Teqmine’s mission is to make technology and IPR intelligence easy-to-use and affordable. To this end, we offer to our clients solutions based on cutting-edge artificial intelligence and natural language processing, as well as the benefits of advanced automation and easy-to-use Software-as-a-Service.
Our algorithms read mercilessly through all full-text patent descriptions – ranging from few kilobytes to hundreds of megabytes without missing a word or single patent. With machine-learning techniques, we generate high-complexity probability models that taken into different features of all patents and client data. With our solutions you are guaranteed not to miss any of the tens of thousands of new patent applications and grants that are published every month.
What’s different with Teqmine’s AI approach
There are several “smart patent search engines” on the market, but we are not aware of another one that would employ probability models based on minute reading of millions of patent full-text descriptions and their subsequent re-classification with artificial intelligence. The advantage of our approach is that AI takes care on your behalf of the basic search and classification work, and eliminates the redundant work and “noise” in the data.
In contrasts, traditional patent data bases are based on “indexed search”, where you sort information based on key-words or technology classes. This suffers from two known problems: Either you are flooded with too much information, or you suffer from the “street light” effect, where people search where it is easiest and most comfortable for them.
“Smart searches” offered by some of our competitors are typically “truncated” trained or untrained probability models that are based only on abstracts or technology classes, offering narrow and skewed window in patent intelligence modelling.
Teqmine solutions enable also true scaleability and cumulative knowledge bases. Our Patent Monitoring solutions capture every month from new publications only those that really matter, and you have instantly updated graphics and analysis on your patent landscape.
Versatile Big Data and Visualization
Our analysis generates extensive statistical data on technical documents and patents, which is turned into standardized intelligence or for custom needs.
Our custom solutions can re-create a range of new analysis possibilities for you by combining your data and our analysis, and cast light any aspect of new technology and business, such as predictive analytics, IPR portfolio comparison and matching, etc. We can also offer you a range of data visualization options to impress your clients or to convey effectively the key insights to your management.
Complete publication level patent record data for our basic services is obtained directly from issuing offices and (USPTO/EPO/PCT) re-package it for optimal data mining. Additionally, we have at our disposal a broad range of other raw patent data.
Our probability models for Patent Similarity use the latest available English language publications from USPTO, EPO and PCT, and data since 1990 and are updated monthly.
Because all data is hosted on our own servers, Teqmine can easily create customized data for special analysis, such as corporate technology and business portfolio or industry level analysis.
Most importantly, we can mix client confidiential data with the raw data. This can include e.g. invention disclosures, scientific manuscripts, research prosals or whole databases of technical information.