Sentiment analysis

Overview

Sentiment analysis assigns an opinion label to a piece of text. The label set differs between the cited studies: positive and negative, a three-way split into positive, negative and neutral, a five-way split that adds sarcasm and an indeterminate class, a continuous intensity score, or emotion 2301.08826 §2.1 1810.02508 tab. 13 2005.03545 §3.1 2606.25495 §5.1. A sarcasm paper trains one network to label sentiment and sarcasm together, with sarcasm as a yes/no label 1901.08014 §2. The review applies sentiment analysis to student feedback, where it sits next to feature extraction, topic modelling and summarisation 2301.08826 §2. The review is a literature survey with no experiments of its own, so every number from it is a result its authors quote from another study 2301.08826 §5. The sarcasm paper reports its own experiments, on one small benchmark 1901.08014 §3 1901.08014 §4. The multimodal and conversational studies also report their own experiments, on public benchmarks 2005.03545 §4 1810.02508 §4. A third multimodal study adds mutual-information terms to a regressor over the same three channels 2109.00412 §3.2. A hierarchical fusion model over text, audio and video reports its own results on CMU-MOSI and IEMOCAP, with a context variant for clips of utterances 1806.06228 §4.1 1806.06228 §4.4.3. A sarcasm study feeds sentiment, emotion and personality networks into a sarcasm classifier 1610.08815 §4 1610.08815 §5.2. A study of recommendation letters uses a sentiment classifier only to score tone, not to study sentiment models 2310.09219 §4.3.4. A canonical-correlation model over the same channels reports its own benchmark numbers 1911.05544 §4. A survey of emotion recognition in conversation labels each utterance with an emotion class rather than a polarity, and collects results from earlier papers 1905.02947 §1 1905.02947 §4. A comparison of a DNN, a CNN and an LSTM on eight tweet and review sets reports word embeddings ahead of TF-IDF for the CNN and LSTM on every set 2006.03541 tab. 2. A comparison of eight polarity tools measures how many messages each one labels at all, and how often the tools agree with one another 1406.0032 §1 1406.0032 §3.2. The finance study tests one prompted LLM on sentiment and other finance tasks 2401.11641 §4. A fine-tuned Arabic encoder reports five-class sentiment F1 0.75 on private telecom tweets 2606.25495 tab. 4. A position paper argues that the field is less solved than its polarity benchmarks suggest, and lists open problems without running any experiments 2005.00357 §1.

Variants

Dictionary tools

Dictionary methods score text against word lists. VADER and SentiWordNet are named as dictionary baselines for Twitter polarity, with their values not stated 2301.08826 §2.1. TextBlob appears in a feedback analysis tool with recall 96.17%, precision 67.47% and F 79.30% 2301.08826 §2.3.2. Card: Review of NLP methods for education feedback.

Eight tools, from emoticon lookups and LIWC to SentiWordNet and SenticNet, were compared on Twitter events and six labelled sets 1406.0032 §2 1406.0032 §3.1. Coverage, the share of messages a tool labels at all, ranged from about 4% to 95% on real events 1406.0032 §1. SentiWordNet and SenticNet had the highest coverage, at 90% to 95% on the AirFrance and H1N1 events 1406.0032 §4.1 1406.0032 fig. 1. Emoticons had the best average F, 0.846 over the labelled sets, but the lowest coverage 1406.0032 §4.3 1406.0032 tab. 5. A combination of seven tools reached 95% coverage with F 0.730 1406.0032 §5.1 1406.0032 tab. 5. The rule that maps F to each tool's weight, and the rule for conflicting labels, are not stated, so this combination cannot be rebuilt from the paper alone 1406.0032 §5.1. Card: Comparing and Combining Sentiment Analysis Methods.

Trained classifiers

A CNN on student feedback reports accuracy 87.95%, precision 87%, recall 78% and F1 81% 2301.08826 §2.4.5. An LSTM on the same kind of feedback reports 92% accuracy on positive text and 79% on negative text 2301.08826 §2.4.5. Neither entry gives a baseline value 2301.08826 §2.4.5.

