Individuals and companies are snowed under with text. We all receive a constant barrage of emails and web postings, and we have access to vast amounts of data on the internet with only primitive keyword based tools to access it. There are numerous business opportunities in processing, indexing, analyzing, classifying and generating text automatically.
Scientio have several products that are targeted at handling and analyzing text.
We can index very large collections of documents so that you can group documents that are very similar, as in plagiarized versions or multiple revisions. We can do this more efficiently, in terms of processing time and memory, than any other vendor. Customers are using this system to handle millions of documents, finding near duplicates in milliseconds with conventional hardware. See ConceptMine
We can analyze text and classify it according to criteria that our systems can learn. A great example of this is sentiment analysis. A customer of ours is using a web service based on XmlMiner to analyze blogs for their sentiments towards various brand names and celebrities, creating a popularity measure that change daily. See XmlMiner.
Our software can be used to auto-summarize the ideas expressed in large collections of textual responses - from advertising campaigns, complaints or blogs. We're currently working with a customer on just such a system.
The most natural way for humans to interact with computers is via a conversation. Scientio are developing applications in this field in two different ways.
We have constructed a ChatBot, or Virtual Agent, that is based on conventional technology, but which we are extending using our concept mining technology. Our main design goals are to increase the range of correct responses while reducing the programming demand in creating the knowledge bases used by the system. See ScientioBot.
Coming from a totally different direction, but rapidly converging with the above, we have adapted our rule-based technology so that it auto-generates question and answer sessions. Using this, experts have a simple way to encapsulate knowledge of, for instance, legislations, best practice, trouble shooting guides or company standards into our rule sets. These can then be automatically used in a question and answer session, and our software works out from the rules which questions to ask first and it excludes questions that are irrelevant because of previous answers. See XmlMiner.
These facilities can be provided remotely via web services at our site or locally running on a rack in your data centre.
Talk to us about your requirements - it's free and we're 'startup friendly'!