Intelligent Identification System Research Articles

An Intelligent Tutoring System for Identification of Learning Styles and Assignment Educational Strategies

The possibility of improving the learning that a student obtains considering the learning styles that he or she has, refers to the best way to retain learning based on 4 primary stimuli, such as sight, hearing, touch, and experience. In this research work, an intelligent tutoring system is described based on 2 small surveys. The system will obtain sufficient information to determine those educational strategies or techniques that favor not only the students as individuals but the classroom as a group, also favoring the teacher a route learning process focused on how students learn best. To obtain educational strategies, the intelligent tutor system makes use of one of the multiple techniques of artificial intelligence such as fuzzy logic, necessary to obtain a result with greater precision, an issue that would hardly be possible using basic logic or conventional due to the uncertainty it generates. In turn, the software is developed using the Laravel work environment using MySQL as the main database manager.

An intelligent lithology recognition system for continental shale by using digital coring images and convolutional neural networks

Lithology identification is a critical component of reservoir evaluation and hydrocarbon development. However, the continental shale rocks, characterized by complex lithologies and developed laminations, present new challenges to traditional lithology identification techniques. In this study, we develop a centimeter-level intelligent identification system for continental shale lithology, leveraging digital image preprocessing and deep learning methods. Specifically, we construct a dataset from 400 m of drilling cores, encompassing 24 types of lithology and 152 lithological transitions. The downsampling and oversampling methods are employed to balance the dataset. The digital image denoising and restoration techniques are then utilized to enhance rock image features. Different neural network models are examined to obtain the optimum predictive results. The results indicate that data balancing improves the TOP1 and TOP5 accuracies by 10% and 5%, respectively. Gaussian low-pass filter denoising, in comparison to other methods, increases the TOP1 and TOP5 accuracies by 2%, establishing it as the superior denoising technique for lithology recognition. Considering both computational efficiency and prediction accuracy, EfficientNet emerges as the most effective model for continental shale image recognition. The optimized model attains up to 60% TOP1 and 95% TOP5 across 22 categories in drilling core samples. This study delves into the intelligent identification of macroscopic drilling core images of continental shale, and the findings can offer technical support for the interpretation and evaluation of continental shale reservoirs.

A facile fabrication strategy constructed multilayer piezoresistive pressure sensor for intelligent recognition system towards privacy protection

Privacy and security concerns have become a focal point of attention for humans, so that it is crucial to develop a secure and efficient identity recognition smart system. This work proposes a piezoresistive pressure sensor (PPS) based intelligent recognition system for privacy protection, wherein the PPS consists of multiple layers of electrospun fiber membranes with a bionic pine needle shaped structure. The PPS exhibits high sensitivity of 9040.46 kPa−1 at 0–1 kPa, 679.7 kPa−1 at 1–7 kPa, and 50.04 kPa−1 at 7–20 kPa, with a fast response/recovery time of 30 ms and excellent stability. Given these prominent features, the specific application in privacy protection has been construced. PPS was firstly exploited as an intelligent handwriting pad to obtain handwriting signals from different users and combined with convolutional neural network (LeNet-5) and long short-term memory (LSTM) network models to establish a handwriting recognition intelligent system for accurate identification of user identities. Further, a password unlocking control system was succesfully created based on the PPS array and support vector machine (SVM) network model for dual verification of passwords and behavior patterns. This work can easily and effectively detect people’s daily behavioral characteristics for identity authentication, and has enormous application potential in fields such as handwriting recognition, security defense, and privacy protection.

An intelligent identification and classification system of decoration waste based on deep learning model

Efficient sorting and recycling of decoration waste are crucial for the industry's transformation, upgrading, and high-quality development. However, decoration waste can contain toxic materials and has greatly varying compositions. The traditional method of manual sorting for decoration waste is inefficient and poses health risks to sorting workers. It is therefore imperative to develop an accurate and efficient intelligent classification method to address these issues. To meet the demand for intelligent identification and classification of decoration waste, this paper applied the deep learning method You Only Look Once X (YOLOX) to the task and proposed an identification and classification framework of decoration waste (YOLOX-DW framework). The proposed framework was validated and compared using a multi-label image dataset of decoration waste, and a robot automatic sorting system was constructed for practical sorting experiments. The research results show that the proposed framework achieved a mean average precision (mAP) of 99.16 % for different components of decoration waste, with a detection speed of 39.23 FPS. Its classification efficiency on the robot sorting experimental platform reached 95.06 %, indicating a high potential for application and promotion. This provides a strategy for the intelligent detection, identification, and classification of decoration waste.

