Hunting the literature: effective use of classification and indexing
The effective use of classification or indexing schemes as part of any patent or literature search can be problematic for inexperienced searchers who may be unfamiliar with the peculiarities of a particular classification scheme or, indeed, with the theory of classification in general. In fact, even seasoned searchers may struggle if the technology they are trying to research is outside their normal area of work.
Searchers can sometimes be presented with an overwhelming choice of classification schemes which need to be considered, depending on their search requirements.
A search within patent documents, for example, may currently need to incorporate the International Patent Classification (IPC), European Classification (ECLA), and US Patent Classification (USPC) as well as Japanese FI and F-terms, while specialised subject-specific scientific literature databases will make use of their own individual classification schemes. Yet, despite these difficulties, such schemes are often the best way to achieve high quality results, quickly and efficiently.
“IN THE NEW PROQUEST DIALOG SERVICE, THIS ANALYSIS OF THE CLASSIFICATION TERMS IS UNDERTAKEN SIMULTANEOUSLY WITH THE SEARCH AND SEAMLESSLY PRESENTED TO THE SEARCHER ON THE RESULTS SCREEN.”
A number of initiatives over the years from the patent offices and database vendors have sought to simplify this facet of searching. In the latest, coming into force from January 2013, a joint partnership between the US Patent and Trademark Office and the European Patent Office will result in the harmonisation of their own individual classification schemes (USPC and ECLA, respectively) into a single Cooperative Patent Classification which will be applied across both of their patent collections.
The two offices will continue to work with the Japanese Patent Office, the Korean Intellectual Property Office and the State Intellectual Property Office of the People’s Republic of China on even further simplification and standardisation of patent classification schemes.
Information aggregators such as ProQuest Dialog, however, have little or no control over the structure or content of classification schemes in the databases which they make available to searchers. The goal here is to make the classification scheme as accessible and meaningful as possible for users with a wide range of searching expertise.
A quick and simple way to start to explore any classification scheme is to run a basic keyword search for the main concepts of interest within a database and then to analyse the classification terms which have been applied to the documents retrieved. The most frequently occurring relevant classification terms can then be applied to a subsequent search to expand retrieval to documents which did not contain the original keywords used, but which are nevertheless of potential interest.
In the new ProQuest Dialog service, this analysis of the classification terms is undertaken simultaneously with the search and seamlessly presented to the searcher on the results screen. Additional lateral search links on classification terms within documents also allow the searcher to retrieve other, similarly indexed, documents.
Both scenarios allow searchers to immerse themselves in a classification scheme which they may know little about, having started from a basic keyword search. For more advanced searchers, thesauri that can be fully browsed and searched also help in the identification of appropriate classification terms to use.
An interesting advancement of the use and simplification of classification schemes has been an attempt to map a classification scheme intended for one type of document or technology into other areas. The business intelligence tool Innography, for example, has taken US trademark goods and services classification information and mapped this to patent classifications, with the aim of correlating trademarks and the products they represent to any underlying patents.
This starts to open up novel search scenarios such as determining the patents that a trademarked product may be infringing.
In addition, anybody with an interest in electronics, computer science, physics, electrical, control, production or mechanical engineering should be aware that the Institution of Engineering and Technology’s Inspec database has mapped its own indexing schemes to the International Patent Cooperation (IPC), to facilitate the retrieval of relevant non-patent literature by means of a classification structure used for patents.
In both cases a new mass of documentation has been opened up to searchers who may be unfamiliar with the content or classification schemes normally used with them.
