Expand your knowledge base
iKVA can index very unstructured data from both internal and external sources and across multiple languages including documents, chat and video and bring all this information together in one place
Fast and accurate search
iKVA’s solutions integrate into your workflow and anticipate the information you need, meaning you don’t have to stop work to look for it. This reduces time spent searching as well as business risk since iKVA can provide critical information you may not even knew existed
Easy to deploy
iKVA’s solutions do not need clean taxonomies or meta-data – simply point iKVA’s data indexers at the source data in its rawest form and we do all the hard work.
Available as both SaaS and on premises solutions
What we do
iKVA’s underlying technology is based on innovative research from some of the best minds at Cambridge University’s Computer Laboratory.
Using advanced Machine Learning, NLP and Semantic technology iKVA’s solutions solve the problems of working with siloed unstructured data from almost any source including document archives, email, chat, and video. This allows you to search for and index relevant information easier, or iKVA can bring you relevant information without even having to look for it.
Features and benefits
Fast. Queries typically take less than 200ms enabling truly conversational interactions
Multilingual. Any language – no need for language experts or translators. You can even search in one language against a library created in another.
Any input. Queries can be keywords, APIs, full documents, PDFs, web pages… allowing greater result accuracy and relavancy.
Immersive. iKVA can proactively bring you information – providing additional data to help decision making before you even know you need it
Secure and Private. Source data is not stored by iKVA and no personal profiles are used. No GDPR compliance issues and you control your data
Data Efficient. iKVA doesn’t rely on accessing terabytes of archived data which means data transfer is very efficient, portable, and aids remote working
Research papers
-
Kvasir: Scalable Provision of Semantically Relevant Web Content on Big Data Framework
Liang Wang, Sotiris Tasoulis, Teemu Roos, Jussi Kangasharju
-
Kvasir: Seamless Integration of Latent Semantic Analysis-Based Content Provision into Web Browsing
Liang Wang, Sotiris Tasoulis, Teemu Roos, Jussi Kangasharju
-
Fast Nearest Neighbor Search Through Sparse Random Projections and Voting
Ville Hyvönen, Teemu Pitkänen, Sotiris Tasoulis, Elias Jääsaari, Risto Tuomainen, Liang Wang, Jukka Corander, Teemu Roos
-
Efficient Autotuning of Hyperparameters in Approximate Nearest Neighbor Search
Elias Jääsaari, Ville Hyvönen, Teemu Roos
Products
Enterprise search
It is easy to become overloaded from the many high volume information flows that exist on the Internet.
Read more
Intelligent assistant
Intelligent Assistant actively monitors interactions between people to proactively provide information and assistance when they need it.
Read more