Design of Online Intelligent Detection System for Power Quality and Fault Identification in Distribution Networks

Abstract This paper lays the foundation for the subsequent analysis of power quality in distribution networks by briefly describing the indicators of power quality. Utilizing the zero sequence voltage mutation detection fault of a small current routing algorithm, the instantaneous values of voltage and current of each phase of the three-phase circuit are quickly calculated based on the electrical parameters of instantaneous reactive power and the online intelligent detection system is established. The effectiveness and feasibility of the designed detection system and detection algorithm in detecting electrical energy parameters have been verified through experiments. The results show that the error of voltage measurement is within 0.0003-0.002, the frequency error of the system is up to 0.0117%, the measurement error of three-phase unbalance is distributed between 0.001 and 0.1, and the error of negative sequence and zero sequence is between ±0.003, which meet the system design requirements. It shows that the use of online intelligent detection systems can improve the quality of distribution network operation and fault identification accuracy, improve power quality, and effectively improve the reliability of power supply.

CHEIC: Chemical Image Classificator. An intelligent system for identification of volatiles compounds with potential for respiratory diseases using Deep Learning

CHEIC: Chemical Image Classificator. An intelligent system for identification of volatiles compounds with potential for respiratory diseases using Deep Learning

AN INTELLIGENT MULTI-SENSOR SYSTEM FOR IDENTIFICATION AND ASSESSMENT OF THE TECHNICAL CONDITION OF ELECTRICAL EQUIPMENT USING WI-FI

The modern stage of industrial development is characterized by an increasing need for systems to monitor the condition and efficient functioning of electrical equipment. Breakdown or unplanned stoppage of an electromechanical system engine is accompanied by significant economic losses for enterprises. Therefore, the current stage of industrial development is characterized by an increasing need for systems for monitoring the condition and efficient functioning of electrical equipment.
 As a result, due to the rapid development of precision manufacturing technologies, many studies have been conducted in the field of multi-sensor measurement and data fusion technologies to improve monitoring capabilities in terms of measurement accuracy and information richness, thereby increasing production efficiency and accuracy. In a multi-sensor system, each sensor independently measures certain parameters. The system then uses an appropriate signal processing algorithm to combine all the independent measurements into a complete set of measurement results.
 An accurate assessment of the technical condition makes it possible to predict the term of the next repair of the equipment, increases the reliability and efficiency of its operation, reduces the damage from downtime of the equipment due to accidents, the costs of repair and restoration of operability, reduces the costs of maintenance and operation of the equipment.
 This article presents an overview of an intelligent multi-sensor system for monitoring the process and technological state of electrical equipment. Which in real time with the help of sensors makes it possible to obtain information from the environment and draw conclusions about the condition of the diagnosed equipment. The technology of an intelligent multi-sensor system for identification and assessment of the technical condition of electrical equipment has prospects for wide application in industries. Thanks to the most compact dimensions of the installation, we have the opportunity to take measurements in hard-to-reach places, without removing the equipment from work, which minimizes losses from downtime. Also, the use of this system allows detecting the development of a defect in the early stages of development, which prevents and significantly reduces the cost of repairing power equipment when it is out of service.

An intelligent system for high-density small target pest identification and infestation level determination based on an improved YOLOv5 model

