We have twitter accounts to show some demos of NLP technologies.
These accounts get new tweets by news agencies' accounts and analyse the news texts.
Currently, we have two (English/Japanese) accounts.
Extract spatial expressions and estimate the latitude/longitude of them.
Extract temporal expressions and estimate the actual date/time of them.
Separate unsegmented texts into units (words or morphemes), and estimate the pronounciations of each unit.
For details, see KyTea's page.
Convert texts into Kanji (Chinese character) expressions.
For details, see Senn.
Your can send a command to analyse your own texts.
Commands with ✔ are implemented.
Command | Usage | LSTA_Linfer_en | LSTA_Linfer_jp |
---|---|---|---|
SNEL | @account SNEL [target text] | ✔ | |
TNEL | @account TNEL [target text] | ✔ | |
Word Segmentation | @account KYTEA [target text] You can use "KyTea" instead of "KYTEA" |
✔ | |
Kana Kanji Conversion | @account SENN [target text] | ✔ |
The map data used by LSTA-Linfer-en is made with Natural Earth. Free vector and raster map data @ naturalearthdata.com.
The latitude/longitude DB used by LSTA-Linfer-en is made with GeoNames.