{"default": {"description": "\n\u201cUrdu Sentiment Corpus\u201d (USC) shares the dat of Urdu tweets for the sentiment analysis and polarity detection.\nThe dataset is consisting of tweets and overall, the dataset is comprising over 17, 185 tokens\nwith 52% records as positive, and 48 % records as negative.\n", "citation": "\n@inproceedings{khan2020usc,\n  title={Urdu Sentiment Corpus (v1.0): Linguistic Exploration and Visualization of Labeled Datasetfor Urdu Sentiment Analysis.},\n  author={Khan, Muhammad Yaseen and Nizami, Muhammad Suffian},\n  booktitle={2020 IEEE 2nd International Conference On Information Science & Communication Technology (ICISCT)},\n  pages={},\n  year={2020},\n  organization={IEEE}\n}\n", "homepage": "https://github.com/MuhammadYaseenKhan/Urdu-Sentiment-Corpus", "license": "", "features": {"sentence": {"dtype": "string", "id": null, "_type": "Value"}, "sentiment": {"num_classes": 3, "names": ["P", "N", "O"], "names_file": null, "id": null, "_type": "ClassLabel"}}, "post_processed": null, "supervised_keys": null, "builder_name": "urdu_sentiment_corpus", "config_name": "default", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 161190, "num_examples": 1000, "dataset_name": "urdu_sentiment_corpus"}}, "download_checksums": {"https://github.com/MuhammadYaseenKhan/Urdu-Sentiment-Corpus/archive/master.zip": {"num_bytes": 51583, "checksum": "793561fd1cd4931024f551c0ea4a09d159064c2536f36ac5fec225178ba6a9b2"}}, "download_size": 51583, "post_processing_size": null, "dataset_size": 161190, "size_in_bytes": 212773}}