Mining Structured Data From Recipes
API for extracting ingredients, quantities and other entities from unstructured text.
Read case study →Fact-checking is crucial for maintaining the health of the public debate and the high standards of our democracy. One of the most time-consuming parts of a fact checker’s workflow is finding “check-worthy” claims in the sea of political fluff.
The problem is only getting worse with political communication shifting from traditional channels to the much more fragmented communication through social media. If an AI could automatically monitor various channels and pinpoint “check-worthy” factual statements, fact checkers could spend drastically more time on verifying the claims themselves.
Bohemian AI partnered with the leading Czech fact-checking organization, Demagog.cz, and researchers from Palacky University Olomouc, to create a unique fact checking automation initiative, first of its kind in Central Europe.
We leveraged Demagog’s historical archives, manually annotated the data and trained a sentence-scoring model to predict which sentences are more likely to be “check-worthy” than others. To maximize our impact in 2020, we plan to adapt the system to other Slavic languages, starting with Slovak and Polish.
API for extracting ingredients, quantities and other entities from unstructured text.
Read case study →We're helping an NYC startup turn unstructured text in enterprises into value and insights.
Read case study →See how AI opens new revenue streams for law firms and legal tech companies alike.
Read case study →Redefining fact-checking workflow with artificial intelligence.
Read case study →