ABSTRACT Sarcasm, a complex linguistic form combining humor and criticism, often involves conveying meanings opposite to literal meanings, posing significant challenges in sentiment analysis. This endeavor becomes complicated when it comes to resource‐poor Indian indigenous languages like Telugu, which require more adequate resources and exhibit intricate morphology. Detecting sarcasm in such contexts necessitates the creation of a well‐balanced and annotated corpus. To address this gap, the primary aim of the paper is to curate a Telugu corpus of 10,000 conversations, including 5000 sarcastic and 5000 non‐sarcastic instances, using a multi‐annotator strategy. Additionally, this paper proposes a novel feature fusion network, that is, “FF‐Net”, that integrates neural features with manually extracted handcrafted features to detect sarcasm in Telugu conversational texts. To test and validate the proposed work, extensive experimentations were done and the results demonstrate the significance of using handcrafted features for sarcasm detection, with the proposed model achieving 95.65% accuracy, outperforming existing baselines by 2.75% and surpassing state‐of‐the‐art models by 2.80%. These results underline the efficacy of fusing automated and manual feature extraction techniques, offering a robust approach to sarcasm detection with improved performance.

In this paper, we propose a novel ensemble of Convolutional Neural Network-Long Short-Term Memory with an Extra Tree Classifier for automatic feature engineering in the spatiotemporal domain and classification of different human activities collected using inbuilt smartphone sensors. Prominent sensor-based HAR systems are trained and validated in extracted features of datasets collected in a controlled and simulated environment, which can adversely impact the reliability in real-world scenarios. Our approach utilizes unsimulated data from diverse users collected using inbuilt smartphone sensors in real world environment. Also, the proposed model was effectively trained and tested with multiple standard public datasets for validation and benchmarking with diverse evaluation metrics. With effective parameter tuning and lightweight feature extraction architecture, the proposed model achieved an average performance accuracy of 99 % with their own generated dataset, mHealth and MotionSense dataset, in comparatively optimized computational time compared to the benchmark models.

ABSTRACT Language prior is a pressing problem in the VQA domain where a model provides an answer favoring the most frequent related answer. There are some methods that are adopted to mitigate language prior issue, for example, ensemble approach, the balanced data approach, the modified evaluation strategy, and the modified training framework. In this article, we propose a VQA model, “Ensemble of Spatial and Channel Attention Network (ESC‐Net),” to overcome the language bias problem by improving the visual features. In this work, we have used regional and global image features along with an ensemble of combined channel and spatial attention mechanisms to improve visual features. The model is a simpler and effective solution than existing methods to solve language bias. Extensive experiment show a remarkable performance improvement of 18% on the VQACP v2 dataset with a comparison to current state‐of‐the‐art (SOTA) models.

The vast majority of languages in the world at present are considered to be low-resource languages. Since the availability of large parallel data is crucial for the success of most modern machine translation approaches, improving machine translation for low-resource languages is a key challenge. Most unsupervised techniques for translation benefit closely related languages with monolingual data of substantial quantity. To facilitate research in this direction for the extremely low resource language pair English ( en ) and Mizo ( lus ), we have developed a parallel and monolingual corpus for the Mizo language from various news websites. We explore Unsupervised Neural Machine Translation (UNMT) based on the developed monolingual data. We observe that cross-lingual embedding (CLWE) initializations on subword segmented data during pre-training, based on both masked language modelling and sequence-to-sequence generation tasks, improve translation performance. We experiment with cross-lingual alignment and combined alignment and joint training for learning the cross-lingual embedding representations. We also report baseline performances and the impact of CLWE initialization using semi-supervised and supervised neural machine translation. Empirical results show that both CLWE initializations work well for the distant pair English-Mizo compared to the baselines.

Human activity recognition (HAR) is a process of identifying the daily living activities of an individual using a set of sensors and appropriate learning algorithms. Most of the works on HAR are done using a mix of sensor data that is collected in a simulated environment, and due to that, the real-time recognition suffers. This article proposes an efficient adaptive batch size-based-CNN-LSTM model for recognizing different human activities in an uncontrolled environment. It uses adaptive batch sizes from 128 to 1024 for iterative model training and validation. The proposed model can handle imbalanced classes and un-normalized data efficiently. A state-of-art HAR dataset is also generated in an open environment to get the activity data of an uncontrolled scenario. With minimal data preprocessing and data augmentation, the model is tested, and the proposed model managed to get the highest accuracy of 99.29% with an average loss of 0.08$\pm$0.136%. The presented model is also tested with two public datasets named- mHealth and MotionSense and achieved an accuracy of 99.5% and 99.8%, respectively. The proposed model outperforms all the previous benchmarks and approaches by a good margin.

Machine Translation is an effort to bridge language barriers and misinterpretations, making communication more convenient through the automatic translation of languages. The quality of translations produced by corpus-based approaches predominantly depends on the availability of a large parallel corpus. Although machine translation of many Indian languages has progressively gained attention, there is very limited research on machine translation and the challenges of using various machine translation techniques for a low-resource language such as Mizo. In this article, we have implemented and compared statistical-based approaches with modern neural-based approaches for the English–Mizo language pair. We have experimented with different tokenization methods, architectures, and configurations. The performance of translations predicted by the trained models has been evaluated using automatic and human evaluation measures. Furthermore, we have analyzed the prediction errors of the models and the quality of predictions based on variations in sentence length and compared the model performance with the existing baselines.

With the advancement of image editing tools in today's world, the manipulation of images like cropping, cloning, resizing, etc., becomes an easy proposition and on the other end, checking or determining whether an image has been manipulated or not, becomes a great challenge. Copy‐move forgery in images is the most popular tampering method in which a portion of an image is copied and pasted in some other location of the same image. The detection of copy‐move forgery has become a prominent research area. This study presents a detailed review and critical discussions with pros and cons of each of copy‐move forgery detection techniques from 2007 to 2017. This study also addresses the variation in databases, issues, challenges, future directions and references in this domain.

In this study, the problem of detecting if an image has tampered is inquired; especially, the attention has been paid to the case in which the portion of an image is copied and then pasted onto another region to create a duplication or to hide some important portion of the image. The proposed copy‐move forgery detection system is based on the scale‐invariant feature transform (SIFT) features extraction and density‐based clustering algorithm. The extracted SIFT features are matched using the generalised two nearest neighbours (2NN) procedure. Thereafter, the density‐based clustering algorithm is utilised to improve the detection results. The proposed system is tested using MICC‐F220, MICC‐F2000 and MICC‐F8multi datasets. Due to the generalised 2NN matching procedure, the proposed system is able to detect multiple forgeries present in the image. Experimental results show that the performance of the system is quite satisfactory in terms of computational time as well as detection accuracy.