High-resource Languages Research Articles

Exploring Explainability in Arabic Language Models: An Empirical Analysis of Techniques

Transformer-based language models have advanced Natural Language Processing (NLP), achieving state-of-the-art results across various tasks. However, their complex architectures often obscure the decision-making processes, making transparency a critical challenge, especially in sensitive applications. The emerging field of eXplainable Artificial Intelligence (XAI) seeks to address this by enhancing model transparency. Nonetheless, the primary focus of XAI has largely been on high-resource languages, neglecting low-resource ones such as Arabic. In this paper, we first show the importance of studying XAI in the Arabic language. We then detail our methodology, which involves adapting AraBERT and AraGPT models to specific tasks including Arabic sentiment analysis and semantic question similarity. Then, we conduct an empirical study to evaluate various XAI methods, specifically gradient-based and perturbation-based approaches. These methods are assessed using two key metrics: faithfulness and plausibility. Our findings suggest that while gradient-based methods are more faithful, perturbation-based methods align better with human judgment.

Hallucinations in Large Multilingual Translation Models

Abstract Hallucinated translations can severely undermine and raise safety issues when machine translation systems are deployed in the wild. Previous research on the topic focused on small bilingual models trained on high-resource languages, leaving a gap in our understanding of hallucinations in multilingual models across diverse translation scenarios. In this work, we fill this gap by conducting a comprehensive analysis—over 100 language pairs across various resource levels and going beyond English-centric directions—on both the M2M neural machine translation (NMT) models and GPT large language models (LLMs). Among several insights, we highlight that models struggle with hallucinations primarily in low-resource directions and when translating out of English, where, critically, they may reveal toxic patterns that can be traced back to the training data. We also find that LLMs produce qualitatively different hallucinations to those of NMT models. Finally, we show that hallucinations are hard to reverse by merely scaling models trained with the same data. However, employing more diverse models, trained on different data or with different procedures, as fallback systems can improve translation quality and virtually eliminate certain pathologies.

An adaptive cyclical learning rate based hybrid model for Dravidian fake news detection

Fake news has evolved into a pervasive issue in the era of information overload, influencing public opinion and challenging the credibility of news sources. While various approaches have been proposed to combat fake news, most existing research focuses on high-resource languages, leaving low-resource languages vulnerable to misinformation. In this study, we propose a hybrid deep learning model architecture that integrates dilated temporal convolutional neural networks (DTCN), bidirectional long-short-term memory (BiLSTM), and a contextualized attention mechanism (CAM) to address the problem of detecting fake news in low-resourced Dravidian languages. DTCN is employed to capture temporal dependencies due to its sequential nature, BiLSTM is employed to seize long-range dependencies efficiently, and CAM is used to emphasize important information while downplaying irrelevant content. Additionally, we incorporate an adaptive-based cyclical learning rate with an early stopping mechanism to enhance model convergence. The results demonstrate that the proposed model surpasses the state-of-the-art and baseline models and achieves a higher average accuracy of 93.97% on the Dravidian_Fake dataset in four Dravidian languages.

Optimal Transport Posterior Alignment for Cross-lingual Semantic Parsing

Abstract Cross-lingual semantic parsing transfers parsing capability from a high-resource language (e.g., English) to low-resource languages with scarce training data. Previous work has primarily considered silver-standard data augmentation or zero-shot methods; exploiting few-shot gold data is comparatively unexplored. We propose a new approach to cross-lingual semantic parsing by explicitly minimizing cross-lingual divergence between probabilistic latent variables using Optimal Transport. We demonstrate how this direct guidance improves parsing from natural languages using fewer examples and less training. We evaluate our method on two datasets, MTOP and MultiATIS++SQL, establishing state-of-the-art results under a few-shot cross-lingual regime. Ablation studies further reveal that our method improves performance even without parallel input translations. In addition, we show that our model better captures cross-lingual structure in the latent space to improve semantic representation similarity.1

Neural Machine Translation from Bengali Language to English language and vice-versa

