Ebook: Advances in Artificial Intelligence, Big Data and Algorithms
Computers and automation have revolutionized the lives of most people in the last two decades, and terminology such as algorithms, big data and artificial intelligence have become part of our everyday discourse.
This book presents the proceedings of CAIBDA 2023, the 3rd International Conference on Artificial Intelligence, Big Data and Algorithms, held from 16 - 18 June 2023 as a hybrid conference in Zhengzhou, China. The conference provided a platform for some 200 participants to discuss the theoretical and computational aspects of research in artificial intelligence, big data and algorithms, reviewing the present status and future perspectives of the field. A total of 362 submissions were received for the conference, of which 148 were accepted following a thorough double-blind peer review. Topics covered at the conference included artificial intelligence tools and applications; intelligent estimation and classification; representation formats for multimedia big data; high-performance computing; and mathematical and computer modeling, among others.
The book provides a comprehensive overview of this fascinating field, exploring future scenarios and highlighting areas where new ideas have emerged over recent years. It will be of interest to all those whose work involves artificial intelligence, big data and algorithms.
The 3rd International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA 2023) was held as a hybrid conference in Zhengzhou, China, from 16-18 June 2023. The purpose of CAIBDA 2023 was to discuss the present status and future perspectives of artificial intelligence, big data and algorithms. The following topics were selected in order to cover most of the scientific program and highlight an area where new ideas have emerged over recent years:
(1) Artificial Intelligence Tools and Applications;
(2) Intelligent Estimation and Classification;
(3) Representation Formats for Multimedia Big Data;
(4) High-performance Computing;
(5) Mathematical and Computer Modeling, etc.
Around 200 participants attended the conference. In addition to the keynote speeches, many presentations were selected from contributed papers, including oral and poster sessions. These proceedings contains those papers presented as keynote and selected talks, together with other contributions.
Conference participants discussed a wide range of issues related to the theoretical and computational aspects of research in artificial intelligence, big data and algorithms. In addition, thanks to the participation of leaders from a number of important projects, the conference provided a comprehensive overview of this fascinating field and of future scenarios. Professor Meng Zheng (Chinese Academy of Sciences, China) delivered a keynote speech, and shared his own research experience and professional views with the report ‘Design and Optimization Methods in Wireless Networks for Industrial Automation’. He first introduced representative fundamental results on wireless networks for process automation, including the topology design for large-scale and low-overhead networking and cognitive networking protocols. He also presented work on wireless networks for factory automation, including the topology design for dense and mobile networking and for on-demand reliable retransmission and resource allocation schemes for heterogeneous coexisting traffic. Finally, he discussed the applications of industrial wireless networks in industrial automation systems.
We would like to take this opportunity to thank our host institutions, sponsors, and the many people whose work made this conference possible, and express our gratitude to the members of the International Technical Program Committee and Organizing Committee for their efforts, which made this conference a success. In particular, we would like to thank the editors and other colleagues at IOS Press, publishers of the series Frontiers in Artificial Intelligence and Applications, for their patience and kind assistance in publishing this volume.
The Committee of CAIBDA 2023
Smart agriculture in China still faces difficulties in the transformation of agricultural achievements by AI technology and the scattering of land. The current problems in smart agriculture are a serious shortage of talent, scattered data, a weak mass base, unfocused land and weak agricultural technology companies. To address the problems in the development of smart agriculture. Based on the earlier research, the hypothesis that AI technology plays a positive role in the development of smart agriculture. Combining case study and phenomenology of qualitative analysis theory, field observation, focus group discussion and in-depth interviews were used to conduct an in-depth qualitative analysis using the digital agriculture base as a typical smart agriculture construction. Artificial intelligence technology has been widely used in agricultural production, especially in the Digital Agriculture Base, and is an important part of the construction of smart agriculture. It has played a catalytic role in the development of the agricultural economy and the improvement of agricultural production efficiency. This paper explores the impact of AI technology on the development of smart agriculture, and the obstacles to implement AI technology into the construction of smart agriculture, with a view to propose strategies to sustainably promote the construction of smart agriculture. The results show that the hypothesis of AI technology has a positive effect on the construction of smart agriculture holds true.
