The COVID-19 pandemic overwhelmed social media platforms with many discussions, requiring detailed research of the tweets connected to it. Nevertheless, existing techniques frequently struggle to rectify the disparity between useful and uninformative tweets in the training dataset, compromising the categorization’s precision. This study presents ICNM-BERT, a hybrid model that combines Convolutional Neural Networks (CNNs) with Bidirectional Encoder Representations from Transformers (BERT). ICNM-BERT seeks to improve the classification of tweets by categorizing COVID-19-related tweets into awareness, irrelevant, report, and treatment categories. Combining Convolutional Neural Networks (CNNs) with BERT can generate bidirectional encoder representations, enhancing the capabilities of text analysis and Natural Language Processing (NLP) applications. By utilizing Hybridized Deep Learning, the model efficiently combines the advantages of both architectures to handle heterogeneous data. ICNM-BERT effectively tackles socioeconomic difficulties related to tweet analysis using deep ensemble learning, as confirmed through experimental validation on real-world datasets.

Rapid advances in machine learning, deep learning, and AI are changing business and society. The history of these technologies reveals their transformative impact on a variety of industries. Convolutional, Deep, and Recurrent Neural Network studies show their evolution. Deep learning has facilitated the development of novel applications in several industries and stimulated technological progress. Reinforcement learning, a core component of artificial intelligence, has made notable progress, particularly in the realm of autonomous systems. This paper discusses the latest algorithms and their applications, stressing reinforcement learning's impact on robotics and automated decision-making. The study of natural language processing is crucial. Through language modelling, sentiment analysis, and translation, it shows banking, healthcare, and customer service applications. After that, the paper examines real-world AI applications. These technologies help doctors detect, treat, and forecast medical disorders. Fraud detection, risk management, and algorithmic trading benefit financial institutions. Industry 4.0 combines AI-driven autonomous vehicle navigation, control, and ethical decision-making with intelligent manufacturing and predictive maintenance.

With ever-growing internet users, Twitter has evolved as one of the most popular social media platforms for voicing opinions. Emotion analysis in tweets is helpful in many applications. ChatGPT, the recent AI buzz, received significant attention for its extraordinary abilities. The effectiveness of ChatGPT has not yet been evaluated for emotion analysis in tweets. This paper investigates how effective ChatGPT is for tweet emotion analysis. Experimental results show that directly applying ChatGPT in zero or few-shot settings does not yield good results. ChatGPT is more effective for data augmentation, and fine-tuning smaller pre-trained models on the augmented dataset achieves new state-of-the-art results on the TweetEval emotion analysis dataset. It is the first work to show that ChatGPT augmented tweets can help smaller pre-trained models achieve the best results for tweet emotion analysis.

Classifying tweets containing valuable information about COVID-19 is crucial for developing monitoring systems that provide the latest updates. Existing approaches for informative tweet classification considers only the last layer vector of a special token by ignoring the vectors of other tokens and the token vectors from the previous layers. The paper addresses this drawback by proposing a novel approach which (i) makes use of all the token vectors from the last four layers and (ii) leverages additional information in the form of POS tags and informative words. Experiment results show that the proposed approach outperforms all the existing approaches and achieves an accuracy of 92% and F1-score of 92.01% on the COVID-19 informative tweets dataset. The uniqueness of this paper is the attempt to leverage token vectors from the last four layers, additional information in the form of POS tags and informative words from COVID-19 informative tweets for classification.

Early warning systems for earthquakes can mitigate their destructive potential by spreading information about the quake's magnitude and location long before destructive waves reach populated areas. Source-location estimations in these systems need to be timely and accurate for them to be useful. This study presents a novel approach for enhancing the precision and speed of seismic early warning using machine learning techniques. Timely warnings may be delayed due to the precision but slowness of traditional seismic techniques for calculating earthquake sites. The purpose of the random forest (RF) model for fast earthquake localization is to aid in the quick decision making required by earthquake early warning (EEW) systems. This approach takes use of the P-wave arrival times recorded by the first five stations to record an earthquake and calculates the variations in these timings with regard to the first station. In order to determine the epicenter, the RF model categorizes these differences in Pwave arrival timings and station locations. The model is used to train and validate the proposed method using a Japanese earthquake dataset. The RF model is quite accurate in predicting earthquake epicenters, with a Mean Absolute Error (MAE) of just 2.88 kilometers. Additionally, the suggested RF model may learn from as little as 10% of the information and as little as three recording stations while still producing usable results (MAE5 km) in most cases. This novel algorithm provides a robust and flexible method for predicting the location of EEW sources in real-time.

Recent years have seen the emergence of computer vision, a subfield of artificial intelligence (AI), as a technology that has the potential to revolutionize agricultural practices. This might have an impact on many agricultural practices and crop management techniques. This is due to the fact that computer vision can examine pictures and identify patterns of data. This article provides a summary of the uses of computer vision in agriculture as well as the consequences such applications have had. The issues of precision agriculture, disease diagnosis, crop monitoring, and yield computation may be overcome with the use of computer vision technologies such as image recognition, object detection, and pattern analysis. Specifically, it investigates the ways in which these strategies are useful. In addition to this, it investigates the benefits, drawbacks, and possible future applications of computer vision in agriculture, with a particular emphasis on the potential for the sector to improve its levels of productivity, sustainability, and profitability. According to this in-depth analysis, computer vision has revolutionized the agricultural industry and contributed significantly to economic growth. In order to evaluate the financial impacts that computer vision has had on the agricultural industry, this study looks at a wide range of academic papers, publications, and reports. The findings highlight the advancements, benefits, challenges, and future opportunities presented by computer vision technology in the areas of crop monitoring, precision farming, animal management, and harvesting. According to the evaluation, computer vision has the potential to improve farming in terms of productivity, resource allocation, cost reduction, and sustainable practices.