Bengali ranks among the first ten spoken languages in the world with a native speaker numbering about 230 million people. With UNESCO declaring 21st February as International Mother Language Day to commemorate the laying down of lives by five Bangladeshi students for the cause of their mother tongue, Bengali has come into the radar of worldwide attention . Though significant amount of prose, poetry have been written in Bengali language and large number of newspapers in Bengali get published daily, technically it is still considered a Low Resource Language (LRL) unlike English or French which are High Resource Language (HRL). The reason is not far to seek as corpora in varied domains such as short stories, sports, politics, agriculture etc is less in number and even when they are available, the size is less. Machine translation (MT) is difficult to perform in Bengali as parallel corpora from Bengali to other languages and vice versa is few and far between and when they are available they suffer from the problems of size and quality. This work is aimed at implementing one state of the art model in Neural Machine Translation (NMT) which is called the self-attention transformer model to perform translation from English to Bengali and vice versa. Though a couple of research work has been published in the recent years on MT from English to Bengali, they are mostly domain specific. This paper does not focus on any specific domain for NMT from English to Bengali and as such may be conceived as a more of general domain NMT from English to Bengali which is more difficult than domain specific NMT. Performance evaluation of the model was done using BLEU version-4 vis-à-vis translations of well known English-Bengali MTsystems.

Metadata for Efficient Management of Digital News Articles in Multilingual News Archives

The digital news preservation and management of low-resource languages are challenging tasks, especially in vast collections. Unique identification of individual digital objects is possible with well-defined attributes to assure efficient management, such as access, retrieval, preservation, usability, and transformability. The metadata element set is required to maximize the available attributes related to the digital objects. To create a comprehensive metadata set that contains all the necessary attributes and data about the digital news objects. It is more challenging and complicated when the archive contains articles from low-resourced and morphologically complex languages like Urdu and Arabic, which is difficult for machines to understand. The study presents challenges in low-resource languages (LRL) and research challenges. This metadata will help to link news articles based on similarity with other news articles stored in the digital news stories archive (DNSA) and ensures accessibility. In this study, we introduced 38 metadata elements set for the digital news stories preservation (DNSP) framework, of which 16 are explicit and 12 are implicit metadata elements. The paper presents how the digital news stories archive (DNSA) is enhanced to a multilingual archive and discusses the digital news stories extractor, which addresses major issues in implementing low-resource languages and facilitates normalized format migration. The extraction results are presented in detail for high-resource languages, that is, English, and low-resource languages (HRL), that is, Urdu and Arabic. The LRL encountered a high error rate during preservation compared to HRL, 10%, and 03%, respectively. The metadata extraction results show that HRL sources support all metadata elements as compared to LRL. The LRL has good support for explicit meta elements and many implicit meta elements with low extraction percentages. The LRL needs a more detailed study for accurate news content extraction and archiving for future access.

A novel Data and Model Centric artificial intelligence based approach in developing high-performance Named Entity Recognition for Bengali Language.

Named Entity Recognition (NER) plays a significant role in enhancing the performance of all types of domain specific applications in Natural Language Processing (NLP). According to the type of application, the goal of NER is to identify target entities based on the context of other existing entities in a sentence. Numerous architectures have demonstrated good performance for high-resource languages such as English and Chinese NER. However, currently existing NER models for Bengali could not achieve reliable accuracy due to morphological richness of Bengali and limited availability of resources. This work integrates both Data and Model Centric AI concepts to achieve a state-of-the-art performance. A unique dataset was created for this study demonstrating the impact of a good quality dataset on accuracy. We proposed a method for developing a high quality NER dataset for any language. We have used our dataset to evaluate the performance of various Deep Learning models. A hybrid model performed with the exact match F1 score of 87.50%, partial match F1 score of 92.31%, and micro F1 score of 98.32%. Our proposed model reduces the need for feature engineering and utilizes minimal resources.

T3L: Translate-and-Test Transfer Learning for Cross-Lingual Text Classification

Abstract Cross-lingual text classification leverages text classifiers trained in a high-resource language to perform text classification in other languages with no or minimal fine-tuning (zero/ few-shots cross-lingual transfer). Nowadays, cross-lingual text classifiers are typically built on large-scale, multilingual language models (LMs) pretrained on a variety of languages of interest. However, the performance of these models varies significantly across languages and classification tasks, suggesting that the superposition of the language modelling and classification tasks is not always effective. For this reason, in this paper we propose revisiting the classic “translate-and-test” pipeline to neatly separate the translation and classification stages. The proposed approach couples 1) a neural machine translator translating from the targeted language to a high-resource language, with 2) a text classifier trained in the high-resource language, but the neural machine translator generates “soft” translations to permit end-to-end backpropagation during fine-tuning of the pipeline. Extensive experiments have been carried out over three cross-lingual text classification datasets (XNLI, MLDoc, and MultiEURLEX), with the results showing that the proposed approach has significantly improved performance over a competitive baseline.

