the basic steps of carrying out a Free-Text search for an invention, when you have either Claims of a patent application available - either as draft versions or as filed, or an Invention Disclosure/Project Scope description. It will also cover what is not possible to do with Free-Text Searching (see: Generative Prompting).

English is the native language of IPRally and optimal as input. German, French, Spanish, Chinese, Japanese, Italian, Dutch, Danish, Finnish and Swedish are supported via machine translation (with a maximum 5000 characters). The translation happens inside IPRally, making it secure.

Professionally drafted Claims are an optimal input for IPRally, since the graph AI is trained with millions of professionally drafted claims.

Free-Text Search takes your input and curates a Knowledge Graph for searching. The results are listed in Ai Score order. For more information, please read Processing the Results (including descriptions of 'Ai Scores').

Free-Text Search Is Not a Generative Ai Search - meaning, you must describe the object(s) and features you are searching for, and NOT a 'ChatGPT' style prompt (See: Can't read your mind nor infer). In addition, you you may have the draft description or description as filed available, and that can be used to further improve the search is some cases.

The guide is specifically intended for

The basic steps of a Free-Text search are:

1. Enter the whole Claims Set/Invention Disclosure/Project scope to the Free text search field

2. Review the search results

3. Mark the best findings as Favorites

4. Re-run the search using Zoom to Favorites

Additionally you can (but not covered in this guide):

Good: "The invention is a vehicle-to-vehicle (V2V) communication system for autonomous vehicles, using both wireless technologies (WiFi, LTE, 5G) and RFID. It features a dual-module framework that dynamically switches between high-bandwidth and low-bandwidth communication modes to optimize data transmission. This system enhances safety, efficiency, and passenger interaction in autonomous vehicles."

Poor: "A V2V system for autonomous vehicles using wireless tech and RFID to enhance safety."

Why? The good example has technical details around the objects, and how they interact to deliver the function being searched. There are functional relationships described, and the language is descriptive vs. prompting. The poor example is too brief, lacking functional relationships, and doesn't describe the 'novelty' well enough. The results will be too 'vague' compared to the good input.

* (Curated using ChatGPT 4.o - resemblance to any prior art is by chance)

Please...

It will greatly diminish the ability for our Graph-Ai to understand the subject, and therefor provide lower-quality results.

Example of a poor 'Prompt' inspired search input:

"A V2V system for autonomous vehicles using wireless tech (WiFi, LTE, 5G) and RFID to enhance safety. Focus on highlighting the dual-module communication approach that integrates high-bandwidth wireless technologies for real-time data and low-bandwidth RFID for stable communication. Emphasize how the system dynamically switches between these modes to optimize performance, targeting improvements in safety, efficiency, and passenger experience in autonomous vehicles."

Why? Our Graph Ai is not a LLM (As all GPT's are), but rather a Knowledge Graph Neural Network. This requires the input to be similar to the patent documents being searched, and does not operate as an LLM would in 'filling in the blanks'

The four steps are described in more detail and with screenshots below.

For some FAQ/Best Practice, please click here (or scroll down to the end of the article)

Statistically, the whole Claims set alone is proven to provide almost as accurate search results as using the whole patent text, and is usually a sufficient input for a high-quality search. If available, we suggest copying it completely into Free Text Search.

If you don't have a Claim Set, a technical description/invention disclosure is a great input. The key is an input which describes embodiments according to the technical requirement in more concrete terms and with more specific examples, or the part that gives more context or technical field information for the invention.

What is NOT recommended here is a simple Title; Terms; Patent number - these inputs are for a Boolean Search.

Free-Text Search is NOT a 'Semantic' input, nor is it a Generative Ai input.

IPRally provides many functionalities for processing the search results both inside the platform and by exporting data into another review platform.

The result list and expanding the bibliographic data:

Result List:

Expanded Bibliographic Data:

Showing image mosaics:

Smart image viewer by clicking the mosaic images:

AI-based relevant passage highlighting:

If the closest prior art appears among the results of IPRally, you can click the heart symbol to make the hit a Favorite.

If, for some reason (yes, that happens 😊), the closest prior art is not among the hits of IPRally, you can manually enter the publication in the Favorites tab:

You are able to add multiple documents here as well:

We recommend using 1-5 best hits as favorites. Provided that they do not contain a lot of contradictory information, carrying out the next step (step 4) usually improves the quality of the search results. Please see our article on Zoom to Favorites for more guidance

Toggling Zoom to Favorites on and pressing SEARCH PATENTS again, will utilize the Favorites information (together with the original graph) to find more relevant hits.

Note: you can track the new hits in the list using the VIEWED status of the hits (Blue: Un-Viewed / White: Viewed). You can easily 'hide viewed' under the View Options to only see which new results appear with the new search:

The viewed status changes also automatically, if you open the full document view.

Once you have re-run the search, get back to the analysis step 2 and repeat iteratively, if needed.

When using the Free-Text tool, please use the below insights to drive your searches: