Enhanced Sarcasm Detection in Telugu Dialogue Systems Using Self Attention-Based RNN and Gated Recurrent Unit Models

Sarcasm detection is used to single out natural language statements where intended meaning differs from what the surface meaning implies. A number of tasks in natural language processing areas like sentiment analysis and also mining of opinions use sarcasm detection underneath. Many of the key research in the area of sarcasm detection primarily focus on only text-based input. In the present day scenario , there has been a sudden explosion in the amount of multimodal data mainly due to social media. As a result of that, users these days are not just limited to text while expressing themselves , but also make heavy use of visuals like in images and videos. The objective of our paper is to incorporate multimodal data so as to enhance the performance of present sarcasm detection algorithms. Multiple methods that leverage data in the form of image and text, and also as a combination of both, have been presented thus far.We present a unique architecture which works on the Robustly Optimized BERT pre-training approach(RoBERTa) which is a facebook modified version of well known model BERT with a co¬attention layer on the top for including the context incongruency between input text and attributes of the image. Using Gated Recurrent Unit(GRU), the text-based features are extracted and the features of the image are retrieved by conditioning the image through Feature-wise Linearly Modulated ResNet Blocks. These combined with CLS token from the model RoBERTa are used to obtain the final predictions. In the end, we show performance enhancements from our model when it is stacked against the other latest models with competitive results which make use of the same publicly available Twitter dataset.

Sarcasm detection plays an important role in Natural Language Processing as it has been considered one of the most challenging task in sentiment analysis and opinion mining appli- cations. Our work aims to recognize sarcasm in social media sites, microblogs and discussion forums, exploiting the potential of Deep Learning tools such as Deep Neural Network and Word Embeddings. In this thesis, we (a) develop multiple types of neural models and analyze their efficiency when combined with word embeddings; (b) create a new multitasking frame- work that exploits the strong correlation between sarcasm and sentiment detection (c) test the performances of our models on two pre-labelled datasets; (d) compare our results with other state-of-the-art models; (e) apply our models on real word data to evaluate the efficiency of their prediction. We then discuss on the benefits of our research in the field of sarcasm detection and sentiment analysis, and put the basis for some future researches.

As one of the biggest data sources, conversations on social media can be used for various purposes, one of which is sentiment analysis. By utilizing sentiment analysis, we can obtain an overview of public opinion on certain topics, such as business and politics. Through sentiment analysis, text can be grouped into positive, negative, or neutral sentiments. Sarcasm can cause inaccurate classification of sentiment. In this study, we examined the effect of sarcasm detection on sentiment analysis accuracy. The dataset used in this study contains a group of 1000 Indonesian language Tweet. The features used for sarcasm detection are sentiment score and interjection words. The algorithm used for sarcasm detection is Random Forest. Sentiment analysis was performed using the TF-IDF as the feature and the NaYve Bayes as the classification algorithm. The test results show that the sarcasm detection does not improve the accuracy of sentiment analysis. The average sentiment analysis accuracy is 74.3%. The average sentiment analysis accuracy with sarcasm detection is 72.9%. The average accuracy of sarcasm detection is 71.7%.

Satire detection and sentiment analysis are intensively explored natural language processing (NLP) tasks that study the identification of the satirical tone from texts and extracting sentiments in relationship with their targets. In languages with fewer research resources, an alternative is to produce artificial examples based on character-level adversarial processes to overcome dataset size limitations. Such samples are proven to act as a regularization method, thus improving the robustness of models. In this work, we improve the well-known NLP models (i.e., Convolutional Neural Networks, Long Short-Term Memory (LSTM), Bidirectional LSTM, Gated Recurrent Units (GRUs), and Bidirectional GRUs) with adversarial training and capsule networks. The fine-tuned models are used for satire detection and sentiment analysis tasks in the Romanian language. The proposed framework outperforms the existing methods for the two tasks, achieving up to 99.08% accuracy, thus confirming the improvements added by the capsule layers and the adversarial training in NLP approaches.

A Recurrent Neural Network (RNN) is a specialized form of neural network that is adept at handling sequential data by retaining information from prior inputs. In contrast to conventional feedforward neural networks, RNNs incorporate loops in their architecture, allowing them to leverage data from previous time steps to affect the current output. This characteristic renders RNNs especially effective for applications that involve sequences, including time-series forecasting, natural language processing, and speech recognition. A fundamental component of RNNs is their hidden state, which acts as a dynamic memory that is refreshed with each incoming input. This allows RNNs to capture dependencies across time steps, which is crucial for understanding context in sequences. In language modeling, the interpretation of a word often relies on the words that come before it, a task that Recurrent Neural Networks (RNNs) handle well. However, RNNs struggle with issues like vanishing gradients, which hinder their ability to capture long-range dependencies. To overcome this, models such as Long Short-Term Memory (LSTM) networks and Gated Recurrent Units (GRUs) were introduced. These models incorporate gates that regulate the flow of information, allowing them to better learn long-term dependencies. RNNs remain a powerful tool for working with sequential data, facilitating the modeling of temporal relationships, but their effectiveness depends on careful design and optimization. Research significance: Recurrent Neural Networks (RNNs) hold significant research value because of their capacity to simulate temporal and sequential data, which is essential in many fields. They are frequently employed in natural language processing for tasks such as sentiment analysis, language translation, and text generation. In time-series analysis, RNNs enable accurate forecasting in finance, healthcare, and climate modeling. They also are essential in speech recognition and video processing, handling dependencies across time steps. Research focuses on improving RNNs, addressing challenges like vanishing gradients, and enhancing efficiency through architectures like LSTMs and GRUs, solidifying their relevance in advancing AI and machine learning applications. Methodology: A technique for analyzing the relationships between several variables, particularly in situations when data is limited or unclear, is called gray relational analysis, or GRA. In order to comprehend the relationships between variables, it evaluates how similar or different they are. GRA aids decision-makers in identifying critical factors, prioritizing actions, and improving processes in complex fields like engineering, finance, and management. By converting both qualitative and quantitative data into gray numbers, GRA addresses uncertainty and provides valuable insights for problem-solving, decision-making, and performance improvement, leading to more informed and effective strategies. Alternative taken as Simple RNN, LSTM, GRU, Bidirectional RNN, Deep RNN, Vanilla RNN, Echo State Network, Attention-based RNN, Transformer RNN, GRU with Attention. Evaluation preference taken as Prediction Accuracy, Model Robstness, Learning Efficiency, Training Time, Complexity. Attention-based RNN has the lowest score, Deep RNN has the highest rank, according to the results.

Computational linguistics refers to an interdisciplinary field associated with the computational modelling of natural language and studying appropriate computational methods for linguistic questions. The number of social media users has been increasing over the last few years, which have allured researchers’ interest in scrutinizing the new kind of creative language utilized on the Internet to explore communication and human opinions in a better way. Irony and sarcasm detection is a complex task in Natural Language Processing (NLP). Irony detection has inferences in advertising, sentiment analysis (SA), and opinion mining. For the last few years, irony-aware SA has gained significant computational treatment owing to the prevalence of irony in web content. Therefore, this study develops Computational Linguistics with Optimal Deep Belief Network based Irony Detection and Classification (CLODBN-IRC) model on social media. The presented CLODBN-IRC model mainly focuses on the identification and classification of irony that exists in social media. To attain this, the presented CLODBN-IRC model performs different stages of pre-processing and TF-IDF feature extraction. For irony detection and classification, the DBN model is exploited in this work. At last, the hyperparameters of the DBN model are optimally modified by improved artificial bee colony optimization (IABC) algorithm. The experimental validation of the presented CLODBN-IRC method can be tested by making use of benchmark dataset. The simulation outcomes highlight the superior outcomes of the presented CLODBN-IRC model over other approaches.

