Tiring-Text
Words are more powerful than actions!
π΅οΈ Introduction
"Words are more powerful than actions", said the great speaker Chichnas. Help him spread his wise work by segregating the words he said.
π€ Problem Statement
For this challenge, your input will consist of multiple text transcript covering various domains such as math, news, technology wildlife, food, fitness, chess and programming. Given an abstract of a text transcript, your task is to identify which domain does it belong to and label it accordingly.
Understand with code! Here is the getting started code for you.π
πΎ Dataset
The training dataset train.csv
contains two columns text
and tag text
. The categories of the text are [math, news, tech, wildlife, food, fitness, chess, programming]
. The training dataset comprises of 79,376
text data points each corresponding to a specific category. The test dataset test.csv
contains just a single text
column. This comprises 19,844
data points for which tags have to be predicted.
π Files
Following files are available in the resources
section:
train.csv
- (79,376
samples) This CSV file contains two headers. The first header is thetext
which contains the transcripts and the second header is thetag
which contains the label corresponding to that text transcript.test.csv
- (19,844
samples) The file that will be used for actual evaluation for the leaderboard score. It only contains the text transcripts for which the tags have to be predicted.
π Submission
- Prepare a CSV containing only the header
tag
which contains the predicted category of the corresponding text transcript in thetest.csv
from the above-specified list of categories with name assubmission.csv
. - Sample submission format available at
sample_submission.csv
in the resources section.
π Evaluation Criteria
During the evaluation, F1 score will be used to test the efficiency of the model where,
π± Contact
Notebooks
0
|
0
|