To solve the problem of low accuracy of commodity identification due to frequent update of commodity library in automatic checkout system, an image retrieval-based commodity identification method for automatic checkout system is proposed. The YOLOv5 algorithm is used to detect commodities in the checkout image, generate a set of localization frames and segment the targets. For the problem that YOLOv5 is insensitive to small targets, a detection layer is added to the YOLOv5 network structure, and the loss function and bounding box localization error are optimized to improve the model detection performance. On this basis, the SCDA algorithm is used to match the registered commodity map in the database with the segmented commodity map to complete the commodity detection and recognition. The experimental results show that the algorithm in this paper improves the accuracy rate by 3.75% relative to DPNet and achieves 80.79% mAP on the whole test set.
Modern site management has made significant progress, and many advancing equipment and technologies have been introduced into the site management process. During power transmission and transformation construction, safety accidents often occur due to fatigue construction or negative emotions. Embedded Devices plus AI is an advanced solution that monitors and identifies worker sentiment in real-time. The LERM model optimized in this paper can run well on embedded devices with reliable accuracy. This model can be well applied to surveillance camera equipment, at low cost, fast response, and high recognition accuracy. The application of cameras with embedded LERM models on power transmission and transformation sites can identify staff emotions in real-time and alarm managers. As a result of this application, fatigue construction or negative emotions can be avoided as a safety hazard in power transmission and transformation construction personnel.
Vulnerability database data sources use semi-structured and unstructured data expression, which can provide convenience for researchers in vulnerability analysis and systematic study of vulnerability mechanism and vulnerability content. However, the traditional vulnerability database has some problems, such as weak correlation attribute, redundant data format and low visualization degree, which are difficult to be understood and analysed by machines. This paper proposes a CVE vulnerability intelligent association based on chain reasoning. By analysing the vulnerability description and the relationship between different vulnerability databases, investigating the software and systems affected by each vulnerability, linking relevant knowledge, inferring the hidden relationship of each vulnerability, building the relationship between the entity nodes of the vulnerability knowledge graph, and finally importing the data into the neo4j diagram database to build the vulnerability knowledge graph. By calculating the number of software associated with vulnerabilities, the CWE chain relationship is deduced, and the vulnerability knowledge graph constructed is used to conduct preliminary inference on the vulnerability scanning results, so as to obtain the hidden relationship between vulnerabilities and optimize the vulnerability scanning results.
With the development of human society, the problem of aging population is becoming increasingly serious; and the topic of elderly care and a healthy retirement has attracted more public attention. Many elderly people are seriously affected by emotional distress, which results in deterioration in physical health. As a result, the development of smart wearable devices has created the possibility of realizing the idea of real-time monitoring of the elderly people’s mental health. This paper will start from the smart wearable products developed to recognize negative emotions the elderly, then understand the functions of smart wearable devices such as collection, analysis, processing, and recognition of physiological signals, and discusses how to use human-computer interaction to monitor the negative emotions of the elderly in real time, so that negative emotions can be quickly transformed into positive emotions. The elderly can maintain a good mental state, promote physical health, and stimulate and motivate emotional communication between the young and the elderly.
The most important form of artificial intelligence technology applied in the broadcasting and hosting industry is “artificial intelligence anchor”. In the context of artificial intelligence, today’s artificial intelligence anchors have received great technical improvements in terms of image, language appearance, and humanized demeanor. Artificial intelligence anchors serve the audience as well as traditional hosts. Through the research and analysis of artificial intelligence anchors, summarize the advantages of artificial intelligence anchors, guide the future development direction of traditional anchors, and hand over low-level repetitive work to artificial intelligence. The completion of the anchor allows traditional anchors to have more time and energy to dig deep into the work of high-level broadcasting and hosting. This research can provide practical guidance for AI anchors to be better selected and accepted by the audience, so that traditional anchors and artificial intelligence anchors can find growth points in the development of the times, so that artificial intelligence anchors and traditional anchors can play a role in broadcasting and hosting. A good cooperation model has been formed in the industry.
