In Japanese patent law, F-term is a system for classifying Japanese patent documents according to the technical features of the inventions described in them. It is not a replacement for the International Patent Classification (IPC) or other patent classifications, but complements other systems by providing a means for searching documents from different viewpoints. A symbol attached to a patent document, indicating that the invention disclosed in the document has a particular technical feature, is also called an F-term.
Contents |
The F-term system was developed in 1987 by the Japan Patent Office (JPO) and is maintained by the same body. The F-term system does not have any legal foundation, whereas the IPC system was agreed by the Strasbourg Agreement of 1971.
The F-term system is used by examiners in the JPO who give appropriate F-terms, together with IPC categories, to each patent document published by the JPO.
As with other patent classification systems, F-terms attached to a patent document do not affect the technical scope of the patent right, they simply serve as an index for patent documents.
The IPC system is similar to the scientific classification of organisms; both being hierarchical classification systems based on a single viewpoint. In contrast, the F-term system is like a picture book that presents various views such as "lives in rainforests" or "creatures having wings". The IPC system could be said to be systematic, while the F-term system is intuitive.
The F-term system was developed on the principle that computers will be employed for searching documents using the system. This is evident from the fact that F-terms did not appear on patent documents printed on paper before 2000, only being present in the electronic records up to that date. In particular, the system assumes the availability of set operations such as union or intersection, because all applicable F-terms are applied to each document. (See the fictional example below.) This contrasts with the original basic idea of the IPC that each patent document falls into one subgroup only and documents are stored in book stacks.
The F-term classification system consists of themes and terms.
The coverage area of the IPC is divided into approximately 2900 themes with each theme spanning a range of IPC subgroups. A theme is identified by the title describing the range or the theme code which consists of five digits allocated uniquely to each theme; for example, a theme spanning IPC range A01K 87/00
–87/06
is identified by its title "Fishing rods" and its theme code is 2B019
. Some themes only span one IPC subgroup, such as theme 2F011
, "Tape measures" which covers IPC G01B 3/10
.
Each theme has a number of viewpoints and each document is classified repeatedly for each viewpoint in the respective theme. For example, theme 2F011
"Tape measures" has the following four viewpoints: tape measures, housings, driving of tape measures and accessories.
From each viewpoint, documents are classified into several groups and labeled with a four-digit code called the term or F-term. For example, from the viewpoint "tape measures", which focuses on measuring tape itself rather than winding mechanism or housing of tape, all measuring tape documents are classified into eight groups such as AA02
"scales for special applications", AA05
"tapes with cores containing synthetic resins", etc.
All terms concerning a theme are put into a tabular form called an F-term list, which are available online, e.g., 2B019 and 2F011. (Some descriptions of F-terms may seem strange because of incorrect translation from the original Japanese.)
Below is a fictional example of classifying a patent application within the F-term system. The fictional example of a cooking recipe database will be used as this avoids the need for any subject-specific knowledge.
The fictional IPC-style classification of recipes may be as follows:
X99Y 1/00 Dishes X99Y 1/02 . Meat dishes X99Y 1/04 . . comprising beef X99Y 1/06 . . comprising pork X99Y 1/08 . . comprising lamb X99Y 1/10 . comprising seafood X99Y 1/12 . comprising vegetables
This classification of recipes is based on a single viewpoint—the main ingredient of the dishes. Hence we can browse all recipes of meat dishes by selecting recipes labeled X99Y 1/02
, X99Y 1/04
and X99Y 1/06
, and likewise we can browse all recipes of beef dishes by selecting recipes labeled X99Y 1/04
.
However, the system does not allow the user to find all Chinese recipes, or all halaal lamb recipes, because the system is based on a single viewpoint. This is solved by the F-term system by the use of multiple viewpoints.
A fictional F-term classification of recipes may be as follows:
9Z999 Dishes (X99Y 1/00—1/12) AA INGREDIENTS AA11 Meat AA12 . Beef AA13 . Pork AA14 . Lamb AA21 Seafood AA22 . Fish AA31 Vegetable BB CUISINE BB41 Asian BB42 . Chinese BB43 . . Cantonese BB44 . Turkey BB51 European BB52 . French BB53 . Italian CC COOKING CC11 Boiled CC21 Fried CC31 Roasted CC41 Steamed DD SPECIAL DD11 Halaal DD12 Vegetarian
9Z999
identifies an F-term sheet, or theme. X99Y 1/00--1/12
in the parentheses indicates that the sheet covers the IPC range X99Y 1/00
–1/12
. AA
, BB
, or CC
identify the different viewpoint of the theme and the combination of two letters and two numbers AA11
etc. identifies features viewed from a specific viewpoint. This 4-symbol construction is known as an F-term. We can therefore find all Chinese recipes by selecting recipes having an F-term of BB42
.
Since all applicable F-terms are applied to a given document a boolean-logic search can be used to find combinations of features—for example to find halaal lamb recipes a search can be made for recipes having both AA14
and DD11
F-terms. The latter search is conducted by calculating the intersection of set AA14
and DD11
.
Because some F-term sheets of practical use are fairly large and complicated like 2H171 and 3C007, examiners sometimes drop applicable F-terms by mistake when they classify documents. In addition, some documents are classified into inappropriate IPC and therefore do not have any appropriate F-terms.
Taking these errors into account, a "defensive" search is conducted by
at the expense of additional time for browsing more documents.