Christopher L. Benson

Quantitative Determination of Technological Improvement from Patent Data

The results in this paper establish that information contained in patents in a technological domain is strongly correlated with the rate of technological progress in that domain. The importance of patents in a domain, the recency of patents in a domain and the immediacy of patents in a domain are all strongly correlated with increases in the rate of performance improvement in the domain of interest. A patent metric that combines both importance and immediacy is not only highly correlated (r = 0.76, p = 2.6*10-6) with the performance improvement rate but the correlation is also very robust to domain selection and appears to have good predictive power for more than ten years into the future. Linear regressions with all three causal concepts indicate realistic value in practical use to estimate the important performance improvement rate of a technological domain.

Using Enhanced Patent Data for Future-Oriented Technology Analysis

Patents represent one of the most complete sources of information related to technological change, and they also contain much detailed information not available anywhere else. Thus, patents are the ‘big data’ source most closely related to future-oriented technology analysis (FTA). Not surprisingly, therefore, there is very significant practical and academic use of the patent database for understanding past technical change and attempting to forecast future change. This paper summarizes several new methods and demonstrates their combined effectiveness in establishing a cutting-edge capability for patent study not previously available. This capability can be stated as a link between the information in patents and the dynamics of technological change. The demonstrated capability relies upon the use of a database containing the rates of improvement for various technologies. We also specify the term we use for the analysed units of technology: a technological domain is a set of artefacts that meets a specific generic function while utilizing a specific set of engineering and scientific knowledge. This definition is unambiguous enough so technological domains can be linked with progress rates and are sufficiently flexible to accommodate the large scale and complexity of the patent database. The existence of an improvement rate database and its quality is a critical foundation for this paper. Establishing the overall capability also involves relating the rate of improvement of a technological domain to the patents in that domain. We show that a recently developed method called the classification overlap method (COM) provides a reliable and largely automated way to break the patent database into understandable technological domains where progress can be measured. In this paper, we show how this method overcomes the third limitation of the patent database. The major conclusion of the paper is that there is now an overall objective method named Patent Technology Rate Indicator (PTRI) for using just patent data to reliably estimate the rate of technological progress in a technological domain. Thus, the first link between the patent database information and the dynamics of technological change is now firmly established; robustness and back-casting tests have shown that the assertion of reliability is meaningful and that the estimate has predictive value. We demonstrate the key methodology of new elements (use of COM and rate estimation from the selected patent sets) for 15 technologies that some have thought have possible future importance. The 15 cases also demonstrate the usefulness of the overall method by estimating technological improvement rates that are significantly different for this group of technologies.

Correction: Quantitative Determination of Technological Improvement from Patent Data.

Fig 1 appears incorrectly in the published article. Please see the correct Fig 1 and its legend here. Fig 1 Technological Improvement Rates vs Simple Patent Count (A), ratio of patents with greater than 20 citations (B), and average number of forward citations within 3 years of publication (C); the Pearson correlation coefficient (cp), the null hypothesis acceptance ... There is an error in the second sentence of the third paragraph in the Results section. The correct sentence is: There seems to be a slight visual trend in the figure, the Pearson correlation is a moderate 0.38 and the p-value is slightly lower than is generally accepted for statistical significance, at 0.043. Table 4 appears incorrectly in the published article. Please see the correct Table 4 and its legend here. Table 4 Least Squares Linear Regression Models for Predicting Technological Improvement Rates with R2 shown for each model and the coefficients shown for each metric included in the model and its p value.