Detection of offensive terms in resource-poor language using machine learning algorithms.

The use of offensive terms in user-generated content on different social media platforms is one of the major concerns for these platforms. The offensive terms have a negative impact on individuals, which may lead towards the degradation of societal and civilized manners. The immense amount of content generated at a higher speed makes it humanly impossible to categorise and detect offensive terms. Besides, it is an open challenge for natural language processing (NLP) to detect such terminologies automatically. Substantial efforts are made for high-resource languages such as English. However, it becomes more challenging when dealing with resource-poor languages such as Urdu. Because of the lack of standard datasets and pre-processing tools for automatic offensive terms detection. This paper introduces a combinatorial pre-processing approach in developing a classification model for cross-platform (Twitter and YouTube) use. The approach uses datasets from two different platforms (Twitter and YouTube) the training and testing the model, which is trained to apply decision tree, random forest and naive Bayes algorithms. The proposed combinatorial pre-processing approach is applied to check how machine learning models behave with different combinations of standard pre-processing techniques for low-resource language in the cross-platform setting. The experimental results represent the effectiveness of the machine learning model over different subsets of traditional pre-processing approaches in building a classification model for automatic offensive terms detection for a low resource language, i.e., Urdu, in the cross-platform scenario. In the experiments, when dataset D1 is used for training and D2 is applied for testing, the pre-processing approach named Stopword removal produced better results with an accuracy of 83.27%. Whilst, in this case, when dataset D2 is used for training and D1 is applied for testing, stopword removal and punctuation removal were observed as a better preprocessing approach with an accuracy of 74.54%. The combinatorial approach proposed in this paper outperformed the benchmark for the considered datasets using classical as well as ensemble machine learning with an accuracy of 82.9% and 97.2% for dataset D1 and D2, respectively.

Multilingual personalized hashtag recommendation for low resource Indic languages using graph-based deep neural network

Users from different cultures and backgrounds often feel comfortable expressing their thoughts on trending topics by generating content in their regional languages. Recently, there has been an explosion in multilingual information, and a massive amount of multilingual textual data is added daily on the Internet. Using hashtags for multilingual low-resource content can be an effective way to overcome language barriers because it allows content to be discovered by a wider audience and makes it easier for people interested in the topic to find relevant content, regardless of the language in which it was written. To account for linguistic diversity and universal access to information, hashtag recommendation for multilingual low-resource content is essential. Several approaches have been put forth to recommend content-based and personalized hashtags for multimodal content in high-resource languages. Data availability and linguistic differences often limit the development of hashtag recommendation methods for low-resource Indic languages. Hashtag recommendation for tweets disseminated in low-resource Indic languages has seldom been addressed. Moreover, personalization and language usage aspects to recommend hashtags for tweets posted in low-resource Indic languages have yet to be explored. In view of the foregoing, we propose an automated hashtag recommendation system for tweets posted in low-resource Indic languages dubbed as TAGALOG, capable of recommending personalized and language-specific hashtags. We employ user-guided and language-guided attention mechanisms to distill indicative features from low-resource tweets according to the user’s topical and linguistic preferences. We propose a graph-based neural network to mine users’ posting behavior by connecting historical tweets of a particular user and language relatedness by linking tweets according to language families, i.e., Indo-Aryan and Dravidian. Experimental results on the curated dataset from Twitter demonstrate that the proposed model outperformed recognized pre-trained language models and extant research, showing an average improvement of 12.3% and 12.8% in the F1-score, respectively. TAGALOG recommends hashtags that align with the user’s interests and linguistic predilections, leading to a heightened level of tailored and engaging user experience. Personalized and multilingual hashtag recommendation systems for low-resource Indic languages can help to improve the discoverability and relevance of content in these languages.