Neural classifiers on tweets and reviews

A DNN, a CNN and an LSTM ("RNN" in the paper) are compared for binary polarity on eight tweet and review sets. Each model is fed either word embeddings (WE) or TF-IDF 2006.03541 §4.3 2006.03541 tab. 2. On Sentiment140, WE gives 0.8006 for the CNN and 0.8282 for the LSTM, against 0.7669 and 0.5696 with TF-IDF 2006.03541 tab. 2. The LSTM with WE beats the CNN with WE on six of the eight sets. Book reviews (0.7663 for the CNN against 0.7335 for the LSTM) and music reviews (0.7445 against 0.7310) are the exceptions 2006.03541 tab. 2. The paper reports one number per cell, with no seeds or spread [arxiv:2006.03541 tab. 2]. Card: Sentiment Analysis Based on Deep Learning: A Comparative Study.

Fine-tuned pretrained encoder on Arabic tweets

A MARBERT encoder, pretrained on dialectal and standard Arabic tweets, is fine-tuned on more than 22,000 manually labelled customer tweets from one telecom company, with five sentiment classes 2606.25495 §5.1 2606.25495 §4.2. The best of six settings, with training batch size 96, learning rate 2e-5, maximum length 256 and five epochs, reaches F1 0.75 2606.25495 tab. 4. At batch size 96 the per-class values are 0.85 (negative), 0.92 (positive), 0.89 (neutral), 0.54 (sarcasm) and 0.56 (indeterminate). At batch size 16 the sarcasm value is 0.33 2606.25495 tab. 5. The authors rejected over- and undersampling, and raised the sarcasm examples from about 200 to about 1,400 by collecting more tweets 2606.25495 §5.3. The data is held privately by the client, and the paper's comparison table uses figures from other datasets 2606.25495 §5.1 2606.25495 tab. 2. Card: Spam and Sentiment Detection in Arabic Tweets Using MARBERT Model.

Attention and embeddings for emotion

For emotion in student feedback, a plain LSTM reaches 86.27%, an LSTM with attention 87.49%, and multi-head attention 90.03%. Multi-head attention with GloVe and CoVe embeddings reaches 94.13% 2301.08826 §2.1. The review also reports GloVe about 5–7% better than Word2Vec in a cited study, without the task detail 2301.08826 §2.1.

Aspect-level opinions

A MOOC study reports F1 86.13% for the aspect category and 82.10% for the aspect's sentiment 2301.08826 §2.3.2. Another reports 91% accuracy for aspect extraction and 93% for sentiment detection 2301.08826 §2.4.2. The review lists aspect-based sentiment analysis as a challenge of its own 2301.08826 §3.5. Card: Review of NLP methods for education feedback.

Joint sentiment and sarcasm

One GRU reads the sentence with attention pooling for each task. A neural tensor network fuses the two task vectors before the sarcasm softmax, and each task has its own two-class softmax 1901.08014 §2 1901.08014 fig. 1. On 994 sentences with 10-fold cross-validation, the best variant reaches average F 86.66. A simple multi-task model reaches 84.37, and the cited state of the art 82.13 1901.08014 tab. 1. For sentiment alone, the best variant reaches F 83.03 against 77.30 for the state of the art 1901.08014 tab. 1. The paper reports that the sarcasm signal helps mainly when there is a clear sentiment shift 1901.08014 §4. The baseline uses eye-movement features that this model does not use 1901.08014 §4, so the gap is not a like-for-like test. The paper has no official code and does not state the learning rate, batch size or epochs 1901.08014 §2. Card: Sentiment and Sarcasm Classification with Multitask Learning.

Sarcasm with features from sentiment, emotion and personality networks

Three CNNs are first trained on sentiment, emotion and personality corpora. Their outputs are concatenated with the features of a baseline sarcasm CNN, and an SVM makes the final decision 1610.08815 §4 1610.08815 §5.2. Word vectors are word2vec and are trained along with the network 1610.08815 §4.1. On Dataset 1 the all-feature SVM reaches F1 97.71, against 97.60 for the baseline CNN-SVM and 95.04 for the baseline CNN 1610.08815 tab. 2. On Dataset 2 the figures are 94.80 against 92.32, and on Dataset 3, 93.30 against 92.20 1610.08815 tab. 2. Adding only the sentiment features to the baseline features moves Dataset 1 from 97.60 to 97.67 in the CNN-SVM setting 1610.08815 tab. 2. Trained on Dataset 3 and tested on Dataset 1, the model reaches 76.78, against 53.02 for Ptacek et al. (2014) in the authors' re-implementation 1610.08815 tab. 3. Training on Dataset 1 and testing on Dataset 3 drops F1 to 33.05 1610.08815 §5.6. All numbers are F1 from 5-fold cross-validation, and the prior methods are the authors' own re-implementations 1610.08815 §5.2. Card: A Deeper Look into Sarcastic Tweets Using Deep Convolutional Neural Networks.

