Dataset for annotation
Files and their description
combined_all_stats_csv
- This is the CSV File containing statistics for the patent dataset
can be downloaded from the Git Repo.
- What are the statistics it shows?
This file contains columns such as filename, total publications, total positive samples, total negative samples, and total neutral samples per every week of the year from 2010-2020.
- This is the CSV File containing statistics for the patent dataset
combined_neut_data_csv
- This is the CSV File, that contains details regarding neutral samples
Can be downloaded from here
- What data this file contains?
This file contains columns such as publication_number, patent_title, appl_type, and neutral_text for each patents.
- This is the CSV File, that contains details regarding neutral samples
combined_pos_data_csv
- This is the CSV File, that contains details regarding positive samples
Can be downloaded from here.
- What data this file contains?
This file contains columns such as publication_number, patent_title, appl_type, and positive_text for each patent.
- This is the CSV File, that contains details regarding positive samples
combined_neg_data_csv
- This is the CSV File, that contains details regarding negative samples
Can be downloaded from here
- What data this file contains?
This file contains columns such as publication_number, patent_title, appl_type, and negative_text for each patent.
- This is the CSV File, that contains details regarding negative samples
combined_data_150k
- This is the CSV File, that contains 150k cleaned data samples(50k from each pos, neg, and neut) required to train models.
Can be downloaded from here
- What data this file contains?
This file contains two columns text and target. Text can be neutral, positive, or negative.
Target can be label 0, 1, 2 for neutral, positive, and negative respectively.
- This is the CSV File, that contains 150k cleaned data samples(50k from each pos, neg, and neut) required to train models.
- Downloads last month
- 5