The informality and lack of structure in social media messages make emotion detection difficult. Our study introduced a novel approach to Twitter data analysis using a Genetic Algorithm (GA). This strategy combines various information to comprehend better users' emotions, including writing style, sentiment, and linguistic patterns. A fine-tuned ensemble classifier using GA was employed to enhance the detection accuracy. Using a Twitter dataset for training and testing, we discovered that our model outperformed more conventional machine learning techniques. We explored many models and combinations, such as RF + SVM + XGB, Decision Tree, Random Forest, XGBoost, and SVM. The Average Soft-Voting Classifier (RF + MLP + LGBM) is an ensemble technique we investigated; they demonstrated promise for even better outcomes, with the ability to achieve 98% accuracy. Our results demonstrate that ensemble approaches greatly enhance the precision of social media emotion identification.

The motivation of the research work is Respiratory Rate (RR) monitoring which is susceptible to environmental and physiological stimuli, knowing it may help in assessing the health of patients. In this work, at rest, six trials on a treadmill were to be completed by 10 healthy males, while low-speed jogging in order to evaluate a new approach based on Principal Component Analysis (PCA). Random Forest (RF) with PCA for sensory selection using a special wearable system with six piezoresistive sensors. For instance, a single sensor is needed for a breathing evaluation while at rest, three sensors are needed for a low-speed walk, and four sensors are needed for a high-speed walk and run assessment. The findings may be helpful in the deployment of specialized algorithms to continuously and accurately monitor RR, as well as the creation of the best instrumented wearable devices for RR monitoring both while the subject is at rest and when they are engaging in physical activity. A smart garment was utilized to obtain the breathing information delivering six respiration signals in the three compartmental parts, and a data gathering board (DAQ NI USB6002 from National Instruments) was used to gather the reference respiratory data. The performance measurements are often calculated using it (e.g., accuracy, sensitivity, specificity, precision and F1 score).

Code-mixing refers to the blending of languages. The most serious issue with code-mixing is that people move between languages and type in English instead of transcribing words in Dravidian scripts. Traditional NLP models are trained on large monolingual datasets. Code-mixing is difficult since they can’t manage data code-mixing. A significantly less percentage of research is going on code-mixed data. In this paper, a Kannada-English code-mixed dataset is used. In this dataset, YouTube comments are categorized into a negative, positive, unknown state, not-intended language (not-Kannada) and mixed feelings, and categories based on message-level polarity. The summation of hidden layers of the pre-trained model (SHLPM) like BERT and XLM-RoBERTa classification models was used for this complex and challenging task. The SHLPM-XLM-RoBERTa has given F1-score of 64% for the Kannada-English YouTube comments dataset and outperforms other existing models.

In recent days, social media users are drastically increasing, and they are very interested in participating in discussions and expressing their feelings in the form of comments. Most of the users use their native language, which is written in English(Code-Mixed Language). But the existing sentiment classification models can analyze the text sentiment if it is in English vocabulary or the script is in the native language. If the YouTube comments are in the code-mixed language, existing methodologies’ performance is not promising. To solve this classification problem, we use the Paraphrase XLM-RoBERTa model. We train the model on Tamil, Malayalam, and Kannada code-Mixed language datasets, and achieve F1-scores of 71.1, 75.3, and 62.5 respectively. Our team ranks first, second and third on Tamil, Malayalam, and Kannada code-Mixed language datasets.

Code mixing is the mixing of language while writing text. The biggest problem in Malayalam-English code-mixing is that people switch between languages (e.g. Malayalam and English) and use English phonetic typing instead of writing Malayalam words using Dravidian scripts. Traditional NLP models are trained on extensive monolingual tools (e.g. Malayalam or English); code-mixing is challenging since they cannot handle data code-mixing. DravidianCodeMix FIRE 2020 is to classify comments into positive, negative, unknown_state, mixed_feelings and not-malayalam categories based on message-level polarity using Malayalam - English dataset. The classification model used for this challenging task is sentence-level BERT. Achieved an F1-score of 0.71 (ranked 4th) for the Malayalam-English (Manglish) comments dataset submitted under the username: CIA_NITT.

Ad hoc network is basically infrastructure less in nature and it is vulnerable to attacks. In such a network it is necessary to create the security autonomously without the involvement of centralized architecture. To solve the above issue, a novel security protocol is proposed for the artless network creation. Initial network creation done based on trust. For each newly added user authorization is performed using certificate generation. Authentication is carried out using certificate verification mechanism. Session key is used by all the nodes in the network, for the secure communication. Symmetric key encryption is used for data encryption. Asymmetric cryptography is used for session key encryption and signature generation. Session key updating is carried out based on expiration time. To revoke the attacker from participating in the network, User revocation mechanism is contributed to the proposed security protocol. Security is enhanced with revocation capabilities.