Purpose:A deep learning-based intelligent system has been developed for the identification and detection of high-density small target pests with the aim of addressing the limitations observed in previous detection systems and manual sorting. The novel system promises to overcome issues of low detection accuracy, low detection efficiency, and serious leakage and error detection. It is expected to significantly improve the efficiency of pest detection, thereby offering a potential solution to the challenges posed by the presence of these pests in agricultural settings. Methods:In this paper, we propose an enhanced YOLOv5s model to tackle challenges in pest detection, such as small targets, high densities, and species diversity. Our method embeds Channel Attention (CA) modules into the algorithm and broadens the shallow feature detection scale of the original FPNet, bolstering small-target detection. Network convergence is accelerated using the Kmeans++ algorithm for prior frame generation and YOLOv5s’ intrinsic weights for transfer learning. The Efficient Intersection over Union (EIOU) loss function is adopted to address unbalanced data labels. Furthermore, we develop an algorithm to determine pest infestation levels, transforming detection results into pest grade evaluations. These advancements form the core of our proposed intelligent pest identification and detection system. Results:In our enhanced YOLOv5s model, we assessed the system’s performance using a dataset comprising six prevalent pest species. This dataset was specifically curated to emulate real-life detection scenarios, featuring small pest targets with similar appearances and imbalanced data labels. We compared our model with a common one-stage model and noted average improvements of 5.5%, 2%, and 3.95% in accuracy, recall, and detection precision respectively across both sparse and dense scenarios. Importantly, the enhanced model maintained a detection time of roughly 10ms per image and was only 1MB larger than the original model, illustrating a balance between accuracy and efficiency. Furthermore, our algorithm enables the assessment of pest grades based on the input detection results, illustrating the degree of pest infestation. Conclusion:The present paper describes the development of an intelligent system for the automatic identification and detection of pests. Experimental results demonstrate the high efficiency and efficacy of the proposed system in identifying and detecting common pests, even in complex scenarios characterized by high pest density and data label imbalance. Notably, the proposed system has the potential to assist practitioners in the fields of agriculture and forestry to make informed decisions on pest control, thereby improving work efficiency and productivity.

Intelligent identification system of gastric stromal tumors based on blood biopsy indicators

BackgroundThe most prevalent mesenchymal-derived gastrointestinal cancers are gastric stromal tumors (GSTs), which have the highest incidence (60–70%) of all gastrointestinal stromal tumors (GISTs). However, simple and effective diagnostic and screening methods for GST remain a great challenge at home and abroad. This study aimed to build a GST early warning system based on a combination of machine learning algorithms and routine blood, biochemical and tumour marker indicators.MethodsIn total, 697 complete samples were collected from four hospitals in Gansu Province, including 42 blood indicators from 318 pretreatment GST patients, 180 samples of gastric polyps and 199 healthy individuals. In this study, three algorithms, gradient boosting machine (GBM), random forest (RF), and logistic regression (LR), were chosen to build GST prediction models for comparison. The performance and stability of the models were evaluated using two different validation techniques: 5-fold cross-validation and external validation. The DeLong test assesses significant differences in AUC values by comparing different ROC curves, the variance and covariance of the AUC value.ResultsThe AUC values of both the GBM and RF models were higher than those of the LR model, and this difference was statistically significant (P < 0.05). The GBM model was considered to be the optimal model, as a larger area was enclosed by the ROC curve, and the axes indicated robust model classification performance according to the accepted model discriminant. Finally, the integration of 8 top-ranked blood indices was proven to be able to distinguish GST from gastric polyps and healthy people with sensitivity, specificity and area under the curve of 0.941, 0.807 and 0.951 for the cross-validation set, respectively.ConclusionThe GBM demonstrated powerful classification performance and was able to rapidly distinguish GST patients from gastric polyps and healthy individuals. This identification system not only provides an innovative strategy for the diagnosis of GST but also enables the exploration of hidden associations between blood parameters and GST for subsequent studies on the prevention and disease surveillance management of GST. The GST discrimination system is available online for free testing of doctors and high-risk groups at https://jzlyc.gsyy.cn/bear/mobile/index.html.

Deep learning-based intelligent system for fingerprint identification using decision-based median filter

Fingerprint recognition has emerged as one of the most reliable biometric authentication methods, owing to its uniqueness and permanence. However, the security and confidentiality of the user’s data are key considerations in modern biometric systems. In this study, we describe an intelligent computational technique for automatically validating fingerprints for identification and verification purposes. The feature vector is created by fusing Gabor filtering features with deep learning techniques like the faster region-based convolutional neural network (Faster R-CNN). This study uses linear and decision-based median filtering (DBMF) techniques to minimize visual impulse noise. Faster-R-CNN with DBMF was applied to the feature vectors to reduce overfitting problems while improving classification precision and reliability. For fingerprint matching, the Euclidean distance between the associated Harris-SURF feature vectors of two feature points is used to measure feature-matching similarity between two fingerprint images. Furthermore, for fine-tuned matching an iterative technique known as RANSAC (Random Sample Consensus) is used. The experimental results collected from the public-domain fingerprint databases FVC-2002 DB1 and FVC-2000 DB1 show that the proposed design is viable and performs well with an accuracy of 99.43%, MSE value of 43.321%, and an execution time of 3.102 ms which was more exact than existing models.

