The manhole sealing performance of a kettle agitator directly affects the production efficiency and safety of the whole pressure vessel. This paper investigated the market demand for the reliability of manhole seal rings. First, the objective weight was determined by the rough set method and the subjective weight was determined by an analytic hierarchy process. The subjective weight and objective weight were integrated organically to obtain a scientific and effective comprehensive evaluation model. Then, the method of Quality Function Deployment is applied to transform the user requirements into the technical characteristics of the seal ring. The technical characteristics of the sealing ring are converted into the influencing factors of the reliability of the sealing ring by the transformation matrix, and the weight of the factors is determined according to the triangular fuzzy number. Finally, the most important factors affecting the reliability of the seal ring were studied by finite element analysis. This paper provides a theoretical basis for studying the reliability design of seal ring.

ETRAG is a tractor manufacturing company that has been the pride of the Algerian agricultural equipment industry for decades. However, despite the advantages that distinguish them from other industries, the length and complexity of the supply chain make it vulnerable to a variety of unexpected risks that can directly affect supply chain activities. Therefore, it is essential to explore the key risks and formulate risk mitigation actions, which are the objectives of our research. On the basis of extensive bibliographic research, the House of Risk (HOR) model, through a SCOR-based approach, was selected as a working tool in this study in order to identify the priority of risk agents and risk preventive action. The study results showed there were 13 business activities, 29 risk events, and 30 risk agents. Based on the Aggregate risk potential (ARP) value, 16 risk agents are classified as priority risks in HOR1. In HOR2, 17 preventive actions are proposed for priority risk agents to be implemented in the company in descending order based on the effectiveness to difficulty (ETD) value. There are five strategies to implement the 17 mitigation actions: operations planning, supplier relationship management, multiple sourcing, demand management, and skill and efficiency development.

Cloud Computing has provided opportunities for organizations to get rid of their infrastructure and instead utilize the services from the cloud vendors. However, the openness, the multi-tenancy nature of the cloud and the volume of the critical data that it stores lures the intruders to launch attacks on the cloud. To counter such attacks and protect the critical data in the cloud, Network Intrusion Detection System (NIDS) is used in the cloud environment. NIDS can detect these attacks in a timely manner and help to minimize the damages to the cloud resources. Various researchers have proposed NIDS models for cloud using machine learning techniques. However, the major characteristics of the cloud that can impact the performance of NIDS are the high volume of incoming network traffic and the high dimensionality of this traffic. NIDS in the cloud must have ability to process this high volume of traffic quickly and accurately. One way to do this is by reducing the number of features in the traffic data so that the NIDS have fewer features to process. However, NIDS must be able to predict the attacks using fewer features but with higher accuracy. This research work proposes a Multi-layer NIDS based on Individual Component Optimization Technique where each component is optimized individually and independently before integrating them to create a multi-layer NIDS. This model uses only 7 features from the dataset to predict the attacks with higher accuracy. Proposed model was evaluated repeatedly using NSL-KDD dataset and it outperformed the exiting Network Intrusion Detection systems in terms of number of features as well as accuracy.

When driving, the most crucial factor to consider is you and your passengers’ safety. Drivers must be kept under observation for any potential harmful act, whether intentional or inadvertent, in order to ensure a safe navigation. As a result, a real-time emotion detection system for a driver has been developed to detect, exploit, and evaluate the driver's emotional state. This paper discusses how to recognize emotions using facial expressions for application in active security systems for drivers. We discuss our research and development of a Convolutional Neural Network-based intelligent system for face image-based expression classification in this paper.

Software defect prediction is a major concern for estimating many factors of software products such as reliability maintenance, estimating the cost, and quality assurance. Under different circumstances/phases, the defects can be expected before scheduling each stage of software development. However, most of the software products are being developed by individuals, which leads to unwanted types of defects in different scenarios. Software structural quality refers to analyzing the source code, its inner structure, and compliance with the functional requirements. In this work, an Adaptive Boosting Meta-estimator has been proposed for software defect prediction using characterize code features associated with software quality. The proposed method has been tested with various performance metrics and compared with existing machine learning-based methods to prove its superiority.

Traditional pattern recognition methods rely on manual feature-extraction, which may result in the poor generalization of the model. With the increase in the popularity and success of deep learning methods, it is widely adopted in Human Activity Recognition (HAR). The ability of HAR can be extended to automated surveillance systems. In this paper, a deep learning and motion flow based Incept_LSTM is proposed. The proposed method extends the capability of pre-trained Inception-v3 and Long Short-Term Memory (LSTM). The hybridization of these models sustains a spatio-temporal convergence which is validated by the results so obtained. The proposed model is trained and validated on UCF-Crime dataset. The obtained results are then compared with the work done in the literature on the UCF-Crime dataset, KTH, and UCF-Crime2Local. It has achieved an accuracy of 98.2% and 94.57% on training and validation, respectively. Testing the effectiveness of RMSProp optimizer (as opposed to Adam) with 1e-6 learning rate has given best fit with 0.2 training and 0.38 validation loss. The model takes the advantage of motion flow computed using Lucas-Kanade Method. Motion flow is the important paradigm for considering video data. The proposed method outperforms the state-of-the-art methods in terms of accuracy, number of parameters and processing time. Also, various hyper-parameter settings are performed for the best training results.

One of the major platforms today to express your emotions, expressions, feelings, and views about something or about any product is social media. People put forth what they feel, therefore their individual posts can be judged as a re?ection of their personality, but how to judge a personality of a person based on social media posts and that too of say large data, say thousands, and produce analysis on it. In our studies, we specialize in using deep learning to build a classifier that can classify individuals according to their Myers-Briggs Type Index (MBTI) solely using textual content samples from their social media posts as an input. There are two reasons for developing such a classifier. First, due to the fact social media is so ubiquitous, one of these classifiers might have lots of statistics to degree character, permitting extra human beings to realize their MBTI character type, possibly extra reliably and faster. Our second incentive is the opportunity that our classifier can be more correct than the cutting-edge assessments, as evidenced with the aid of using the low retest mistakes price for character assessments executed with the aid of using expert psychologists. It may be possible to use our classification method to verify the accuracy of these first assessments, to give individuals more confidence in their results. In some cases, a text classification may be more comprehensive than a single personality assessment.

Artery Vein Nicking (AVN) is a retinal vessel abnormality at the crossings of the artery and vein. Early detection of the AVN is very important to predict systemic health disorders and life threatening risk factors. Such retinal vascular abnormality is identified and graded by using the proposed multi class supervised Random Forest machine learning classifier. It is trained with 4 grades of AVN abnormality on the publicly available AV Nicking dataset. The statistical, texture based, grey level and abnormality related features are extracted from the retinal AV Nicked images. The optimal feature set is selected for the classification in such a way that selected features are highly correlated with the target class and less correlated between the features using variance threshold, Pearson correlation, and information gain. Finally, various machine learning (feature based) classifiers namely Random Forest (RF), Naive-Bayes (NB), Adaboost (AB) and Quadrant Discriminative Analysis (QDA), are trained with the optimal feature set. The accuracies in classifying the AVN abnormality as normal, mild, moderate and severe by the classifiers are 94.3%, 93.6%, 70.6% and 50% respectively. The Random Forest classifier with maximum depth of 5 has achieved the highest classification accuracy while maintaining a 0.97 true positive rate at the lowest false positive rate of 0.023. The performance metric: F score for each class individually by multiclass RF classifier is 0.97 (normal), 0.95 (mild), 0.92 (moderate), and 0.94 (severe), and the average precision of the classifier is 0.92.