metadata

language:
  - 'no'
  - nb
  - nn
license: cc-by-nc-4.0
size_categories:
  - 10K<n<100K
task_categories:
  - token-classification
pretty_name: NoReC TSA
dataset_info:
  - config_name: default
    features:
      - name: idx
        dtype: string
      - name: tokens
        sequence: string
      - name: tsa_tags
        sequence: string
    splits:
      - name: train
        num_bytes: 2296476
        num_examples: 8634
      - name: validation
        num_bytes: 411562
        num_examples: 1531
      - name: test
        num_bytes: 346288
        num_examples: 1272
    download_size: 899078
    dataset_size: 3054326
  - config_name: intensity
    features:
      - name: idx
        dtype: string
      - name: tokens
        sequence: string
      - name: tsa_tags
        sequence: string
    splits:
      - name: train
        num_bytes: 2316306
        num_examples: 8634
      - name: validation
        num_bytes: 414972
        num_examples: 1531
      - name: test
        num_bytes: 349228
        num_examples: 1272
    download_size: 902284
    dataset_size: 3080506
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
      - split: test
        path: data/test-*
  - config_name: intensity
    data_files:
      - split: train
        path: intensity/train-*
      - split: validation
        path: intensity/validation-*
      - split: test
        path: intensity/test-*

Dataset Card for NoReC TSA

This is a dataset for Targeted Sentiment Analysis (TSA) in Norwegian, derived from the the fine-grained annotations of NNoReC_fine. The dataset contains tokenized Norwegian sentences where each token is tagged for sentiment expressed within the sentence towards that token.

Since a sentiment target may be the target of several sentiment expressions, these are resolved to a final sentiment polarity (and intensity) using the conversion script in NoReC_tsa. There is no "mixed" sentiment category. When a target is the receiver of both positive and negative sentiment, the strongest wins. If a tie, the last sentiment wins.

Curated by: The SANT project (Sentiment Analysis for Norwegian Text) at the Language Technology Group (LTG) at the University of Oslo
Funded by: The SANT is funded by the Research Council of Norway (NFR grant number 270908).
Shared by: The SANT project (Sentiment Analysis for Norwegian Text) at the Language Technology Group (LTG) at the University of Oslo
License: The data is distributed under a Creative Commons Attribution-NonCommercial licence (CC BY-NC 4.0). The licence is motivated by the need to block the possibility of third parties redistributing the orignal reviews for commercial purposes. Note that machine learned models, extracted lexicons, embeddings, and similar resources that are created on the basis of NoReC are not considered to contain the original data and so can be freely used also for commercial purposes despite the non-commercial condition.
Language: Norwegian ("no"): Predominantly Bokmål (nb) written variant.

variant	split	sents	docs
nb	dev	1531	44
nb	test	1272	47
nb	train	8556	323
nn	train	78	4

Dataset Sources

Repository: https://github.com/ltgoslo/norec_tsa
Paper: The underlying NoReC_fine dataset is described in the paper A Fine-Grained Sentiment Dataset for Norwegian by Øvrelid et al., published at LREC 2020.

Uses

The data is intended to be used for training and testing models for TSA token classification; identifying and classifying sentiment targets in Norwegian sentences.
Example models fine-tuned on this dataset can be found at huggingface.co/collections/ltg/sentiment-analysis

Dataset Structure

The dataset comes in two flavours:

default configuration yields labels with binary Positive / Negative sentiment description
intensity configuration yields labels with additional sentiment intensity, 1: Slight, 2: Standard, and 3: Strong.

The config is required for accessing the version with intensity. tsa_data = load_dataset("ltg/norec_tsa", "intensity") The dataset comes with predefined train, dev (vallidation) and test splits.

Data Instances

Config "default" example instance:

{'idx': '701363-08-02',
 'tokens': ['Vi', 'liker', 'det', '.'],
 'tsa_tags': ['O', 'O', 'B-targ-Positive', 'O']}

Config "intensity" example instance:

{'idx': '701363-08-02',
 'tokens': ['Vi', 'liker', 'det', '.'],
 'tsa_tags': ['O', 'O', 'B-targ-Positive-2', 'O']}

Data Fields

idx(str): Unique document-and sentence identifier from NoReC_fine. The 6-digit document identifier can also be used to look up the text and its metadata in NoReC.
tokens: (List[str]): List of the tokens in the sentence
tsa_tags: (List[str]): List of the tags for each token in BIO format. There is no integer representation of these in the dataset.

Data Splits

DatasetDict({
    test: Dataset({
        features: ['idx', 'tokens', 'tsa_tags'],
        num_rows: 1272
    })
    train: Dataset({
        features: ['idx', 'tokens', 'tsa_tags'],
        num_rows: 8634
    })
    validation: Dataset({
        features: ['idx', 'tokens', 'tsa_tags'],
        num_rows: 1531
    })
})

Dataset Creation

Source Data

The sentiment annotations are aggregated from the NoReC_fine dataset, which in turn comprises a subset of the documents in the Norwegian Review Corpus (NoReC), which contains full-text professional reviews collected from major Norwegian news sources and cover a range of different domains, including literature, movies, video games, restaurants, music and theater, in addition to product reviews across a range of categories. The review articles NoReC were originally donated by the media partners in the SANT project; the Norwegian Broadcasting Corporation (NRK), Schibsted Media Group and Aller Media. The data comprises reviews extracted from eight different Norwegian news sources: Dagbladet, VG, Aftenposten, Bergens Tidende, Fædrelandsvennen, Stavanger Aftenblad, DinSide.no and P3.no. In terms of publishing date the reviews of NoReC mainly cover the time span 2003–2019, although it also includes a handful of reviews dating back as far as 1998.

Annotators

The original annotations of NoReC_fine that the sentence-level labels here are derived from, were originally created by hired annotators who were all BSc- or MSc-level students in the Language Technology study program at the Department of informatics, University of Oslo.

Personal and Sensitive Information

The data does not contain information considered personal or sensitive.

Recommendations

Results obtained on this data might not generalize to texts from other domains or genres. Any biases in the sentiments expressed by the original review authors may carry over to models trained on this data.

Citation

BibTeX:

@InProceedings{KutBarVel21,
  author = {Andrey Kutuzov and Jeremy Barnes and Erik Velldal and Lilja {\O}vrelid and Stephan Oepen}, 
  title = {Large-Scale Contextualised Language Modelling for Norwegian},
  booktitle = {{Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa 2021)}},
  year = 2021
}

@InProceedings{OvrMaeBar20,
  author = {Lilja {\O}vrelid and Petter M{\ae}hlum and Jeremy Barnes and Erik Velldal},
  title = {A Fine-grained Sentiment Dataset for {N}orwegian},
  booktitle = {{Proceedings of the 12th Edition of the Language Resources and Evaluation Conference}},
  year = 2020,
  address = "Marseille, France, 2020"
}

Dataset Card Authors

Egil Rønningstad and Erik Velldal

Dataset Card Contact

egilron@ifi.uio.no and erikve@ifi.uio.no