Deep learning based intelligent identification system for ripening stages of banana

Ripening of banana hands during handling, on board transit, shipping and storage leads to higher post-harvest loss and impede the trade. Identification of ripening is paramount importance to reduce loss. Bulk handlers and food processing industries requires automated non-destructive methods of ripening stage identification methodologies. This paper proposes a deep learning based non-destructive method of classification of banana fruit under four categories – unripe, under ripe, ripe and over ripe. A customized dataset was prepared with sufficient images in each class. A convolution neural network (CNN) combined with an eXtreme Gradient Boosting (XgBoost) algorithm (CNN-XgBoost) is introduced for the effective determination of the ripening stage of banana. CNN acts as the trainable feature extractor of the images and XgBoost acts as the identifier of ripening stage. Linear Discriminant Analysis (LDA) is incorporated in order to eliminate the need to have data augmentation or a huge data set. Thus, the proposed deep learning approach possesses capability to perform classification even with a smaller data set compared to conventional deep and machine learning techniques. The performance accuracy of the proposed duo is found to be 91.25 % and it is higher than that obtained with a Support Vector Classifier (SVC), Gaussian Naive Bayesian Classifier (GNB) or k-Nearest Neighbours (KNN) algorithms.

Non-destructive testing technology for intelligent identification of foreign objects in cosmetics based on BP algorithm

To solve the problem that the presence of foreign matters in cosmetics will affect the safety and health of consumers and is not conducive to the development of the cosmetics industry, an intelligent identification system for foreign matters in cosmetics is established using the improved BP algorithm. Scan cosmetic samples to identify foreign matters and extract foreign matter features, so as to achieve non-destructive detection of foreign matters in cosmetics. Comparing the traditional BP algorithm, Faster R-CNN algorithm and the improved BP algorithm, the results show that the convergence time of the improved BP algorithm is 60 s and 30 s earlier than that of the traditional BP algorithm and Faster R-CNN algorithm respectively; Whether there is noise or not, the recognition rate of the improved BP algorithm is always higher than that of the traditional BP algorithm and Faster R-CNN algorithm. The accuracy rate of the improved BP algorithm is between 0.88 and 0.96, the accuracy rate of the traditional BP algorithm is between 0.57 and 0.75, and the accuracy rate of the Faster R-CNN algorithm is between 0.76 and 0.81. This shows that the improved BP algorithm can realize the nondestructive detection of foreign matters in cosmetics, ensure a high accuracy and fast speed, and provide consumers with a sense of safe use of cosmetics, it can also improve consumers’ satisfaction with the use of cosmetic products.

Deep learning‐based recognition method of red bed soft rock image

In order to improve the investigation efficiency, improve the traditional engineering investigation work mode, realize the intelligent investigation, and improve the economic and social benefits of enterprises, a red soft rock image intelligent analysis and recognition system is proposed based on deep learning methods. The intelligent recognition system includes two core algorithms: soft rock image decomposition and soft rock lithology/weathering degree recognition. The research shows that the identification model of weathering degree and the lithology identification model based on the convolutional neural network (CNN) algorithm have a good identification effect, and the identification probability of moderately weathered rock reaches 95.22% and the accuracy of lithology identification is 91.34%. With the increase in training data, the recognition effect will be further improved. The intelligent identification system has been integrated into the WeChat mini programme, App, and Web system, which can be directly applied to field geological survey operations to assist geological workers in the investigation work, and realize the intelligent identification and classification of red bed soft rock.

An intelligent identification and classification system for malicious uniform resource locators (URLs).