Speaker identification is a classification task which aims to identify a subject from a given time-series sequential data. Since the speech signal is a continuous one-dimensional time series, most of the current research methods are based on convolutional neural network (CNN) or recurrent neural network (RNN). Indeed, these methods perform well in many tasks, but there is no attempt to combine these two network models to study the speaker identification task. Due to the spectrogram that a speech signal contains, the spatial features of voiceprint (which corresponds to the voice spectrum) and CNN are effective for spatial feature extraction (which corresponds to modeling spectral correlations in acoustic features). At the same time, the speech signal is in a time series, and deep RNN can better represent long utterances than shallow networks. Considering the advantage of gated recurrent unit (GRU) (compared with traditional RNN) in the segmentation of sequence data, we decide to use stacked GRU layers in our model for frame-level feature extraction. In this paper, we propose a deep neural network (DNN) model based on a two-dimensional convolutional neural network (2-D CNN) and gated recurrent unit (GRU) for speaker identification. In the network model design, the convolutional layer is used for voiceprint feature extraction and reduces dimensionality in both the time and frequency domains, allowing for faster GRU layer computation. In addition, the stacked GRU recurrent network layers can learn a speaker’s acoustic features. During this research, we tried to use various neural network structures, including 2-D CNN, deep RNN, and deep LSTM. The above network models were evaluated on the Aishell-1 speech dataset. The experimental results showed that our proposed DNN model, which we call deep GRU, achieved a high recognition accuracy of 98.96%. At the same time, the results also demonstrate the effectiveness of the proposed deep GRU network model versus other models for speaker identification. Through further optimization, this method could be applied to other research similar to the study of speaker identification.

Sarcasm is one of the problem that affect the result of sentiment analysis. According to Maynard and Greenwood (2014), performance of sentiment analysis can be improved when sarcasm also identified. Some research used Naïve Bayes and Random Forest method on sentiment analysis process. On Salles, dkk (2018) research, in some cases Random Forest outperform the performance by Support Vector Machine that known as a superior method. In this research, we did sentiment analysis on comment section on Instagram account of Indonesian politician. This research compare the accuracy of sentiment analysis with sarcasm detection and analysis sentiment without sarcasm detection, sentiment analysis with Naïve Bayes and Random Forest method then Random Forest for sarcasm detection. This research resulted in accuracy value in sentiment analysis without sarcasm detection with Naïve Bayes 61%, with Random Forest method 72%. Accuracy on sentiment analysis with sarcasm detection using Naïve Bayes – Random Forest method is 60% and using Random Forest – Random Forest method is 71%.

Sarcasm can change the polarity of a sentence and it becomes the opposite. While sentiment analysis on social media has been widely used, but it is still rare to find sentiments and analyze them, considering the detection of sarcasm in it. Sarcasm detection in sentiment analysis is a challenging task. After successful identification of sarcasm the quality of sentiment analysis improves drastically. Experiments about sentiment analysis by detection of sarcasm are more often found in the language used in context with some special words. Therefore, taking into account research done on English tweets, this study analyzes the sentiment analysis sarcasm in Tweets agreed within context (specific topic) using the interjection and unigram features as features The main task is to detect sarcastic sentences and compare using classification methods namely Support Vector Machine with polynomial kernels. Thereafter incorporating interjection feature words that were expressing one's feelings and intentions and the unigram feature which is a collection of words a single obtained from the corpus automatically. Results of experiments show that the use of interjection features and unigram as detection of sarcasm in tweets using SVM will enhance the accuracy by 91%.

