Contex2Vec mines/groups words based on the context used.
Like the famous Word2vec it finds the similarity of words based on context and uses machine learning techniques to generate context vector. Read More in our blog and Take a look at words extracted in US Patent data
Natural Text Patent Similarity Search uses Machine Learning algorithms to identify similar phrases/lines between patent descriptions.
This search provides results from US Patents registered between Jan 2016 to May 2016. Older patent data and other countries patents will be available soon. Read more in our blog and Take a look at Patent Search Demo
Pattern rules learned from Machine Learning algorithms used to correct the grammar. Raw text is fed to the algorithm to generate the patterns. Rules for any language or any sequencing pattern can be mined using our algorithms.
Created using open source technologies. Used Python, MongoDB, RocksDB, Tornado, Ngnix running on Ubuntu. Big datasets handled using custom in-house developed MapReduce algorithms