Here's what we cover in this article. Click the links to jump to the relevant section:

A technology cluster refers to a grouping of patent families relating to the same technical area. The clustering tool groups patent families with similar characteristics into technology clusters, and will do so based on the patent families that have been included in your report. All families included in your report will be clustered.

Lets take a look at how clustering is applied to a report containing multiple organisations:

Clustering can be useful in understanding:

Clustering is done by creating similarity matrices based on patent meta data. Clustering involves no human intervention or hard coded categories.

The platform uses meta-data available to create technology clusters that are as accurate as possible. Machine learning plays a role here as the algorithms identify the technology domain of the patents and give different weights to different factors (e.g. codes tend to be poor at clustering software). Meta data used to create clusters includes:

How are cluster names determined?

Cluster names are machine generated, by reference to the title and abstract. Clusters are given a name that most closely describes all patent families in the Cluster using text summarisation, and natural language processing (NLP) techniques.

The clustering and naming algorithms are separate, ensuring that there is no possibility of a self-fulfilling prophecy in the clustering results. If the clustering were based on the occurrence of a certain phrase then it would bias that cluster towards containing only patents that used the phrase, and not other closely related technologies irrespective of words, skewing the clustering results.

How do you determine a cut off for the number of clusters in report as this could be endless?

The platform will present a maximum of sixteen technology clusters. If a group of patent families creates more than sixteen clusters, the clustering tool will group the remaining families into ‘miscellaneous’.

If relevant to your reporting, this miscellaneous cluster can then be broken down into further clusters for your analysis.

Can I cluster to a selected portfolio to see how others compare?

Yes absolutely. You can choose to cluster one portfolio first and see how others read onto that. The 'unrelated' cluster will show the number of patent families that do not fit into any of the technology clusters in the report i.e. clusters generated by the selected ‘target’ portfolio.

e.g. If the platform clustered company X and you are benchmarking company Y and Z: unrelated shows the number of patent families in Y and Z's portfolio that don’t cross over, or fit into, any of X’s clusters.

Please note that when applying classifiers to a report, you can choose to capture those patent families that are not a match for the classifiers. In this instance, the 'unrelated' cluster will show the number of patent families that do not fit into any of the technology classifiers in the report.

You may choose to cluster all of the information in your report as we've just seen, but it is also possible to group patent families into technology clusters formed by a target organisation. This means clustering the target organisation's portfolio first, and seeing how the other organisation's in the report read into these technology areas.

Clustering to a target organisation or portfolio is useful to:

Click here to learn more about this approach.

2. Search for the portfolios you want to add to your report. Click to add to your report basket:

3. Once you've added all required portfolios to the report basket, select 'Next' without applying any classifiers.

4. On the next screen your clustering options are selected. Select 'Run' to start your report:

Need help? Request support and we'll connect you with an expert.