Bitext Platforms Overview

Bitext Deep Linguistic Analysis Platform

Bitext offers a multilingual platform to analyze & tag text at three levels:

• lexical
• syntactic and
• semantic.

The platform has been designed for enterprise use: one single API for any language, platform independent (runs on Linux, Windows, Android, iOS), maximum performance/throughput, minimum footprint and easy scalability. As for deployment, customized solutions are available both for cloud or on-premise, as well as through our NLP API platform.

Lexical Level and Lemmatization

At the lexical level, the main component is the lemmatizer, integrated with tools for decompounding or word segmentation (required by some languages to perform proper lemmatization).

The lemmatizer can be additionally packaged to cover a language analysis full pipeline, from sentence segmentation to full parsing, including tools like spell checking.

Both components of the lemmatizer, data and software, can be distributed integrated or separately. All these tools are available in 77 languages and 25 language variants.

Syntactic Level and Parsing

At the syntactic level, the parser is the main component. The parser analyzes the structure of the sentences in the text and is used for tasks like POS Tagging and Phrase Extraction. Additionally, it is used as the base component for various semantic level tasks like Named Entity Recognition (NER), Topic-Level Sentiment Analysis or Generation of Synthetic Text. We have developed parsers for 21 languages and are always adding new languages.

For a full list of services, at the lexical, syntactic and semantic levels, check our linguistic services.