Fake news detection in Dravidian languages using transfer learning with adaptive finetuning

Fake news has become a major challenge for online platforms and society as a whole, with potentially harmful consequences for individuals and organizations. While there has been a lot of research on detecting fake news in high-resource languages, very little attention has been paid to low-resource languages. Due to a lack of corpora and annotated data, the classification of fake news in low-resource languages remains in its infancy. In this research, we present a novel transfer learning strategy for detecting fake news in Dravidian languages. We introduced a Dravidian_Fake a new dataset for fake news classification in Dravidian languages, and we created multilingual datasets by combining the English ISOT with the Dravidian_Fake datasets. We fine-tuned the mBERT and XLM-R pretrained transformer models with adaptive learning using English and Dravidian language fake news datasets. The classification model is evaluated using transfer learning, and the suggested model outperforms current approaches and provides a viable solution for sentence-level fake news classification in a resource-constrained environment. Experimental results on a Dravidian fake news dataset of low-resource languages demonstrate the efficacy of our approach in detecting fake news with an average accuracy of 93.31 percent in multilingual transfer learning.

Understanding the Research Challenges in Low-Resource Language and Linking Bilingual News Articles in Multilingual News Archive

The developed world has focused on Web preservation compared to the developing world, especially news preservation for future generations. However, the news published online is volatile because of constant changes in the technologies used to disseminate information and the formats used for publication. News preservation became more complicated and challenging when the archive began to contain articles from low-resourced and morphologically complex languages like Urdu and Arabic, along with English news articles. The digital news story preservation framework is enriched with eighteen sources for Urdu, Arabic, and English news sources. This study presents challenges in low-resource languages (LRLs), research challenges, and details of how the framework is enhanced. In this paper, we introduce a multilingual news archive and discuss the digital news story extractor, which addresses major issues in implementing low-resource languages and facilitates normalized format migration. The extraction results are presented in detail for high-resource languages, i.e., English, and low-resource languages, i.e., Urdu and Arabic. LRLs encountered a high error rate during preservation compared to high-resource languages (HRLs), corresponding to 10% and 03%, respectively. The extraction results show that few news sources are not regularly updated and release few new news stories online. LRLs require more detailed study for accurate news content extraction and archiving for future access. LRLs and HRLs enrich the digital news story preservation (DNSP) framework. The Digital News Stories Archive (DNSA) preserves a huge number of news articles from multiple news sources in LRLs and HRLs. This paper presents research challenges encountered during the preservation of Urdu and Arabic-language news articles to create a multilingual news archive. The second part of the paper compares two bilingual linking mechanisms for Urdu-to-English-language news articles in the DNSA: the common ratio measure for dual language (CRMDL) and the similarity measure based on transliteration words (SMTW) with the cosine similarity measure (CSM) baseline technique. The experimental results show that the SMTW is more effective than the CRMDL and CSM for linking Urdu-to-English news articles. The precision improved from 46% and 50% to 60%, and the recall improved from 64% and 67% to 82% for CSM, CRMDL, and SMTW, respectively, with improved impact of common terms as well.

Multilingual NLP

The subject area of multilingual natural language processing (NLP) is concerned with the processing of natural language data in several languages. NLP systems that can translate between languages are becoming more and more necessary as the globe gets more interconnected in order to promote understanding and communication among speakers of various languages. To be effective, communication must overcome a number of obstacles presented by multilingual NLP. Lack of language standardization, which results in major variations in the grammatical constructions, vocabulary, and writing systems used in many languages, is one of the fundamental problems. The requirement for substantial amounts of annotated data for machine learning model training presents another difficulty. The creation of high-quality annotated datasets in numerous languages is time- and money-consuming, which restricts the supply of multilingual NLP resources. The problem of creating NLP systems that can handle several languages at once is the last one. This necessitates the deployment of sophisticated algorithms that can handle and evaluate data in numerous languages while producing precise findings. Researchers and developers are working on a variety of methods to address these issues. Creating standardized formats for multilingual data representation, like Universal Dependencies, which offers a unified framework for annotating linguistic data in several languages, is one strategy. Using transfer learning techniques to transfer knowledge from high-resource languages to low-resource languages is an alternative strategy. The amount of annotated data required for training NLP models in low-resource languages can bede creased with the use of this method. Last but not least, researchers are working to create multilingual NLP models that can manage numerous languages at once. To deliver precise results across numerous languages, these models employ cutting-edge methodologies like neural machine translation and multilingual word embedding’s. Despite the fact that multilingual NLP presents a number of difficult issues, with continuing study and development, it is possible to create NLP systems that are capable of processing natural language data from several languages.