Uniform Resource Locator (URL) is a unique identifier composed of protocol and domain name used to locate and retrieve a resource on the Internet. Like any Internet service, URLs (also called websites) are vulnerable to compromise by attackers to develop Malicious URLs that can exploit/devastate the user's information and resources. Malicious URLs are usually designed with the intention of promoting cyber-attacks such as spam, phishing, malware, and defacement. These websites usually require action on the user's side and can reach users across emails, text messages, pop-ups, or devious advertisements. They have a potential impact that can reach, in some cases, to compromise the machine or network of the user, especially those arriving by email. Therefore, developing systems to detect malicious URLs is of great interest nowadays.This paper proposes a high-performance machine learning-based detection system to identify Malicious URLs. The proposed system provides two layers of detection. Firstly, we identify the URLs as either benign or malware using a binary classifier. Secondly, we classify the URL classes based on their feature into five classes: benign, spam, phishing, malware, and defacement. Specifically, we report on four ensemble learning approaches, viz. the ensemble of bagging trees (En_Bag) approach, the ensemble of k-nearest neighbor (En_kNN) approach, and the ensemble of boosted decision trees (En_Bos) approach, and the ensemble of subspace discriminator (En_Dsc) approach. The developed approaches have been evaluated on an inclusive and contemporary dataset for uniform resource locators (ISCX-URL2016). ISCX-URL2016 provides a lightweight dataset for detecting and categorizing malicious URLs according to their attack type and lexical analysis. Conventional machine learning evaluation measurements are used to evaluate the detection accuracy, precision, recall, F Score, and detection time. Our experiential assessment indicates that the ensemble of bagging trees (En_Bag) approach provides better performance rates than other ensemble methods. Alternatively, the ensemble of the k-nearest neighbor (En_kNN) approach provides the highest inference speed. We also contrast our En_Bag model with state-of-the-art solutions and show its superiority in binary classification and multi-classification with accuracy rates of 99.3% and 97.92%, respectively.

An Adaptive Identification Method for Potential Landslide Hazards Based on Multisource Data

The effectiveness of landslide disaster prevention depends largely on the quality of early identification of potential hazards, and how to comprehensively, deeply, and accurately identify such hazards has become a major difficulty in landslide disaster management. Existing deep learning methods for potential landslide hazard identification often use fixed-size window modeling and ignore the different window sizes required by landslides of different scales. To address this problem, we propose an adaptive identification method for potential landslide hazards based on multisource data. Taking Yongping County, China, as the study area, we create a multisource factor dataset based on the landslide disaster background in terms of topography, geology, human activities, hydrology, and vegetation as the sample for the identification model after processing. Moreover, we combine differential interferometric synthetic aperture radar (D-InSAR) and multitemporal InSAR (MT-InSAR) to process the surface deformation of the study area, and we measure the deformation richness based on the average of the pixel deformation difference within the current window of a pixel point in the image. Therefore, convolutional neural networks (CNNs) with different window sizes are adaptively selected. The results show that the precision of adaptive identification of potential landslide hazards in the study area is 85.30%, the recall is 83.03%, and the F1 score is 84.15%. The recognition rate for potential hazards reaches 80%, which is better than the fixed-window modeling result and proves the effectiveness of the proposed method. This method can help to improve intelligent identification systems for potential landslide hazards, and also contribute to the identification of other potential geological hazards, such as mudslides and collapses.

Non-intrusive Load Identification Algorithm Based on Stacking Modeling

The non-intrusive power load identification relies on the smart meter’s voltage, current and power signals, which can realize the classification and monitoring of the domestic electric load. The load identification algorithm is the core of the system. Non-intrusive power load intelligent identification systems are generally based on a single model for aggregation, and commonly used models include hidden Markov models, graphical models and deep learning models. Usually, a single model is only effective in identifying certain types of electrical loads, and its universality and generality need to be improved. Improving the decomposition accuracy of the model is the main purpose of implementing non-intrusive powerload monitoring. The stacking integration algorithm is used to optimize the non-intrusive powerload monitoring model. The stacking integration model is divided into two layers. The base model initially decomposes the load and realizes the load accuracy through the meta-model. Experiment results show that the non-intrusive load identification algorithm based on Stacking modeling proposed in this paper can accurately decompose the minute-level sampled data. Compared to a single model, the decomposition accuracy of this method is improved by 8.2%.

Системний аналіз та модель ідентифікації хвороби на основі медичних зображень

In given article, we investigate medical images and develop an intelligent system for identification of the disease on their basis. The paper proposes an approach to finding the affected tissue areas in medical images. To find them, a mask was extracted for training a neural network. Mask extraction was carried out using annotations, where polygons with affected tissues were identified. The studied objects were assigned to different classifications of morbidity.

