Using deep learning, a new sentiment analysis model is designed in our article. The original data (input) is first pre-processed by stemming, stop-word removal, and tokenization. The projected technique includes 4 phases: "pre-processing, feature extraction, feature selection, and sentiment classification." The characteristics “Bigram-BoW (B-BoW), Threshold Term Frequency-Inverse Document Frequency (T-TFIDF), Unigram, and N-Gram” are then retrieved from the pre-processed data. Utilizing the self-improved Honey Badger Algorithm, the best features out of the chosen features will be chosen (SI-HBA). The basic Honey Badger Algorithm (HBA) has been conceptually improved by this SI-HBA model. The review classification will then be conducted via the proposed optimized crossover framework, which is constructed by hybridizing the optimized Bi-Long Short-Term Memory (Bi-LSTM) and Deep Belief Network (DBN) trained with Transfer learning, respectively. The SI-HBA model’s optimally chosen features are used to train the hybrid classifier within the optimized crossover architecture. A self-improved Honey Badger Algorithm is used to fine-tune the weight of the Bi-LSTM classifier to improve the classification performance of the gathered reviews (SI-HBA). The final results will indicate whether the reviews are mostly good, negative, or neutral. The proposed sentiment classification model is then validated by a comparison analysis.

Abstract The probability of redundancy in questions has significantly increased due to the increasing influx of users on different cQA forums such as Quora, Stack overflow, etc. Because of this redundancy, the responses are scattered through various variations of the same question that results in unsatisfactory search results to a specific question. To address this issue, this work proposes the model for discovering the semantic similarity among the cQA questions. We followed two approaches (i) Feature-based: the question embedding is created using four forms of word embeddings and an ensemble of all four. Then Siamese LSTM (sLSTM) is used to find the semantic similarity among the questions. (ii) Fine-tuning: we fine-tuned BERT model on STS and SNLI data, which employs Siamese network architectures to generate semantically meaningful sentence embeddings. Then sBERT is used to assess the similarity between the questions. Experiments were carried out on Quora (QQP) and Stack Exchange cQA dataset with training sets of different sizes and word vectors of different dimensionalities. The model shows significant improvement over the state-of-the-artwork on sentence similarity tasks.

The method wherein a human expert diagnose a patient for COVID-19 with the help of a chest CT scan or X-ray image could be one of the most reliable methods. However, this method of diagnosis is challenging and non-scalable while considering limited medical-care infrastructure and disease spread rate. We train COVID-19 diagnosis models for classification using both the image modalities, chest CT scan and X-ray datasets. We have used fusion approach for multimodal data fusion and proposed two variants. The first model is trained using an automated deep learning approach and in the second model features from the images are extracted using transfer learning approach followed by fine tuning of model. The performance of these models are evaluated with metrics like testing accuracy, recall, precision and f1-score. False negatives are critical and to ensure a smaller number of false negatives, cost-sensitive learning is enforced. The cost-sensitive convnet model achieves an accuracy of 97%.

Reckless driving has always been a major concern for people. Every year it results in collisions and casualties. It is important to develop an automated system that detects reckless driving on highways. In order to overcome this issue, a new Deep Leaning system is designed that detects rash driving and notifies the concerned authority to take necessary action. The proposed system has two main modules: (i) Extracting the trajectory and (ii) Detecting the dents in the vehicle body. Our proposed model demonstrates the promising results in detecting the rash driving. We validated our results using YouTube videos with promising accuracy.

The rise of E-learning systems offers a large number of open online educational videos through portals like YouTube, MOOCs, which result in an enormous volume of data for every inquisitive e-learner. This work proposes a novel approach for recommending educational video clips (VCs) for children's joyful learning. It caters to the need for hard-pressed parents, who are unable to tutor their children efficaciously. As children are unable to choose the recommended VCs on their own, the proposed system offers a precise summary of the recommended VCs for parents to effectively select the learning content for their children. For a joyful learning experience, the selected VCs are then style transferred using our deep style transfer network. The extensive evaluation methods demonstrate the effectiveness and practicability of the proposed approach.

Multiple object detection using deep neural networks can lead to transportation vehicles estimate, a necessary requirement for prediction and management of road traffic and parking lot. Highly overlapped objects that look similar and objects that are there at far distances have lesser probability of detection by state-of-art techniques. We propose techniques to estimate the traffic at regions of poor detection probability in the image based on (i) density based clustering and (ii) exclusive object detection in the regions of poor detection. The proposed techniques lead to better estimation in comparison to state-of-art by approximately 12 %. We have utilized RetinaNet and YOLOv3 networks for object detection.

The two major sources of information for the knowledge seekers are the community question answers (cQA) blogs and textbooks. Textbooks play a vital role in any educational system. Many times, the textbooks that are available in the market are not adequate to fulfill the curiosity of the students, they frequently use the online question answering systems to acquire more knowledge. Due to the high volume of data, there will be high variance in the quality of questions and available answers in cQA forums, hence it takes additional effort to go through all possible question-answers for a better insight. To address this issue, this paper presents a technological solution-“A sentence-level text enrichment process” for a textbook with cQA content. We used techniques of natural language processing and data mining to extract the high-quality question-answers (QA) sets and corresponding links of cQA to enrich the textbooks. Experiments were carried out on the National Council of Educational Research and Training (NCERT) textbooks from India and Pattern Recognition and Machine Learning textbook by Christopher M Bishop, proves that we succeed to enrich textbooks on various subjects and across different grades with high-quality reference materials using automated techniques. The performance of the proposed system is evaluated using precision scores states that our method is competitive.