Multimodal sentiment (text, audio and video)

MISA splits each modality's representation into a part shared across modalities and a part private to one modality, then fuses both for the prediction 2005.03545 §3 2005.03545 tab. 4. Inputs are word-aligned features from short video utterances 2005.03545 §3.1. On CMU-MOSI with BERT text features, the authors report MAE 0.783, correlation 0.761, binary accuracy 81.8% and 83.4% for the two conventions, and 7-class accuracy 42.3% 2005.03545 tab. 1. On CMU-MOSEI they report MAE 0.555, correlation 0.756 and binary accuracy 83.6% and 85.5% 2005.03545 tab. 2. Some baseline rows were copied from the cited works rather than re-run 2005.03545 §4.2. Removing the language input drops MOSI correlation to 0.041, while removing visual or audio input raises MAE to 0.798 or 0.849 2005.03545 tab. 4. Card: MISA: Modality-Invariant and -Specific Representations for Multimodal Sentiment Analysis.

Hierarchical fusion of text, audio and video

HFusion encodes each modality per utterance and fuses each pair of modalities per dimension with learned weights. A GRU adds context across the utterances of a clip, and the three pair vectors are fused again 1806.06228 §3.1 1806.06228 alg. 1. The authors contrast this with early fusion, which concatenates the unimodal vectors 1806.06228 §3.1. On CMU-MOSI without context, the reported accuracy with the Poria et al. features is 74.6 for all three modalities, against baselines of 73.5 and 71.2 1806.06228 tab. 2. With this paper's own features the figure is 77.9, against 77.0 and 77.3 for the baselines 1806.06228 tab. 2. With context (CHFusion) the reported MOSI accuracy for all three modalities is 80.0, against 78.7 and 77.2 for the two comparison methods 1806.06228 tab. 3. On IEMOCAP with four emotion classes, the three-modality accuracy is 76.5 and F-score 76.8 1806.06228 tab. 3. The gains are 1 to 2.4 points, and the number of runs, seeds and any spread are not stated 1806.06228 §4.4.2. The authors' repository is MIT-licensed, but it differs from the paper on several hyperparameters, and some unimodal encoders are not in it, so reproduction needs care (see the card's Pitfalls). Card: Multimodal Sentiment Analysis using Hierarchical Fusion with Context Modeling.

Canonical correlation across modalities

ICCN forms outer products between a BERT text vector and the LSTM-encoded audio and video states, and passes each product through a 2D CNN. The two CNNs are trained with a deep canonical correlation (CCA) loss so that their outputs correlate, and the text vector is joined to both outputs for a small downstream model 1911.05544 §3 1911.05544 eq. 5. On CMU-MOSI it reports binary accuracy 83.07, F 83.02, MAE 0.862, 7-class accuracy 39.01 and correlation 0.714, the best of the compared methods on all five metrics [arxiv:1911.05544 tab. 1]. On CMU-MOSEI it is best on MAE (0.565) and 7-class accuracy (51.58), while MFM is higher on binary accuracy, F and correlation 1911.05544 tab. 1. On IEMOCAP it is best for three of its four emotions 1911.05544 tab. 2. Each of its three ablations scores below the full model on every column of its ablation table 1911.05544 tab. 3. Card: ICCN (Interaction Canonical Correlation Network).

Mutual-information terms (text, visual and acoustic)

MMIM trains a regressor on text, visual and acoustic inputs. Its loss is the task loss plus two mutual-information (MI) terms, weighted by α and β 2109.00412 §3.6 2109.00412 eq. 13. The input-level term is a Barber-Agakov lower bound on the MI between the text and each other channel, with an entropy estimate from a two-class Gaussian mixture 2109.00412 §3.4 2109.00412 eq. 2 2109.00412 eq. 7. The fusion-level term is a contrastive loss that asks the fused vector to predict each channel's vector 2109.00412 §3.5 2109.00412 eq. 11. On CMU-MOSEI it reports MAE 0.526, correlation 0.772 and 7-class accuracy 54.24, against 0.530 and 0.765 for Self-MM in the same table 2109.00412 tab. 2. On CMU-MOSI it reports MAE 0.700 and correlation 0.800 2109.00412 tab. 2. The paper says the gains are small on several metrics 2109.00412 §4.3, and on one MOSEI binary-accuracy column its 82.24 is below Self-MM's 82.81 2109.00412 tab. 2. Removing all input-level MI terms raises MOSEI MAE to 0.541 and lowers correlation to 0.752 2109.00412 tab. 4. Dropping the memory of earlier batches from the entropy estimate gives NaN in every metric 2109.00412 tab. 4. Some baseline rows were copied from other papers and others were reproduced by the authors 2109.00412 tab. 2. The paper card's "Paper vs code" section records that the released code enables the MI terms only with a flag, and that it picks the reported epoch on test MAE. Card: Multimodal InfoMax.