Improving Multilingual Neural Machine Translation System for Indic Languages

The Machine Translation System (MTS) serves as effective tool for communication by translating text or speech from one language to another language. Recently, neural machine translation (NMT) has become popular for its performance and cost-effectiveness. However, NMT systems are restricted in translating low-resource languages as a huge quantity of data is required to learn useful mappings across languages. The need for an efficient translation system becomes obvious in a large multilingual environment like India. Indian languages (ILs) are still treated as low-resource languages due to unavailability of corpora. In order to address such an asymmetric nature, the multilingual neural machine translation (MNMT) system evolves as an ideal approach in this direction. The MNMT converts many languages using a single model, which is extremely useful in terms of training process and lowering online maintenance costs. It is also helpful for improving low-resource translation. In this article, we propose an MNMT system to address the issues related to low-resource language translation. Our model comprises two MNMT systems, i.e., for English-Indic (one-to-many) and for Indic-English (many-to-one) with a shared encoder-decoder containing 15 language pairs (30 translation directions). Since most of IL pairs have a scanty amount of parallel corpora, not sufficient for training any machine translation model, we explore various augmentation strategies to improve overall translation quality through the proposed model. A state-of-the-art transformer architecture is used to realize the proposed model. In addition, the article addresses the use of language relationships (in terms of dialect, script, etc.), particularly about the role of high-resource languages of the same family in boosting the performance of low-resource languages. Moreover, the experimental results also show the advantage of back-translation and domain adaptation for ILs to enhance the translation quality of both source and target languages. Using all these key approaches, our proposed model emerges to be more efficient than the baseline model in terms of evaluation metrics, i.e., BLEU (BiLingual Evaluation Understudy) score for a set of ILs.

Contextual Urdu Text Emotion Detection Corpus and Experiments using Deep Learning Approaches

Textual emotion detection aims to discover human emotions from written text. Textual emotion detection is a significant challenge due to the unavailability of facial and voice expressions. Considerable research has been done to identify textual emotions in high-resource languages such as English, French, Chinese, and others. Despite having over 300 million speakers and large volumes of literature available online, Urdu has not been properly investigated for the textual emotion detection task. To address this gap, this study makes two contributions: (1) the creation of a novel dialog-based corpus for Urdu (Contextual Urdu Text Emotion Detection Corpus). CUTEC contains 30,160 training and 5,509 testing labelled dialogues, where each dialogue consists of three Urdu contextual sentences. In addition, all dialogues are labelled using four emotion classes, i.e., Happy, Sad, Angry, and Other. As a second contribution (2) five deep learning models, i.e., RNN, LSTM, Bi- LSTM, GRU, and Bi-GRU have been trained and tested using CUTEC with different parametric settings. The highest results (Accuracy = 87.28 and F1 = 0.87) are attained using a GRU-based architecture.

Contextual Embeddings based on Fine-tuned Urdu-BERT for Urdu threatening content and target identification

Identification of threatening text on social media platforms is a challenging task. Contrary to the high-resource languages, the Urdu language has very limited such approaches and the benchmark approach has an issue of inappropriate data annotation. Therefore, we present robust threatening content and target identification as a hierarchical classification model for Urdu tweets. This study investigates the potential of the Urdu-BERT (Bidirectional Encoder Representations from Transformer) language model to learn universal contextualized representations aiming to showcase its usefulness for binary classification tasks of threatening content and target identification. We propose to exploit a pre-trained Urdu-BERT as a transfer learning model after fine-tuning its parameters on a newly designed Urdu corpus from the Twitter platform. The proposed dataset contains 2,400 tweets manually annotated as threatening or non-threatening at the first level and threatening tweets are further categorized into individual or group at the second level. The performance of fine-tuned Urdu-BERT is compared with the benchmark study and other feature models. Experimental results show that the fine-tuned Urdu-BERT model achieves state-of-the-art performance by obtaining 87.5% accuracy and 87.8% F1-score for threatening content identification and 82.5% accuracy and 83.2% F1-score for target identification task. Furthermore, the proposed model outperforms the benchmark study.

