mchav/abantu

A collection of resources I'm compiling to learn about African languages. # abantu A collection of resources I'm compiling to learn about African languages. ## Goal I'd like to use this dataset for a mixture of exploration and preservation. Language is rich and diverse and there is probably a lot you cannot learn from looking at words outside their proper contexts. But documenting language is an important part of cultural preservation. The rough plan for the data is to: * Collect lists – Gather Swadesh (or similar) lists. * Normalize forms – Transcribe to a consistent phonemic form (or IPA) to more accurately compare languages with different romanizations. * Align cognates – Group words that are likely inherited from the same root. * Compute similarity – Either by hand (percentage of cognates) or with custom scripts tools. * Visualize – Build dendrograms or distance heatmaps to see clusters and outliers. * Interpret – Cross-check anomalies (unexpected borrowings, divergent semantics) against historical or sociolinguistic context. If you have any ideas or resources we could draw from please submit an issue.