Conversation context

On MELD, a multi-party dataset from TV dialogue, DialogueRNN models the speakers and the turns before the target utterance. For three-way sentiment, its text-only version reaches weighted F 66.10, against 64.25 for a text-CNN baseline. Adding audio reaches 67.56, the best value in the table 1810.02508 tab. 13. The authors report about 1.4 points from adding audio to the text-only DialogueRNN 1810.02508 §4.3. The test split has 2,610 utterances 1810.02508 tab. 8. Card: MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversations.

The survey of emotion recognition in conversation reports DialogueRNN, CMN and ICON on IEMOCAP with six emotion classes. DialogueRNN reaches weighted F1 62.75, against 58.54 for ICON and 56.13 for CMN 1905.02947 tab. 3. The ICON row was re-run by the survey authors 1905.02947 tab. 3. For DialogueRNN, the survey gives 47.5% accuracy on utterances whose emotion changes from the previous one and 69.2% on utterances without a change. It does not name the dataset for these two figures 1905.02947 §4. The survey says CMN, ICON, IANN and DialogueRNN all perform poorly on shifted utterances 1905.02947 §4. Card: Emotion Recognition in Conversation: Research Challenges, Datasets, and Recent Advances.

Sentiment as a measurement tool (tone in generated text)

A study of ChatGPT and Alpaca reference letters scores each sentence for formality, positivity and agency, using a sentiment classifier tuned on SST-2 for positivity 2310.09219 §3.2.2 2310.09219 §4.3.4. Here the classifier is a measuring instrument for bias, not the object of study. In the biography setting, positivity differs between men's and women's letters with t = 5.93 (p = 1.58e-09) for ChatGPT, and with t = 1.47 (p = 0.07) for Alpaca 2310.09219 tab. 4. Scores from a classifier of this kind depend on its checkpoint, so any reuse of such a measure should pin the model and report its version. The card records that the default checkpoint is not pinned in the released code. Card: Gender-bias study of ChatGPT and Alpaca reference letters.

Prompted large language models

The finance study prompts GPT-4 with zero-shot or one-shot instructions and reports accuracy 0.78 on the Financial PhraseBank three-class task and 0.79 on FiQA sentiment 2401.11641 tab. 2. No baseline model is in the table 2401.11641 tab. 2. The prompts are described in words only, and the model snapshot and decoding settings are not stated 2401.11641 §4.2. Card: Revolutionizing Finance with LLMs: An Overview of Applications and Insights.

Open problems (position paper)

This is a position paper, not a method, so it has no results table. Its argument is that near-perfect scores on polarity benchmarks such as IMDB and SST-2 hide unsolved problems 2005.00357 §1 2005.00357 fig. 1. The open areas it lists are aspect-level sentiment, multimodal sentiment, conversational context, sentiment reasoning (who holds the opinion and what causes it), domain adaptation, multilingual and code-mixed text, sarcasm, sentiment-aware generation and bias 2005.00357 §3.1–2005.00357 §3.9. It states that simple concatenation performs about as well as most multimodal fusion methods, but gives no numbers for this 2005.00357 §3.2.1. It also argues that a bag-of-words input may explain the gains of an adversarial domain-adaptation method 2005.00357 §3.5. Both claims are the authors' experience, not measured results in the paper. The paper defines no measurable target for "true sentiment understanding" 2005.00357 §1. Its dataset sizes and the Wikipedia contributor share it quotes are not checked against their sources 2005.00357 §3.7.1 2005.00357 §3.9.1. Card: Beneath the Tip of the Iceberg: Current Challenges and New Directions in Sentiment Analysis Research.

