The input table is a collection of categorized news articles in Simplified Chinese, from news.data.
To create the input table, use this statement:
CREATE FACT TABLE news_test ( doc_id VARCHAR(10), content TEXT, category VARCHAR(8) ) DISTRIBUTE BY HASH(doc_id);
To load the input table with data, use this command:
ncluster_loader -h queen_ip_address -U username -w password news_test news_test.data;
NaiveBayesTextClassifierPredict Chinese Example Test Data news_test