SexWEs: Domain-Aware Word Embeddings via Cross-Lingual Semantic Specialisation for Chinese Sexism Detection in Social Media

The goal of sexism detection is to mitigate negative online content targeting certain gender groups of people. However, the limited availability of labeled sexism-related datasets makes it problematic to identify online sexism for low-resource languages. In this paper, we address the task of automatic sexism detection in social media for one low-resource language -- Chinese. Rather than collecting new sexism data or building cross-lingual transfer learning models, we develop a cross-lingual domain-aware semantic specialisation system in order to make the most of existing data. Semantic specialisation is a technique for retrofitting pre-trained distributional word vectors by integrating external linguistic knowledge (such as lexico-semantic relations) into the specialised feature space. To do this, we leverage semantic resources for sexism from a high-resource language (English) to specialise pre-trained word vectors in the target language (Chinese) to inject domain knowledge. We demonstrate the benefit of our sexist word embeddings (SexWEs) specialised by our framework via intrinsic evaluation of word similarity and extrinsic evaluation of sexism detection. Compared with other specialisation approaches and Chinese baseline word vectors, our SexWEs shows an average score improvement of 0.033 and 0.064 in both intrinsic and extrinsic evaluations, respectively. The ablative results and visualisation of SexWEs also prove the effectiveness of our framework on retrofitting word vectors in low-resource languages.

Cross-Lingual and Cross-Domain Crisis Classification for Low-Resource Scenarios

Social media data has emerged as a useful source of timely information about real-world crisis events. One of the main tasks related to the use of social media for disaster management is the automatic identification of crisis-related messages. Most of the studies on this topic have focused on the analysis of data for a particular type of event in a specific language. This limits the possibility of generalizing existing approaches because models cannot be directly applied to new types of events or other languages. In this work, we study the task of automatically classifying messages that are related to crisis events by leveraging cross-language and cross-domain labeled data. Our goal is to make use of labeled data from high-resource languages to classify messages from other (low-resource) languages and/or of new (previously unseen) types of crisis situations. For our study we consolidated from the literature a large unified dataset containing multiple crisis events and languages. Our empirical findings show that it is indeed possible to leverage data from crisis events in English to classify the same type of event in other languages, such as Spanish and Italian (80.0% F1-score). Furthermore, we achieve good performance for the cross-domain task (80.0% F1-score) in a cross-lingual setting. Overall, our work contributes to improving the data scarcity problem that is so important for multilingual crisis classification. In particular, mitigating cold-start situations in emergency events, when time is of essence.

English–Assamese neural machine translation using prior alignment and pre-trained language model

In a multilingual country like India, automatic natural language translation plays a key role in building a community with different linguistic people. Many researchers have explored and improved the translation process for high-resource languages such as English, German, etc., and achieved state-of-the-art results. However, the unavailability of adequate data is the prime obstacle to automatic natural language translation of low-resource north-eastern Indian languages such as Mizo, Khasi, and Assamese. Though the recent past has witnessed a deluge in several automatic natural language translation systems for low-resource languages, the low values of their evaluation measures indicate the scope for improvement. In the recent past, the neural machine translation approach has significantly improved translation quality, and the credit goes to the availability of a huge amount of data. Subsequently, the neural machine translation approach for low-resource language is underrepresented due to the unavailability of adequate data. In this work, we have considered a low-resource English–Assamese pair using the transformer-based neural machine translation, which leverages the use of prior alignment and a pre-trained language model. To extract alignment information from the source–target sentences, we have used the pre-trained multilingual contextual embeddings-based alignment technique. Also, the transformer-based language model is built using monolingual target sentences. With the use of both prior alignment and a pre-trained language model, the transformer-based neural machine translation model shows improvement, and we have achieved state-of-the-art results for the English-to-Assamese and Assamese-to-English translation, respectively.

Context-aware Emotion Detection from Low-resource Urdu Language Using Deep Neural Network

Emotion detection (ED) plays a vital role in determining individual interest in any field. Humans use gestures, facial expressions, and voice pitch and choose words to describe their emotions. Significant work has been done to detect emotions from the textual data in English, French, Chinese, and other high-resource languages. However, emotion classification has not been well studied in low-resource languages (i.e., Urdu) due to the lack of labeled corpora. This article presents a publicly available Urdu Nastalique Emotions Dataset (UNED) of sentences and paragraphs annotated with different emotions and proposes a deep learning (DL)-based technique for classifying emotions in theUNEDcorpus. Our annotatedUNEDcorpus has six emotions for both paragraphs and sentences. We perform extensive experimentation to evaluate the quality of the corpus and further classify it using machine learning and DL approaches. Experimental results show that the developed DL-based model performs better than generic machine learning approaches with an F1 score of 85% on the UNED sentence-based corpus and 50% on the UNED paragraph-based corpus.