Dariah Year Prediction Challenge

This is a simple challenge revolving around predicting the year in which a noisy, OCR-ed fragment text was written. A solution should take in a paragraph of text in the form of a string as input and return a date prediction formatted as a float.

The training split of the dataset consists of 677665 fragments of text along with their corresponding dates. Note that it might not be necessary to use all of the training data to devise a good solution!

Metrics: rmse, mean_absolute_error, mse

Dataset authors: Prof. Filip Graliński (et al.) (filipg@amu.edu.pl)

Challenge authors: Kacper Dudzic (kacper.dudzic@amu.edu.pl)

License: CC BY-NC 4.0