search

SidneyFanFan / CNELL

Chinese Never Ending Language Learning
4 stars 1 forks source link

what CNELL was? #1

Open syw2014 opened 7 years ago

syw2014 commented 7 years ago

Hi, I am constructing my own knowledge graph, and found your project about chinese NELL, does it can extact entities and relationship from open IE, can you make a description for your project, thx!

SidneyFanFan commented 7 years ago

Hi Jerry,

Thank you for your interests and sorry for the delay. C-NELL has been shut down for a while, and I'm no longer in the Lab. The version on Github is just a demo backup. If you are interested, you can follow "src/cnell/experiments" to check those Dryruns to get an idea, but unfortunately I am not allowed to give you relation pattern configurations.

C-NELL is a middleware of Knowledge Works 2 years ago. The frontware crawler (which is not included in this project) gathers daily news from the Internet, and export JSON format data. C-NELL makes use of the news data, extracts entities (use Stanford NLP), and finally extracts & clusters entity relationships. The backware makes use of the entity taxonomy for entity relationship refinements.

Hope this helps!

Best, Siqi

On Tue, May 9, 2017 at 7:29 PM, Jerry.Shi notifications@github.com wrote:

Hi, I am constructing my own knowledge graph, and found your project about chinese NELL, does it can extact entities and relationship from open IE, can you make a description for your project, thx!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/SidneyFanFan/CNELL/issues/1, or mute the thread https://github.com/notifications/unsubscribe-auth/AGbEiyGJB_BLG_kE7hei30q_m_MOyrJkks5r4SEZgaJpZM4NWGzr .

--

Sidney(Siqi) Fan | Graduate Student T: (+1) 213-716-5645 | E: SiqiFan@usc.edu or SidneyFan716@gmail.com

Study at Viterbi Engineering School, University of Southern California Los Angeles, CA 90089-0911

syw2014 commented 7 years ago

@SidneyFanFan Thank you for your interpretation， now I am clear about your project, I am going to have a try and may find some ideas about entities and relationships extraction. Thanks again.