Natural Language Processing Text Classification Learning Series (2)

Answer 1: The average text length is 872 characters, with a minimum of 64 characters and a maximum of 7 125 characters, most of which are below 1000.

The corresponding relationships of labels in the data set are as follows: {'technology': 0,' stock': 1,' sports': 2,' entertainment': 3,' current affairs': 4,' society': 5,' education': 6,' finance': 7,' home':

Answer 2: It can be seen that "sports" and "stocks" account for the highest proportion, followed by "technology" and "entertainment", and the distribution of categories is not very balanced.

Answer 3: The maximum number of characters is 30 times per article, and the high-frequency characters are probably punctuation marks or stop words, which need to be filtered.

Homework in this chapter