-
Welcome to B4X forum!
B4X is a set of simple and powerful cross platform RAD tools:
- B4A (free) - Android development
- B4J (free) - Desktop and Server development
- B4i - iOS development
- B4R (free) - Arduino, ESP8266 and ESP32 development
New feature is coming: OpenNLP - Text analysis
- Thread starter Erel
- Start date
- Similar Threads Similar Threads
It can do all kinds of things that can be relevant when you want to programmatically analyze text: https://opennlp.apache.org/docs/1.9.3/manual/opennlp.html
Analyzing text is a very complicated task. Just the language detection demonstrated above, is a task that we can't solve without a framework such as OpenNLP.
Analyzing text is a very complicated task. Just the language detection demonstrated above, is a task that we can't solve without a framework such as OpenNLP.
Will "do" all the features of OpenNLP ?
tokenization, sentence segmentation, part-of-speech tagging, named entity extraction, chunking, parsing,language detection and coreference resolution
Named Entity extraction will be useful too..
ofcourse all features can help for example sentence segmentation will be useful for education apps (learning english, greek etc)
tokenization, sentence segmentation, part-of-speech tagging, named entity extraction, chunking, parsing,
Named Entity extraction will be useful too..
ofcourse all features can help for example sentence segmentation will be useful for education apps (learning english, greek etc)
I wonder where we will see this functionality show up as used in B4X. Is this something you needed for B4X Pleroma, Erel?
If only B4X with such features existed 20 years ago... We had to write NLP from scratch for a CAT (Computer Assisted Translation) and MT (Machine Translation) tool. The Fuzzy Matching engine was one of the best and fastest available at that time. I remember writing the "split" on sentences and words alone took us weeks to get it right for all languages. But the engine could also align/match sentences from two texts in different languages with 98% accuracy. It used our NLP to find which sentence was the best translation in the other text. This was very helpful to build Translation Memories that could feed the Machine Translation, which then assisted the human translators using the CAT tool.
The tool could extract texts from about 70 different file formats (from software source code files, over word processor documents and PDFs to powerpoints) and then rebuild the translations back to its original format preserving all the layout.
Fun times...
Align/Match tool:
CAT tool:
Alwaysbusy
The tool could extract texts from about 70 different file formats (from software source code files, over word processor documents and PDFs to powerpoints) and then rebuild the translations back to its original format preserving all the layout.
Fun times...
Align/Match tool:
CAT tool:
Alwaysbusy
No, though I might use it to extend that app at some point.
I wanted to push B4X boundaries and after testing several directions, decided on text analysis.
great addition after IoT, waiting........
My next project will depend on it: Medical Diagnosis using machine learning..
My next project will depend on it: Medical Diagnosis using machine learning..
All
Member
Dear Erel!
We are just trying to wrap these libraries for text analysis. But we were unable to find an exact guide on how to connect a very large library.
Just look, can you tell us this or is it suitable for you for text analysis?
Thanks!
We are just trying to wrap these libraries for text analysis. But we were unable to find an exact guide on how to connect a very large library.
Just look, can you tell us this or is it suitable for you for text analysis?
Thanks!
All
Member
Dear Erel!
Can your word processing functions work (customize them) to work with the Russian language? This is very important for us, since the Russian audience in the CIS countries is more than 300 million people, and a very rapidly developing sector of IT VTSZ and specialists.
Will Russian be supported?
thanks
Can your word processing functions work (customize them) to work with the Russian language? This is very important for us, since the Russian audience in the CIS countries is more than 300 million people, and a very rapidly developing sector of IT VTSZ and specialists.
Will Russian be supported?
thanks
I'm wrapping OpenNLP so it is not really "my" word processor 
It can handle any language however you need to have a "model" file that is relevant to your domain. You can also create one yourself. It is called training.
I see that someone has published models for Russian: https://github.com/livelace/opennlp-models/pkgs/container/opennlp-models
It can handle any language however you need to have a "model" file that is relevant to your domain. You can also create one yourself. It is called training.
I see that someone has published models for Russian: https://github.com/livelace/opennlp-models/pkgs/container/opennlp-models
When you release it, we are waiting...
First version released: https://www.b4x.com/android/forum/threads/nlp-opennlp-text-analysis.133883/