Dictionaries for Sentiment Analysis
Bibliographic record
Abstract
This repository was created for my Master's thesis in Computational Intelligence and Internet of Things at the University of Córdoba, Spain. The purpose of this repository is to store the dictionaries found that were used in some of the studies that served as research material for this Master's thesis. Below are the dictionaries specified, along with the details of their references, authors, and download sources. ----------- SentiWordNet 3.0 ---------------- SentiWordNet is based on WordNet 2.0 and has been built by automatically associating each WordNet synset to three scores: Obj for objective terms, Pos and Neg for positive and negative terms. Each score ranges from 0.0 to 1.0. Reference: Catelli, R.; Pelosi, S.;Esposito, M. "Lexicon-Based vs.Bert-Based Sentiment Analysis: A Comparative Study in Italian". Electronics 2022, 11, 374.https://doi.org/10.3390/electronics11030374 Download source: https://github.com/aesuli/sentiwordnet File name: SentiWordNet_3.0.0.txt ----------- Multi-perspective Question Answering (MPQA) Subjectivity Lexicon ---------------- The MPQA Subjectivity Lexicon has 8222 words: 2719 positive, 4914 negative and 591 neutral words. It includes adjectives, adverbs, verbs, nouns and 'anypos' (any part-of-speech). The lexicon was aggregated from a variety of sources, including manually developed sources as well as automatically constructed sources. Reference: Khoo, Christopher SG, and Sathik Basha Johnkhan. "Lexicon-based sentiment analysis: Comparative evaluation of six sentiment lexicons". Journal of Information Science 44.4 (2018): 491-511. Download source: http://mpqa.cs.pitt.edu/lexicons/subj_lexicon/ File name: subjectivity_clues_hltemnlp05.zip ----------- NRC Word-Emotion Association Lexicon ---------------- The NRC: National Research Council Canada (NRC) Word-Sentiment Association Lexicon (aka EmoLex) is one of the most well-known domain-specific lexicons incorporating the sentiment polarity and the emotions in the same lexicon for a crowd-sourcing scenario. Reference: Taborda, Bruno, et al. "SA-MAIS: Hybrid automatic sentiment analyser for stock market". Journal of Information Science (2023): 01655515231171361. Download source: https://saifmohammad.com/WebPages/NRC-Emotion-Lexicon.htm File name: NRC-Emotion-Lexicon.zip ----------- Sentiment Italian Lexicon ---------------- The Sentiment Italian Lexicon (Sentix) is a lexicon for Sentiment Analysis of Italian. It is the result of the alignment of several resources: WordNet , MultiWordNet, BabelNet and SentiWordNet. Reference: TWITA. (s. f.). https://valeriobasile.github.io/twita/sentix.html Download source: https://valeriobasile.github.io/twita/downloads.html File name: sentix.gz ----------- SentIta ---------------- SentIta is a sentiment lexicon for the Italian language that has been semiautomatically generated on the base of the richness of the Italian lexical databases of Nooj (Silberztein, 2003; Vietri, 2014) and the Italian Lexicon-grammar (LG) resources (Elia et al., 1981; Elia, 1984). Reference: Pelosi, S. SentIta and Doxa. "Italian Databases and Tools for Sentiment Analysis Purposes". In Proceedings of the Second ItalianConference on Computational Linguistics CLiC-it 2015; Accademia University Press: Turin, Italy, 2015; pp. 226–231. https://books.openedition.org/aaccademia/1537 Download source: https://github.com/NicGian/SentITA File name: sentita_0.2.0.zip ----------- The Distributional Polarity Lexicon ----------------The Distributional Polarity Lexicons (DPLs) are large-scale polarity lexicons acquired with an unsupervised methodology and are publicly available in English and Italian. In this page I uploaded 4 lexicons in English and Italian. In each language we release two versions of the lexicons. The first version contains a list of plain words with polarity scores, e.g. good, suffered, loved, smile. The second version contains a list of words that have been pre-processed, to produce lemma::pos pairs, i.e. good::j indicates the adjective good, while pain::n indicates the noun pain. We call the plain words lexicon DPL-EN and DPL-IT, respectively for English and Italian. The pre-processed versions are called DPLp-EN and DPLp-IT, respectively for English and Italian. References: Castellucci, G., Croce, D., & Basili, R. (2016). "A language independent method for generating large scale polarity lexicons". Language Resources and Evaluation, 38-45. https://www.aclweb.org/anthology/L16-1007.pdf Distributional Polarity Lexicon | Semantic Analytics Group @ Uniroma2. (s. f.). http://sag.art.uniroma2.it/demo-software/distributional-polarity-lexicon/ Download source: http://sag.art.uniroma2.it/demo-software/distributional-polarity-lexicon/ Files names: DPL-EN_lrec2016.txt.gz DPL-IT_lrec2016.txt.gz DPLp-EN_lrec2016.txt.gz DPLp-IT_lrec2016.txt.gz ----------- SenticNet ---------------- SenticNet is a sentiment lexicon for concept-level sentiment analysis that was automatically constructed by applying graph-mining and multi-dimensional scaling techniques on the affective commonsense knowledge collected from three different sources, namely: WordNet-Affect, Open Mind Common Sense and GECKA. It performs tasks such as polarity detection and emotion recognition. It supports different languages such as English, Arabic, Spanish, Portuguese, French, German, Italian, Russian, Turkish, Korean, Polish, Indonesian and more. References: Cambria, E.; Li, Y.; Xing, F.Z.; Poria, S.; Kwok, K. "SenticNet 6: Ensemble Application of Symbolic and Subsymbolic AI for Sentiment Analysis". In Proceedings of the CIKM '20: The 29th ACM International Conference on Information and Knowledge Management, Virtual Event, Ireland, 19–23 October 2020; d'Aquin, M., Dietze, S., Hauff, C., Curry, E., Cudré-Mauroux, P., Eds.; ACM: New York, NY, USA, 2020; pp. 105–114. https://dl.acm.org/doi/10.1145/3340531.3412003 SenticNet. (s. f.). https://sentic.net/ Download source: https://sentic.net/downloads/ File name: senticnet.zip (file is in English language, for files in other languages please use the download source)
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.129 | 0.010 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; both teacher heads agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".