About the client
An innovator in Natural Language Understanding, our client develops semantic fingerprinting technology as the foundation for text mining and artificial intelligence software. Our client’s company, based in Vienna and San Francisco, addresses the challenges of filtering large amounts of unstructured text data, detecting topics in real-time on social media, searching in multiple languages across millions of documents, natural language processing, and text mining. Our client was named a 2016 IDC Innovator in the machine learning-based text analytics market as well as one of the 100 startups using Artificial Intelligence to transform industries by CB Insights.
Cortical came with the challenge
Inspired by the latest findings on how the human brain processes language, this Austria-based startup worked out a fundamentally new approach to mining large volumes of texts to create the first languageagnostic semantic engine. Fueled with hierarchical temporal memory (HTM) algorithms, this text mining software generates semantic fingerprints from any unstructured textual information, promising virtually unlimited use cases and a massive market opportunity.
Our client partnered with us to scale up their development team and bring to life their innovative semantic engine for text mining. Our expertise in REST, Spring, and Java was vital, as our client needed to develop a prototype that was capable of running complex meaning-based filtering, topic detection, and semantic search over huge volumes of unstructured text in real time.
Intellias developed the solution
This language-agnostic service lets our client process terabytes of text data by encoding the semantics of natural language elements into semantically grounded binary code. The resulting code is then further compared and analyzed with standardized metrics, offering great opportunities for text filtering, classification, and search. This service is advantageous for businesses that need to search text repositories in various languages, monitor incoming emails, or detect current topics on social media.
We also presented a prototype of NLP algorithms integrated into KNIME workflows using Java snippet nodes. This is a configurable pipeline that takes unstructured scientific texts as inputs and returns structured data as the output. Users can specify preprocessing settings and analyses to be run on an arbitrary number of topics. The output can then be visualized graphically on the resulting similarity index.
What did our client get?
- An NLP engine that operates on the level of semantics rather than keywords for filtering, classifying, clustering, and searching text
- A horizontally scalable solution that’s ready to support exponential growth of the user base
- A semantic fingerprinting technology to offer as a service on the global market
We achieved great results together
We helped Cortical
- Acquire new investors and raise $2 million in capital
- Implement RESTful APIs available on the Amazon Web Services marketplace to scale their solution with customized databases and applications
- Monitor customers’ reactions in real time through an intelligent filter that converts the Twitter firehose into a stream of semantic fingerprints
- Develop NLP algorithms in KNIME nodes to power a prototype of their NLP system