Patent Search: Few Important Links and Tips

Patent Search: Few Important Links and Tips 

Information Edited and Collected By Prof.Saurabh Mehta,VIT 

Note: This article is prepared by collecting various sources and information available on internet. Main objective of this article is to provide a starting point for an inventor/students/professional for patent filling.  

The patent filling process can be expensive, so before spending much money in preparing patent application it is important do patent search.  Patent search is a very important aspect of Patenting before filling your own patent. Patent search can help applicant in following ways: 

  1. Finding Uniqueness of Invention 
  1. Patent Survey  
  1. Similarity and Applicability 
  1. Usability and Licencing 
  1. Inventor and investor details 

It is important to note that doing your own patent search as an inventor is a very good idea, and perhaps much easier now with many online search engine/patent database available but searching patent is an art and requires enormous practice.   

Few tips to find patent quickly and accurately:  

  1. Utilize Semantic Search

    Semantic searches find relevant patent information based on the intent behind a normal language description. Semantic searching is highly effective at finding relevant patent documents that might use slightly different terms.  Good Semantic search Algorithms examine word frequency, sequencing and patterns to determine the context, regardless of the specific words used in the search. Consequently, semantic searches can find patents disguised by vague language, different terms, and the potential evolution of the industry lexicon. 


  1. Add Boolean Operators 


Boolean operators can help refine Iterative searches results even further. Expert users can put together extremely advanced patent searches using Boolean operators to capture different spellings, and list specific fields in the patent document to search. For example, if searching for the term “plane” related to computer-aided design (CAD), the search string might read “plane NOT (travel OR carpentry)” to eliminate any search results for airplanes or hand tools. 


  1.  Text-Clustering for Terminology

    Clustering text from patents is a way to extract keywords for Boolean search. This can work for an entire set of patents, rather than just one or two, and help to organize large sets of documents into logical themes or concepts. Increased familiarity with how patents are written will ultimately translate into improved keywords and search results. 


  1. Searching metadata of patent 

    Searching the metadata that accompanies a patent, including classification codes, backward and forward citations, previous/current owners, inventors’ names, dates of filing, etc., can also improve search results. Most search engines can search each of these fields individually or grouped as desired. For example, when using keywords to search the International Patent Classification codes, IP searchers can find an alphabetical list of classes and subclasses, each of which has a unique number. Searching these codes may then reveal a relevant patent under a different code.  

In conclusion, use of any method for patent searching must be efficient and effective. Use of Boolean operator, Semantic search, classification codes and text-clustering provide many crucial advantages with more relevant search results. More the practice you put with these search techniques, the better you will find the required results 

Few Patent Search Engine Sites 

  1. Google Patent Search India /Advance Google Patent Search: Google patent search tool is a first step to find relevant patent information.  However, Google Patent Search engine has some limitations. The numbers of fields that can be searched are limited in comparison to the USPTO or Free Patents Online. Thus, if you are going to use Google Patent Search it is probably best used initially because when you start a search you want to search very wide information, so the limitations of the field search are probably not as important initially.  Also, there are some holes in the database.  As you may not get updated patent information even you know though they are existed.    Additionally, the most recent patents are not always available on Google.  What this means is you cannot only rely on Google, but you still must use Google.   


  1. United States Patent and Trademark Office (USPTO) search -:The Patent search engine of USPTO searches about 240 million pages of text. It allows Full Text Searches. This tool is probably the best because it’s from the source itself. You can search by keywords, classification, inventor, assigned entity, attorney, etc.  


  1. provides free access to all US patents and partial European data. It is also normally much faster than the USPTO site as well. It is easier to get PDF version of patents from Free Patents Online because everything here is well hyperlinked. So once you find a handful of relevant patents definitely go to Free Patents Online so you can easily jump back and forth and look at the patents that are cited in each relevant patent you find.  In other words, you can consider using the strengths of both the USPTO and Free Patents Online sites to make your searching easier. 


  1. :PATENTSCOPE allows you to search 30 million patent documents including 2.2 million published international patent applications (PCT). 


  1. The Industrial Property Digital Library (IPDL) offers the public access to IP Gazettes of the Japanese Patent Office free of charge through the Internet. Here, you can access patent abstracts in English. 


  1. With its worldwide coverage and simple search features, Espacenet offers free access to information about inventions and technical developments from 1836 to today. Espacenet is accessible to beginners and experts and is updated daily. It contains data on more than 95 million patent documents from around the world. Supporting information can help you understand whether a patent has been granted and if it is still in force. 


  1. Current version of search engine IPAIRS Version 2.0 is basically a structured search, i.e. interface providing pre-defined Indexed fields for searching in the database. This version has been improved over previous so as to provide:(i) Increased no. of fields (No. of Parameters for search are now 14)(ii) Combination of Search fields (iii) Inclusion of operators (iv) Distinct (non-repetitive) results.(v) Detailed information of Patents (01/01/1995 onwards) 


  1. MCPaIRS,TM the pioneering Indian Patent Database, addresses the need of having a searchable, enriched and indexed full text Patents published from India. The data is hand-curated by domain experts and provided in an easy to use web interface. MCPaIRSTM allows the user to search for full texts of granted patents and published applications from India.The database contains a well-designed front page with bibliographic details, abstract and representative drawing. Data coverage starts from 1912 to till date and is updated every seven days.