Re: Implementation of Chatbot & Adapter project
On Mon 7. May 2018 at 20:27 Dingfan Zhaowrote: > 1. The first one is about the integration issue. Currently there are two > choices about the integration of the Chatbot with the main Fineract > project. The first one is to fork the whole Fineract project to the current > code base and develop it inside the huge code base. The other choice is > that Chatbot only communicate with main project using RESTful API provided > by the Mifos. > I tend to prefer micro repositories, but there are arguments in both directions. If you would like a GitHub repo, let us know. I can create one or show another committer how to do it. Best Regards, Myrle
Implementation of Chatbot & Adapter project
Dear mentors, I summarised the ideas that we come up with after discussion with Mr Alek in this email. I appreciate if anyone can provide any advice on the choice of implementation of the project. 1. The first one is about the integration issue. Currently there are two choices about the integration of the Chatbot with the main Fineract project. The first one is to fork the whole Fineract project to the current code base and develop it inside the huge code base. The other choice is that Chatbot only communicate with main project using RESTful API provided by the Mifos. 2. The choice of NLP framework used for Chatbot. There may be license issue so we may use NLP framework of Apache to avoid such issues. Apache OpenNLP(https://opennlp.apache.org/) or DL4J (https://deeplearning4j.org/nlp) might be good choices. If you have any suggestions on how to choose between those choices, please tell me your ideas. Thanks a lot. Regards, Dingfan