T: +1 347-480-1610
Some of our products are available for evaluation. Please contact us to inquire.
+1 347-480-1610Morphological Analyzer is a software component capable of detecting morphemes in a piece of text. English text is commonly pre-processed before it is indexed and one of the common pre-processing steps is stemming. Stemming is a process through which suffixes are removed and words are converted to their stems. For example, the word "caring" might be stemmed to "car". Stemming rules for English language are simple. Several known algorithms have been published and their implementations are freely available. Many other languages have more complex morphosyntactic characteristics (e.g. different suffixes or prefixes can be used with a single word depending on the tense, gender, number, case, etc.) and thus more complex rules for their stemming are needed. In most cases there are no publicly known algorithms and/or no available stemming products. Our Morphological Analyzer uses Statistical Natural Language Processing (NLP) to learn about the language morphosyntactic structure and uses that knowledge to detect morphemes. It works exceptionally well for highly-inflected languages - languages whose words tend to have lots of affixes, such as Polish, Czech, Slovak, Croatian, Serbian, etc.
Sematext Group, Inc. 540 President St. 3rd Floor Brooklyn, NY 11215, USA
E: info@sematext.com T: +1 347-480-1610 F: +1 718-679-9190
© Copyright 2013 Sematext Group, Inc.
All Rights Reserved
