Can any one suggest me the algorithm or available tools also methodology? Keeping in view that Sindhi is right to left writing language like South and South east Asian Languages, For Example, Persian, Urdu, Siraiki and Panjabi etc.
Chris Biemann
I would recommend training a sequence tagger such as a CRF (Conditional Random Field) on manually labeled data.
You can use WebAnno https://webanno.github.io/webanno/ to annotate right-to-left languages with NER spans.
Then you define your features for the sequence tagger and train a model. Here is a tool that does NER for German and is Open Source - you might consider swapping out the German-specific parts with Sindhi-specific data. https://github.com/tudarmstadt-lt/GermaNER
1 votes
1 thanks
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Writing
More Wazir Ali's questions See All
Similar questions and discussions