Analysis of IPC classification codes frequency in patents concerning "in situ" soil remediation technologies (CPC code B09C2101)

Published: 11 January 2023| Version 1 | DOI: 10.17632/y3jk3twbgn.1
Contributor:
Riccardo Priore

Description

As already explained in our previous publication (Analysis of IPC classification codes frequency in patents concerning “in situ” remediation technologies), our Patlib Centre is involved in the context of Horizon2020 funded projects aimed at soil remediation. Therefore, we are especially interested in identifying the trends of the relevant soil remediation technologies that score the highest frequency of citation within the patent dataset. By means of an iterative procedure based on the t-distributed stochastic neighbor embedding (tSNE) algorithm, the patent families can be included in clusters depending on presence/absence of IPC subgroups or main groups or subclasses, chosen among those most abundant in the dataset. Therefore, following three rounds of clustering, about 90% of the initial ‘dataset 1’ (including 1632 simple families - as defined by the European Patent Office) become "unpacked" and clustered. Further assessments based on the patent bibliographic data can be performed, the essential advantage being that the technical content of each cluster is homogeneous, so that the results are aggregated depending on the pattern of specific IPC codes common to each cluster. Within the abovementioned 1632 simple families of the ‘dataset 1’ initially analyzed, there is evidence of 290 simple families, each one being characterized by the assignment of the CPC classification code B09C2101, specifically dealing with “in situ” remediation of soil. Apart from these 290 simple families, a specific search query made on Patstat online, aimed at selecting those patent documents to whom the CPC code B09C2101 has been assigned, has revealed the presence of additional 2521 simple patent families (included in the ‘dataset 2’ and not included in ‘dataset 1’) that underwent a substantially similar analysis focusing on the IPC classification codes’ distribution already described as far as ‘dataset 1’ is concerned. From the observation that in ‘dataset 1’ IPC classification codes such as B09C1/08 and B09C1/10 are among those most frequently assigned to the patent families, further analysis of the patent families of ‘dataset 2’ has been focused on these two IPC classification codes. In fact, these are the most representative also in ‘dataset 2’. Therefore, the analysis based on codes different from the IPC B09C1/00 (reclamation of contaminated soil), which is significantly represented in both datasets although not referring to a specific soil decontamination technique, reveals that for the time being soil decontamination often entails methodologies based on electrochemical remediation as well as microbiological techniques, for example based on the use of enzymes. The information that can be gathered from both datasets is comprehensive being based on data arguable from 4153 patent families.

Files

Steps to reproduce

The procedure aimed at retrieving the relevant patent documents dealing with "IN SITU" reclamation techniques specifically dedicated to contaminated soils is extensively explained in the Methodology section. The reproducibility implicates the availability of patent databases such as Patstat online (managed by the European Patent Office - https://www.epo.org/searching-for-patents/business/patstat.html) and Orbit Intelligence (provided by Questel: https://www.questel.com/orbit-software-suite/orbit-intelligence/). Further elaboration with the tSNE algorithm is based on elaboration of the data by means of the Python programming language. Scripts that allow for clustering of the patent families can be retrieved from an earlier publication (Python elaboration of Patstat and Orbit data, Published: 11 March 2021|Version 1|DOI:10.17632/gfnhp8r52y.1). Upon following the instructions written in the Methodology section and using the datasets' information included in the Data files' section, the reproducibility of our results is straightforward. The main conclusion of the whole analysis of the dataset specifically dealing with “in situ” remediation of soil is that the frequency of assignment of IPC classification codes such as B09C1/00, B09C1/08 and B09C1/10 is prevalent. This finding confirms the trend already detectable when analyzing the “in situ” remediation information arguable from a dataset analyzed earlier (Analysis of IPC classification codes frequency in patents concerning "in situ" remediation technologies), although not specifically referring to soil decontamination, as instead occurs in the case of the present analysis. Although the B09C1/00 IPC code is not much informative as far as the specific remediation technique is concerned, the specificity of B09C1/08 and B09C1/10 is on the other hand meaningful, the former code specifically referring to electrochemical based remediation, the latter specifically referring to microbiological remediation.

Institutions

Consorzio per l'AREA di Ricerca Scientifica e Tecnologica di Trieste

Categories

Artificial Intelligence, Patent, Business Intelligence, Patent Management, Patent Valuation, Patent Classification, European Patent Office

License