A Deep Learning Method for Facies Recognition from Core Images and Its Application: A Case Study of Mackay River Oil Sands Reservoir

There is a large amount of drilling core data in the Mackay River oil sands block in Canada, and the accurate identification of facies from the cores is important and necessary for the understanding of the subsurface reservoir. The traditional recognition method of facies from cores is by human work and is very time consuming. Furthermore, the results are different according to different geologists because of the subjective judgment criterion. An efficient and objective method is important to solve the above problem. In this paper, the deep learning image-recognition algorithm is used to automatically and intelligently recognize the facies type from the core image. Through a series of high-reliability preprocessing operations, such as cropping, segmentation, rotation transformation, and noise removal of the original core image, that have been manually identified, the key feature information in the images is extracted based on the ResNet50 convolutional neural network. On the dataset of about 200 core images from 13 facies, an intelligent identification system of facies from core images is constructed, which realizes automatic facies identification from core images. Comparing this method with traditional convolutional neural networks and support vector machines (SVM), the results show that the recognition accuracy of this model is as high as 91.12%, which is higher than the other two models. It is also shown that for a relatively special dataset, such as core images, it is necessary to rely on their global features in order to classify them, and, with a large similarity between some of the categories, it is extremely difficult to classify them. The selection of a suitable neural network model can have a great impact on the accuracy of recognition results. Then, the recognized facies are input as hard data to construct the three-dimensional facies model, which reveals the complex heterogeneity and distribution of the subsurface reservoir for further exploration and development.

Intelligent collision risk detection in medium-sized cities of developing countries, using naturalistic driving: A review

Traffic accidents are one of the most serious problems worldwide, being one of the leading causes of death and economic loss in the world. Low-and middle-income countries, mainly their medium-sized cities, are among the most affected by this problem. 93% of traffic accidents occur in low and middle-income countries, even though these countries have approximately 60% of the world's vehicles. This occurs mainly because in these types of countries, especially in medium-sized cities (target context), there are no ideal conditions for driving, such as adequate road infrastructure, good condition of vehicles, and rigorous safety policies. Advanced data analysis techniques including machine learning (ML) have increasingly been used to solve this problem. Naturalistic driving (ND) can be applied as a data collection method that provides information on traffic accidents. ND commonly uses a vehicle's kinematic data to detect high-risk driving behaviors that could cause an accident. The objectives of this document are to present a review of different alternatives that help in data collection and creation of intelligent solutions related to detection of possible traffic accidents, principally using ND; and to propose an intelligent collision risk detection system (ICRDS) for identification of areas with a high probability of TA in the target context. Through the review, it was possible to analyze and evaluate the devices, variables and algorithms that help characterize a risk event in driving, considering the target context. The development of a prototype of an ICRDS for a medium-sized city in a developing country is considered viable, considering the identified components, with the aim of identifying risk events in driving, and areas of high probability of accidents in the city. • A systematic review of analyses and detection of near-crashes and vehicle data collection. • A comparative list of the best devices for naturalistic driving studies. • A comparative list of Machine Learning algorithms for predicting, classifying or analyzing driving events. • Provide a guidance for future research in naturalistic driving study approaches.

Research on Lightweight Lithology Intelligent Recognition System Incorporating Attention Mechanism

How to achieve high-precision detection and real-time deployment of the lithology intelligent identification system has significant engineering implications in the geotechnical, geological, water conservation, and mining disciplines. In this study, a lightweight lithology intelligent identification model is proposed to overcome this problem. The MobileNetV2 model is utilized as the basic backbone network to decrease network operation parameters. Furthermore, channel attention and spatial attention methods are incorporated into the model to improve the network’s extraction of complicated and abstract petrographic elements. In addition, based on the findings of network training, computing power performance, test results, and Grad-CAM interpretability analysis and comparison tests with Resnet101, InceptionV3, and MobileNetV2 models. The training accuracy of the proposed model is 98.59 percent, the training duration is 76 min, and the trained model is just 6.38 megabytes in size. The precision (P), recall (R), and harmonic mean (FI-score) were, respectively, 89.62%, 91.38%, and 90.42%. Compared to the three competing models, the model presented in this work strikes a better balance between lithology recognition accuracy and speed, and it gives greater consideration to the rock feature area. Wider and more uniform, strong anti-interference capability, improved robustness and generalization performance of the model, which can be deployed in real-time on the client or edge devices and has some promotion value.