[smc-discuss] introducing a_മ്മ open data platform for Indic Machine Learning

[fennecfox at openmailbox.org]

Hi,
This email is a follow up on a topic I have shared in this group some days ago (http://lists.smc.org.in/pipermail/discuss-smc.org.in/2018-March/001421.html). 

We have been thinking about an open data platform for Indic Machine Learning. To contribute to the digital transition that's happening in India, we thought basic NLP problems need to be addressed. We are building a data platform, which is currently in its infant state. As the first step towards this, we have decided to work on Named Entity Recognition data collector interface. We have successfully deployed its alpha version. Because it's in alpha state, the application has a lot of issues to be addressed. So, currently, we're inviting only the people who we believe are happy to help make the platform better. We're also looking forward to collecting raw text, which still has difficulties in collecting, especially for the type of data which can be more useful in real life like news / other articles, discussions, office documents, casual communication etc. 

Please take a look at this documentation (https://github.com/a-mma/a-mma-documentation), where we discuss our thoughts on this. SMC has years of experience in community contributions. We believe in that experience and are happy to ask help and guidance on this. Looking forward to your thoughts.