Social media platforms, especially Instagram, have become a significant source of digital communication where users can interact through satire, humour, and irony. Sarcasm is a rhetorical device primarily used to convey hidden meaning, criticism, or humour. This study examined sarcastic comments on the Instagram page @Sarcastic_us, categorising them into different types of Sarcasm based on contextual and linguistic features. A descriptive, qualitative approach was employed to analyse 600 comments from 30 posts using content analysis. Key findings of the study reveal six primary types of Sarcasm employed by Instagram users: absurd Sarcasm, deadpan Sarcasm, polite Sarcasm, playful Sarcasm, ironic Sarcasm, and self-deprecating Sarcasm. The study analysed different types of Sarcasm in Instagram comments. It also presented the most dominant form of self-deprecating Sarcasm (30%), which is used for humour and relevance. Additionally, exaggeration and emojis play a crucial role in amplifying the Sarcasm. The study contributes to the understanding of Sarcasm in digital spaces, taking into account such challenges and enhancing sentiment analysis. Future studies should focus on AI-driven sarcasm detection models incorporating multimodal cues to enhance accuracy. References Austin, J. L. (1962). How to do things with words. Oxford University Press. Fadilah, I., & Wijayanto, A. (2024). Sarcasm in social media: A study of comments on Sam Smith’s Instagram posts. Jurnal Onoma: Pendidikan, Bahasa, dan Sastra, 10(1), 1–15. https://doi.org/10.30605/onoma.v10i1.3077 Farabi, M. (2024). A comprehensive survey on multimodal sarcasm detection. Journal of Artificial Intelligence Research, 75(3), 210–235. Ghosh, D. (2018). Detecting Sarcasm in online conversations: Contextual approaches and challenges. Journal of Computational Linguistics, 47(2), 355–378. Goyal, I. (2022). Multimodal sarcasm and satire detection on social media platforms. Proceedings of the International Conference on Social Media Analysis, 102–115. Goyal, I., Bhandia, P., & Dulam, S. (2022). Finetuning for sarcasm detection with a pruned dataset. arXiv Preprint. https://doi.org/10.48550/arXiv.2212.12213 Grice, H. P. (1975). Logic and conversation. In P. Cole & J. Morgan (Eds.), Syntax and semantics (Vol. 3, pp. 41–58). Academic Press. Igaab, A. (2023). Sarcasm as a communicative strategy: An analysis of expressive and assertive acts. Journal of Pragmatics and Communication Studies, 15(2), 112–127. Jawaid, A., Batool, M., Arshad, W., Haq, M. I. ul, Kaur, P., & Arshad, S. (2025, January 26). English language vocabulary building trends in students of higher education institutions and a case of Lahore, Pakistan. Contemporary Journal of Social Science Review, 3(1), 730–737. https://contemporaryjournal.com/index.php/14/article/view/360 Jawaid, A., Batool, M., Arshad, W., Kaur, P., & Haq, M. I. ul. (2024). English language pronunciation challenges faced by tertiary students. Contemporary Journal of Social Science Review, 2(4), 2104–2111. https://contemporaryjournal.com/index.php/14/article/view/361 Khurdula, H. V., Naik, S. S., & Rusert, J. (2024). Sarcasm through the looking glass: Multi-domain analysis for improved detection. In Proceedings of SoutheastCon 2024 (pp. 650–654). IEEE. https://doi.org/10.1109/southeastcon52093.2024.10500167 Kumar, A., & Garg, G. (2019). Sarc-M: Sarcasm detection in typographic memes. Social Science Research Network. https://doi.org/10.2139/ssrn.3384025 Lestari, D. A. (2024). The types of irony in memes: A study on Instagram. Journal of Language and Literature Studies, 5(2), 45–56. Oshii, A. M. (2023). A deviation-based ensemble algorithm for sarcasm detection in online comments. In IEEE International Conference on Electrical, Computer and Communication Technologies (pp. 1–7). IEEE. https://doi.org/10.1109/ICECCT56650.2023.10179724 Sandor, T. (2023). Detecting Sarcasm in online comments using machine learning models. Journal of Computational Social Science, 6(1), 55–78. Sasirekha, P. (2024). Methodologies for sarcasm detection on social media: A review. International Journal of Computational Linguistics, 12(2), 89–104. Schifanella, R., de Juan, P., Tetreault, J., & Cao, L. (2016). Detecting Sarcasm in multimodal social platforms. In Proceedings of the 2016 ACM Multimedia Conference (pp. 1136–1145). ACM. https://doi.org/10.1145/2964284.2964321 Searle, J. R. (1969). Speech acts: An essay in the philosophy of language. Cambridge University Press. Srivastava, R. (2024). Understanding Sarcasm in digital discourse: A multimodal perspective. International Journal of Linguistics and Communication Studies, 14(1), 77–94. Yasmin, A. P., & Zuhriah, Z. (2024). Analysis of sarcasm language use on Instagram social media (Case study of students from the Faculty of Social Sciences, UINSU). JKOMDIS: Jurnal Ilmu Komunikasi dan Media Sosial, 4(2), 55–64. https://doi.org/10.47233/jkomdis.v4i2.1918

