Syn-Tactic uses a variety of Language Technology components (NLP) to create intelligent solutions like
name recognition (NER, named entity recognition)
text mining
sentiment analysis
terminology extraction
segmentation of translation memories
Syn-Tactic uses a parser augmented with statistic analysis to quickly and efficiently extract domain terminology from any volume of documentation. The technology is easily portable to different languages. The main advantage of this method over purely statistical approaches is that it produces less noise and does not need large quantities of text to produce reliable results. The same technology can be used to segment sentences from translation memories into phrases, to increase reusability.
The Syn-Tactic document classiffication service (DocClassifier) will process text versions of scanned and photographed documents to:
determine the document type (e.g. invoice, warranty, insurance etc.)