Prediction Error on a text based dataset.
Hello everyone, I am new to RapidMiner and I have been stuck on this problem for many hours and I need help.
I am using the movie dataset from UCI
It contains different datasets of movies, actors in those movies and the directors of those movies. The main file is the movies dataset. It has the name of the Movies, Genre, Place it was shot, Release Date, Director and Producer. Except for date all of these are text.
I have tried to use the "Auto Model" of RapidMiner for Predictor of Genres of movies. Basically any model I use it gives an error and I get this "The input data did not have any attributes left after preprocessing. One cause could be that the data only has text columns which have not been processed. Please check your data and settings. "
I searched on the internet but could not find an error like this.
I am using the movie dataset from UCI
http://archive.ics.uci.edu/ml/datasets /MovieIt contains different datasets of movies, actors in those movies and the directors of those movies. The main file is the movies dataset. It has the name of the Movies, Genre, Place it was shot, Release Date, Director and Producer. Except for date all of these are text.
I have tried to use the "Auto Model" of RapidMiner for Predictor of Genres of movies. Basically any model I use it gives an error and I get this "The input data did not have any attributes left after preprocessing. One cause could be that the data only has text columns which have not been processed. Please check your data and settings. "
I searched on the internet but could not find an error like this.
Find more posts tagged with
Sort by:
1 - 1 of
11
Telcontar120
New Altair Community Member
Did you do text preprocessing prior to Auto Model, or did you select the option for Auto Model to do the text processing for you? Either way it needs to be transformed into a word vector before conventional ML algorithms will be able to create predictions.