As the use of automated social media analysis tools surges, concerns over accuracy of analytics have increased. Some tentative evidence suggests that sarcasm alone could account for as much as a 50% drop in accuracy when automatically detecting sentiment. This paper assesses and outlines the prevalence of sarcastic and ironic language within social media posts. Several past studies proposed models for automatic sarcasm and irony detection for sentiment analysis; however, these approaches result in models trained on training data of highly questionable quality, with little qualitative appreciation of the underlying data. To understand the issues and scale of the problem, we are the first to conduct and present results of a focused manual semantic annotation analysis of two datasets of Twitter messages (in total 4334 tweets), associated with; (i) hashtags commonly employed in automated sarcasm and irony detection approaches, and (ii) tweets relating to 25 distinct events, including, scandals, product releases, cultural events, accidents, terror incidents, etc. We also highlight the contextualised use of multi-word hashtags in the communication of humour, sarcasm and irony, pointing out that many sentiment analysis tools simply fail to recognise such hashtag-based expressions. Our findings also offer indicative evidence regarding the quality of training data used for automated machine learning models in sarcasm, irony and sentiment detection. Worryingly only 15% of tweets labelled as sarcastic were truly sarcastic. We highlight the need for future research studies to rethink their approach to data preparation and a more careful interpretation of sentiment analysis.

In recent years, online opinionated textual data volume has surged, necessitating automated analysis to extract valuable insights. Data mining and sentiment analysis have become essential for analysing this type of text. Sentiment analysis is a text classification problem associated with many challenges, including better data preprocessing and sarcasm detection. Sarcasm in the text can reduce sentiment accuracy unless the model is specifically designed to identify such nuances. Sarcastic data conflicts with context, which leads to ambiguity. This article examines the impact of various preprocessing methods on sentiment classification. It presents a sarcasm detection–based sentiment analysis model that utilises an effective preprocessing system and a robust technique for identifying sarcastic text. The preprocessing method includes the impact of removing stop words, whitespaces, usernames, hashtags, and unnecessary URLs, mapping contractions dictionaries, designing word reference sheets, and stemming and lemmatization on the model. Sarcastic headlines are collected, and a word cloud is created using the frequency of words with sarcastic headlines. Custom sarcasm detection models based on long short‐term memory (LSTM), recurrent neural network (RNN), and word embeddings are utilised for sarcasm detection. The model has achieved a validation accuracy of 0.88 and an F ‐measure score of 72% for the overall system.

Machine-learning based sentiment classification has gained increasing popularity for analyzing online content in social media. A new generation of artificial neural networks is deep learning, which has been successfully applied in several domains. In this study, we empirically evaluate two state-of-the-art models of deep recurrent neural networks to detect sentiment polarity of Arabic microblogs. We considered both unidirectional and bidirectional Long Short-Term Memory (LSTM) and its simplified variant Gated Recurrent Unit (GRU). Moreover, due to the complexities and challenges facing the Arabic language to model short dialectical text, which is commonly used in microblogs, we aim to assess non-verbal features extracted from a dataset of 2091 microblogs. We also compared the performance to baseline traditional learning methods and deep neural networks. The experimental results reveal that LSTM and GRU based models significantly outperform other classifiers with a slight difference between them with best results attained when using bidirectional GRU.

