In this section you will find material about search strategies which will help you in finding further insight for your own search. Read how examiners perform a search and how they build strategies to get the best possible search results.
Some aspects of patent searching are similar to searching other kinds of technical literature, such as articles in popular science magazines or papers in professional and academic journals. However some aspects of searching patents are sufficiently different from searching conventional literature, that a dedicated mindset, strategy and tactics need to be applied to the task. The difference between searching conventional publications and patents arise, amongst other things, from the nature of the patent documents themselves, the interrelations between them and the unique way in which patent databases are constructed. This article describe an approach, and initial techniques for searching patents, which will enable the absolute beginner to make a start, and gain confidence is searching patent literature.
In this study we analyse the patent documentation coverage of the Cooperative Patent Classification
system (CPC) in comparison with the coverage of the International Patent Classification system (IPC).
We demonstrate that for patent searchers the coverage in terms of patent families is more relevant than the coverage in terms of patent documents. Of all the patent families classified in the IPC only about one third thereof is currently classified in the CPC. Most of the patent families that are not classified in the CPC are patent families including published patent applications filed at the three major Asian patent offices (JPO, KIPO and SIPO). We demonstrate a method of comparing the CPC patent documentation coverage to the IPC patent documentation coverage down to IPC subgroup level and how this can be applied to improve the efficiency of classification based search. The coverage of the CPC relative to the IPC must be understood in terms of: the CPC being assigned to families, and the IPC being assigned to individual documents. Furthermore, the CPC has only been implemented relatively recently compared to the IPC, with SIPO and KIPO focusing on the classification of their front files in the CPC.
A search for prior art should end with the retrieval of the best available prior art documents and,ideally, it should be performed in the shortest time possible. Optimising search efficiency is key to the work of a searcher. The best tool for a quality and efficient search is a good strategy.
The initial criteria for searching their subject-matter must be selected, involving not just the technical tools or data bases which will be the most appropriate, but also the precise scope of the search: The where and what to search. Then, the searcher must select an initial strategy for carrying out the search in, what would seem, the most efficient manner. The main focus of the search has to be established: The how to search.
And once the search starts, the initial main focus must be continuously revised and adapted in an iterative process in view of the intermediate search results obtained and, if necessary, all the previous criteria and strategies should be changed on the go, steering the search towards the best prior art in the shortest time possible.
Search, display, translate, visualise, integrate into the web: the EPO has made substantial progress in all of these areas in the last few years. This workshop illustrates what is now available for EP data in the EPO’s patent information products, with a special focus on “Patent information services for experts”.
Complex applications with complex claim structures can lead to the EPO issuing a declaration of “no search” or an “incomplete search”. Getting an insight into the way examiners treat such applications will help you when you have to analyse incomplete search reports.
This workshop will introduce participants to the legal requirements related to complex applications. Using examples, the following categories of complex applications will be explained: