The limitations of keyword-based searching are well-documented. A syntactic search engine will deliver results based solely on the words and phrases a searcher enters. Synonyms, homonyms, and translations all prove problematic in this situation. This technology, on its own, is simply not powerful enough to find all of the patent and nonpatent literature related to your novel idea.
Conceptual search, however, considers the meaning of a searcher’s query. It uses this understanding of the search to find the most relevant results, regardless of the words used by a searcher or publication to communicate the concept. The most powerful search solutions are able to identify results beyond just simple synonyms, using AI to truly understand the concepts described.
When you’re pushing a new idea through the innovation pipeline, stakes are high. R&D resources must be allocated to projects with the most potential. The administrative and legal costs associated with patenting must be carefully considered. Determining whether or not an invention is worth bringing to market requires a complete understanding of the technology landscape. This is only possible with a search engine that finds the most relevant patent and nonpatent literature, regardless of how you describe your technology.
With keyword-based search engines, finding relevant results (and eliminating irrelevant results with confidence) required at least some proficiency in specialized techniques like Boolean logic. Conceptual search democratizes patent searching, eliminating bottlenecks within the innovation lifecycle and giving searchers across professional disciplines confidence in their search results.
The AI-backed search engine that powers IP.com’s suite of IP solutions, including InnovationQ Plus®, uses concept searching to deliver highly relevant results. Our mature, unsupervised AI engine is continuously learning from patent and nonpatent literature, improving its understanding of technical and legal language.
Conceptual Searching for More Relevant Results
Keyword-based searching will include a document in the search results if the keyword appears even one time. This leads to thousands of search results that have very little to do with the searcher’s query. Concept-based searching considers how important an idea is within a document to rank search results based on relevancy. This gives you, the searcher, the ability to focus on top results rather than sifting through all results. IP.com’s AI engine gives each search result a relevancy score based on how prevalent a concept is within the document. You can then narrow down results to only those with a high score before visualizing or sorting patent data.
Conceptual search engines not only allow you to use longer phrases to describe what you’re looking for—these tools excel in this situation. You can use a few sentences or your entire invention disclosure to look for technologies similar to yours. If one part of your description is more important than another, a concept modifier can help you communicate that weight to the AI. Now, the AI engine can use all of the details of your idea and an understanding of what’s most important to search for similar inventions. These search techniques deliver more relevant search results than a few keywords can.
Sometimes, you’re not sure how to explain exactly what it is that you’re looking for. Or, you describe it in a way that’s not how someone who works in a different field, or for a different company, or in a different country would. Conceptual searching overcomes this issue because no matter how you articulate what you’re trying to find, the engine understands the concept you’re getting at. The language in the most relevant results can then be used to search again to ensure you’re finding all of the related literature, both in your industry and other industries entirely.