When to use which

Implementation sketch

The review gives no algorithm for sentiment analysis 2301.08826 §2. The sketch below follows the pipeline of one comparative study, which states its cleaning steps, features and folds 2006.03541 §4.2 §4.3 §4.4. It is ours, and the paper gives no code.

# Binary sentiment with cross-validation: our sketch of the study pipeline
# Follows [arxiv:2006.03541 §4.2] [§4.3] [§4.4]; not the paper's code
# Input:  texts with polarity labels; Output: one accuracy per fold

def sentiment_scores(texts, labels, model_name):
    # Keep positive and negative samples only; map them to 1 and 0  [§4.1] [§4.3]
    texts, labels = binarise(texts, labels)
    # Clean each text: remove mentions, links and symbols; lower-case  [§4.3]
    cleaned = [preprocess(text) for text in texts]
    # Word ids with a vocabulary of 15,000 and length 40  [§4.4]
    token_ids = to_token_ids(cleaned, vocab=15000, max_len=40)
    scores = []
    # 10-fold cross-validation, one score per held-out fold  [§4.2]
    for fold in kfold(token_ids, labels, k=10):
        # Train on the training part; optimiser and learning rate not stated
        model = build_model(model_name)               # e.g. CNN or LSTM  [§4.4]
        train(model, fold.train, batch_size=4096)     # batch size  [§5]
        # Score the held-out part  [§4.2]
        scores.append(accuracy(model, fold.test))
    return scores

Contradictions

No direct disagreement between the cited results appears on this page. The accuracies, F1 values, weighted F scores, MAE and correlations come from different datasets, languages, labels, tasks and baselines, and some are Twitter, e-commerce, TV dialogue or finance data rather than student feedback 2301.08826 §2.1 2301.08826 §2.3.2 1810.02508 §3 2005.03545 §4.1 2606.25495 §5.1. Regression scores (MAE, correlation) and classification scores (accuracy, F) cannot be ranked against each other. The review does not claim that its numbers are comparable 2301.08826 §2.

The conversation-emotion survey uses emotion labels, not polarity. Its MELD weighted F1 of 57.03 for DialogueRNN covers seven emotion classes 1905.02947 tab. 5. The MELD weighted F of 66.10 above covers three sentiment classes 1810.02508 tab. 13. The two numbers measure different tasks, so do not read one against the other. Its tables also mix results from several earlier papers. Most of its comparisons are against DialogueRNN, whose paper three of the four survey authors co-wrote 1905.02947 §4.

MISA's numbers differ between its own paper and a later comparison table. MISA reports MOSI MAE 0.783 and MOSEI MAE 0.555 2005.03545 tab. 1 2005.03545 tab. 2. The MMIM table lists MISA at MOSI MAE 0.804 and MOSEI MAE 0.568, with different binary-accuracy values, and marks that row as taken from the Self-MM paper rather than MISA's own 2109.00412 tab. 2. Both sets are reported here as their papers give them. Use one table at a time, and do not compare a MISA row from one paper with an MMIM row from the other. The ICCN table uses yet another MOSI and MOSEI set of baseline rows 1911.05544 tab. 1, so the same caution applies.

ICCN's text calls it state of the art, but its own Table 1 shows MFM ahead on several MOSEI and IEMOCAP metrics 1911.05544 §6 1911.05544 tab. 1 1911.05544 tab. 2. The paper says hyperparameters are chosen on validation performance 1911.05544 §4. The released MOSEI export instead keeps the run with the lowest test MAE, and its best printed run (MAE 0.5693, correlation 0.7122) does not match the MOSEI numbers in Table 1 (0.565 and 0.713) mosei_dcca_exp (2).html mosei_dcca_exp (2).html. Use the paper's tables with that gap in mind.

Inside the lexicon comparison, the text and the table disagree. Section 5.1 says the combined F of 0.730 is higher than every tool except Emoticons, but Table 5 lists SentiStrength at 0.765 1406.0032 §5.1 1406.0032 tab. 5. The SentiStrength row uses figures from its own paper rather than a run on these datasets 1406.0032 §4.3, so it is not directly comparable with the other rows. The F values in Table 5 also do not equal the harmonic mean of the precision and recall in the same table. For Emoticons that mean is about 0.86, against the listed 0.846, and the paper does not say how F was averaged 1406.0032 §3.2 1406.0032 tab. 5. Coverage numbers depend on the order in which tools are combined: one passage reports above 92.75% after two tools, another 70% after two 1406.0032 §4.1 1406.0032 §5.1.

