Summary of Date Predictions

This interactive shows the results of a linear regression model to predict the publication date (or date written) of book-length works published between 1760 and 1990. Many of the texts belong to market genres such as horror/gothic, mystery/crime and science fiction. Each point represents a text, and its position along the x-y axis represents the predicted year and the actual year. The orange line projected through this visualization represents a one-to-one ratio of the predicted value and the metadata value. In other words, the closer a point is to this line, the more accurate the model predicted its date. Similarly, a point farther away from the line represents a less accurate prediction or, in some cases, an inaccurate date in the original metadata.

*Metadata fields are automatically generated from the data found in this repository. Special thanks to Ted Underwood for publishing this repository in connection with his 2016 article "The Lifecycles of Genres."