In recent years, more people have been using the internet and social media to express their opinions on various subjects, such as institutions, services, or specific ideas. This increase highlights the importance of developing automated tools for accurate sentiment analysis. Moreover, addressing sarcasm in text is crucial, as it can significantly impact the efficacy of sentiment analysis models. This paper aims to provide a comprehensive overview of the conducted research on sentiment analysis and sarcasm detection, focusing on the time from 2018 to 2023. It explores the challenges faced and the methods used to address them. It conducts a comparison of these methods. It also aims to identify emerging trends that will likely influence the future of sentiment analysis and sarcasm detection, ensuring their continued effectiveness. This paper enhances the existing knowledge by offering a comprehensive analysis of 40 research works, evaluating performance, addressing multilingual challenges, and highlighting future trends in sarcasm detection and sentiment analysis. It is a valuable resource for researchers and experts interested in the field, facilitating further advancements in sentiment analysis techniques and applications. It categorizes sentiment analysis methods into ML, lexical, and hybrid approaches, highlighting deep learning, especially Recurrent Neural Networks (RNNs), for effective textual classification with labeled or unlabeled data.

• High-resolution data are employed for sea level forecasting, multistep-ahead in the Gulf of Finland. • Deep learning methods of long short-term memory networks (LSTMs) and gated recurrent units (GRU) are applied. • Correlation analysis showed that zonal wind, air pressure, and SST mostly influence sea levels. • Multivariate GRU slightly outperformed LSTM at all horizons (R 2 =0.93, RMSE=4.96 cm). • GRU, validated by satellite altimetry, shows errors within ±5cm; underestimated maxima. Changes in climate, along with increasing marine activities in coastal and offshore regions, highlight the need for effective sea level forecasting methods. In recent years, forecasting techniques, especially those utilizing machine learning/deep learning methods (ML/DL), have shown promising capabilities. However, sea level forecasting is often limited in accuracy and spatiotemporal coverage, primarily due to the challenges posed by available observational data, which complicates the assessment of existing ML/DL techniques in complex and dynamic regions like the Baltic Sea. This study addresses these challenges by utilizing a high-resolution spatiotemporal framework that integrates high-resolution hydrodynamic and marine geoid models available to Baltic countries, enabling further capabilities to be explored in terms of sea level accuracy and validation. Specifically, it examines short-term sea level forecasting in the eastern Baltic Sea and the potential of utilizing two recurrent neural network-based models such as the Long Short-Term Memory Networks (LSTMs), and the Gated Recurrent Unit (GRU) along with high-resolution input data sources. These models were specifically chosen, due to their expected capabilities with time series data and their ability to learn both short and long-term connections of the input datasets. To achieve this, a multivariate multistep-ahead (3, 6, 9, 12, and 24 h) forecasting framework was developed. The DL models' input components are high-resolution sea level data obtained from a bias-corrected hydrodynamic model, wind speed, surface pressure, and sea surface temperature. Results for various time steps (from 3 h to 24 h ahead), during the test period, revealed that the two DL models generally showed similar performance, with slightly superior results with the GRU model. For instance, GRU and LSTM showed an averaged root mean square error (RMSE) of 4.96 cm and 5.3 cm and a coefficient of determination (R²) of 0.93 and 0.92, respectively. Investigations of the time series forecasting performance at selected locations, also demonstrated the superiority of the GRU model, for all time steps, with Willmott's index (WI) values generally above 0.9 and high reliability as reflected in Prediction Interval Coverage Probability (PICP) values mostly exceeding 90 %. The results, however, weren't always perfect; both the GRU and LSTM models encountered limitations with forecasting the sea level maxima. Further examination of the spatial discrepancies also reveals some problematic areas in the eastern Gulf of Finland. This may have been influenced by the exclusion of some input components such as river discharge, salinity and meridional winds, further enhanced by complex hydrodynamics, extreme sea level variations, strong local currents, resonance-induced seiches and seasonal ice cover. In addition, an external validation of the GRU results was performed using along-track satellite altimetry from Sentinel 3A and 3B missions. For most of the satellite tracks, the discrepancy was better than 5 cm, proving the capabilities of the model generalization capabilities. These findings hold significant implications for advancing our comprehension of oceanic dynamics, enhancing maritime safety, and benefiting a wide range of applications that are dependent on accurate sea level forecasting.