Information Search Capabilites of Konstantin Shilan & Co.

Our firm specializes in information search and analysis for various kinds of patent research: simple patent search by IPC groups, prior art search, patent validity and non-infringement searches. We differentiate between the conventional search and heuristic search procedures carried out with the help of automated information retrieval systems.

I. Conventional search

1) Direct documental search. We implemet simple documental search by the descriptors (classification indexes, special terms, title and abstract words, etc.) in patent and scientific literature data bases and can provide to you bibliographic data, abstracts and fulltexts of any Russian patent or published article. Searches are conducted in our own and external data bases with the help of powerfull search instruments.

2) Prior art and validity search. It is well known that the Russian information sources are very difficult to access and given the language barrier, they were and still are are very often neglected by patent examiners of many countries. At the same time the level of technology development in our country is at least not inferior to that of the developed nations and in many areas areas the Russian technologies exceed the world level. It is noteworthy to mention that in the previous years in our country there were published many information materials and specialized editions of rather limited circulation in which technological achievements were described. That is why whenever one has a case of patent infringement suit or just a validity search it is extremely advisable to analyse the Russia published sources. Our experts have been collecting such materials for years and they might be quite successfully used in our investigations. The work includes preparation of an expert's opinion.

II. Heuristic search

There are cases of more complicated patent searches in the data bases with the use of special information retrieval systems that utilise sophisticated mathematical models for determination of the best terms suitable to retrieve the pertinent documents in a certain data base. In such searches the system selects for the user those search terms (classification indexes, title words, names, etc) which should be taken for retrieval of pertinent information.

Our firm possesses such extremely powerful and sophisticated information retrieval systems which it uses for information searches and organization of data bases. These systems envisage automatic reformulation of the search statements and selection of the best terms for further stages of the search and development of search strategies on the basis of "relevance feedback". Utilized in these systems is the so called "Swets model" (structuring of the data base documents into classes of proximity to the relevant documents of the initial output) and multi-dimension analysis of document clasters.

Such instruments make it possible to solve special tasks, when the analyst has to compensate the lack of concrete descriptors by analyzing the co-occurrence of the words and watching the information links from the field of the art he knows to other areas. For example, with the help of said methods it is possible to identify patents obtained by the Russian inventors in foreign countries, including the cases when the documents do not bear specific designation of the genuine country of origin of the inventor (such things happen when the authors work for a long time in a foreign state or change their citizenship. In this case the patents obtained by the Russian scientists for foreign companies, in the USA for example, may point out the valuable technology owners remained Russia (one has only to trace back the inventors' previous activities at home).

These systems are also used by our experts for facilitation of information search conducted through Internet in specialized data bases, maintained by Derwent, Questel Orbit, CAS, and many other less famous firms. The thing is that every concrete data base has its own merits and drawbacks and limitations. It is the task of the information searcher to compensate limitations of one data base by means of using its output for conducting the search in another data base. It is extremely helpful here to process the output of one system to a concrete query by a special information retrieval system in order to refine the search statements and select the best terms for further stages of the search in the same or another data base.