Methodology

To conduct the Canadian-focused analysis from this chapter, a Canadian subset was created from the WIPO Mining Database. The dataset consists of 3,026 INPADOC patent families with a Canadian assignee that filed a patent between 1990 and 2015. The methodology used in this chapter deviates slightly from the one used more broadly in Reference Daly, Valacchi and RaffoDaly et al. (2019) due to the availability of different tools. As such, the members of a patent family were not combined using the first family ID but rather Clarivate’s Derwent Innovation INPADOC Family ID. Utility models and design patents were removed from the dataset, considering that CIPO does not offer utility models and because Canadian industrial design data is not included in the European Patent Office (EPO) PATSTAT database which was used to generate the data for this report.

The Clarivate’s Derwent Innovation database was also used to create the patent landscape maps in this chapter. The approach taken to produce these maps involved a matching exercise to link the patent families extracted from the EPO PATSTAT database to the Derwent Innovation database. Once this exercise was completed, the publication numbers associated with the Derwent Innovation patent families were then loaded into the Derwent Innovation database to produce the landscape maps.

In order to better understand a country’s strengths in each mining category, the Relative Specialization Index (RSI) was used. To ensure consistency throughout this chapter, the Canadian patent family data used to calculate the RSI is based on the methodology using EPO PATSTAT data to construct patent families using first family ID.

The formula used to calculate the RSI is as follows:

Numerator

The sum of patent families by Canadian assignees in a specific mining category is divided by the sum of patent families in the world in the same mining category.

Denominator

The sum of patent families by Canadian assignees in the mining dataset (all mining categories) divided by the sum of patent families for the world in the mining dataset (all mining categories).

$RSI=ln[\left(\frac{{\sum }_{1990}^{2015}{P}_{Can,MiningCat}}{{\sum }_{1990}^{2015}{P}_{World,MiningCat}}\right)/\left(\frac{{\sum }_{1990}^{2015}{P}_{Can,Mining}}{{\sum }_{1990}^{2015}{P}_{World,Mining}}\right)]$

, where P represents patent families.

An RSI greater than 0 suggests that Canadian assignees have a relative specialization in the particular mining category, while an RSI lower than 1 suggests the opposite. An RSI equal to 0 indicates that an economy’s share of patents in that particular mining category equals its share in all mining categories.