Inside the MARBERT paper, the abstract claims accuracy gains over existing techniques. The paper reports F1, and its comparison table uses accuracies from other datasets, so no baseline is run on the telecom data 2606.25495 abstract 2606.25495 tab. 2. The abstract's total of 24,513 tweets also differs from the sum of its per-class counts, 24,477 2606.25495 abstract. The main text gives "more than 22K" sentiment tweets 2606.25495 §5.1. Which set is the training data is not stated.

Inside the sarcasm paper, the abstract describes a 3 to 4 percent gain over the state of the art. The Table 1 values give differences of 3.32 (sarcasm F), 4.53 (average F) and 5.73 (sentiment F) points, by our arithmetic from the table 1901.08014 tab. 1. The two do not match cleanly, so use the table numbers.

Inside the finance study, FiQA sentiment is described as a continuous score from -1 to 1, yet Table 2 reports accuracy, and the paper does not say how scores were turned into classes 2401.11641 §4.1 2401.11641 tab. 2.

Inside the comparative study, the text gives two accounts of the embedding layer. One says the vectors are first trained with Word2Vec, the other that they start from random weights and are learned, and Figure 9 lists the embedding parameters as non-trainable 2006.03541 §4.4 2006.03541 fig. 9. The paper does not say which setup produced its results. Its conclusion also says the CNN "outperforms other models", while its own table shows the LSTM with embeddings ahead on six of eight sets 2006.03541 §6 2006.03541 tab. 2. The claim is best read as the authors' judgement of accuracy against training time.

Papers

• Review of NLP methods for education feedback: a literature review of text methods for student feedback, covering sentiment annotation, aspect-level analysis and the main challenges.
• Comparing and Combining Sentiment Analysis Methods: a comparison of eight polarity tools on coverage and agreement over Twitter events and six labelled sets, with a combined method built from seven of them.
• A Deeper Look into Sarcastic Tweets Using Deep Convolutional Neural Networks: a sarcasm CNN with features from pre-trained sentiment, emotion and personality networks, with author-reported F1 on three Twitter sets.
• Multimodal Sentiment Analysis using Hierarchical Fusion with Context Modeling: a hierarchical fusion model over text, audio and video, with author-reported results on CMU-MOSI and IEMOCAP and a context variant.
• Sentiment and Sarcasm Classification with Multitask Learning: one GRU with attention and tensor fusion that labels sentiment and sarcasm together, with author-reported results on a 994-sentence benchmark.
• Beneath the Tip of the Iceberg: Current Challenges and New Directions in Sentiment Analysis Research: a position paper that traces the field's milestones and lists nine open directions, with no experiments.
• MISA: Modality-Invariant and -Specific Representations for Multimodal Sentiment Analysis: a fusion model that splits each modality into shared and private parts, with author-reported results on CMU-MOSI and CMU-MOSEI.
• ICCN (Interaction Canonical Correlation Network): a model that correlates text-conditioned audio and video features with a deep CCA loss, with author-reported results on CMU-MOSI, CMU-MOSEI and IEMOCAP.
• Multimodal InfoMax (MMIM): a regressor trained with mutual-information bounds on text, visual and audio inputs, with author-reported results on CMU-MOSI and CMU-MOSEI.
• MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversations: a dialogue dataset with sentiment and emotion labels and baselines that use context and audio.
• Emotion Recognition in Conversation: Research Challenges, Datasets, and Recent Advances: a survey of emotion recognition in conversation, with dataset statistics and author-collected results for CMN, ICON, IANN, DialogueRNN and HRLCE.
• Gender-bias study of ChatGPT and Alpaca reference letters: a bias study of ChatGPT and Alpaca reference letters that uses sentiment as one tone measure, with author-reported gaps by gender.
• Revolutionizing Finance with LLMs: An Overview of Applications and Insights: a finance survey with a small author-run GPT-4 evaluation that includes sentiment labelling.
• Sentiment Analysis Based on Deep Learning: A Comparative Study: a DNN, CNN and LSTM compared with word embeddings or TF-IDF on eight tweet and review sets, with author-reported accuracies from 10-fold cross-validation.
• Spam and Sentiment Detection in Arabic Tweets Using MARBERT Model: a MARBERT fine-tune for five-class sentiment on private telecom tweets, with per-class F1 and a note on imbalance handling.