The photovoltaic industry is a key strategic initiative in achieving carbon neutrality and emission peak and receives national support as a sunrise industry. The solar cell module is the central part of a solar power generation system, and its production quality and cost have a direct impact on the overall quality and cost of the system. The EL quality inspection is crucial for ensuring the quality of PV modules. However, traditional methods of EL quality inspection, such as manual inspection or machine vision inspection, are found to be inefficient, prone to false detections, and expensive in terms of labour costs. Additionally, these methods may lead to secondary damage to PV modules due to human intervention during the inspection process. Therefore, this paper proposes an intelligent inspection method for PV modules based on image processing and deep learning to improve the efficiency and accuracy of EL QC. The method pre-segments module images using EL image data acquisition and pre-classifies module types based on a priori defect types and then performs secondary detection of defective types of PV module defects using Faster RCNN. The proposed method’s effectiveness was verified by the EL images collected from an actual PV module production line. The algorithm model was able to label over 12 common defects with strong reliability and achieve a detection accuracy of over 98%. This greatly improves the efficiency and accuracy of EL detection of PV modules and reduces labour costs while improving the quality of PV module detection.
The rapid progress of artificial intelligence (AI) algorithms has opened up new opportunities for optimizing energy consumption and promoting sustainable practices in intelligent energy systems. Artificial intelligence algorithms can analyze energy usage patterns and user behavior patterns, further providing support for load balancing, demand side management, and power grid stability optimization calculations, and ultimately providing recommendations for energy-saving practices. This article explores the application of artificial intelligence algorithms in various stages of energy management and optimization from the above three aspects, discusses the models and implementation steps of mainstream artificial intelligence algorithms in each stage, and provides the challenges of utilizing artificial intelligence algorithms in energy systems in the conclusion.
AI algorithms can analyze historical data, weather patterns, and other relevant factors to predict electricity demand. Accurate load forecasting helps in efficient power generation, resource allocation, and grid stability. They are utilized by energy providers, grid operators, and system planners for short-term load forecasting, medium-term capacity planning, and long-term demand projections. These algorithms support decision-making in power generation, resource allocation, load balancing, demand response, and grid stability management. This article provides a detailed explanation of the principles, steps, applicable conditions, and application scenarios of the most promising AI intelligent algorithms, and provide valuable literature research results.
Rewriting: With the acceleration of China’s urbanization construction, the coverage of the power distribution network is expanding day by day, and the operation and maintenance management work is facing increasing pressure. In order to improve the intelligent level of the operation and maintenance management of power distribution network, a set of intelligent operation and maintenance control system of power distribution network based on big data resources is specially designed. The system integrates a large amount of data resources generated by the substation equipment in the operation state, providing more effective data support for the operation and maintenance management work, so as to improve its intelligence level.
With the rapid advancement of smart city construction, it is imperative for rail transit to adopt innovative approaches and explore new ideas to facilitate the intelligent development of the industry as a whole. As a vital component of intelligent transportation, rail transit plays a crucial role in driving smart cities forward by directing its development towards intelligent solutions. This paper aims to analyze the pivotal technologies that contribute to the intelligent development of the rail transit industry cluster, while also showcasing notable achievements in this domain. By examining the current state of the rail transit industry, this study sheds light on the key areas of focus required to foster intelligent advancements within the sector.
Security policy feasibility assessment is to evaluate the ability of the implemented policy to resist threats, fix system vulnerabilities, etc., whether it is in the user’s acceptable range, and also a measure of security and cost balance. There has been outstanding progress in the current research on automated security policies, and the implementation and application of various security policies are progressing well, but we should pay more attention to security policy effectiveness, which better reflects the risks and the scope and degree of risks accepted by the existing system. However, in many security policy evaluations, there is basically no comprehensive evaluation of existing security policies, and more often the existing vulnerabilities and the degree of risk after being threatened are presented, without quantitative and qualitative feasibility assessment. Based on the existing security policy evaluation methods, a method for assessing the feasibility of automated security protection policies is introduced.
To solve the problem of difficult cultivation of precious potted plants, this paper designed an intelligent potted plant cabinet which can be monitored remotely. Embedded control system was developed based on Keil environment, Wireless communication and real-time monitoring devices are brought together, IP address is created in Alibaba Cloud server, data center is established, cloud middleware is set up, B/S version and mobile App access system are developed, and automatic plant of plant cabinet and intelligent monitoring of potted growth status are realized through temperature and humidity control device, light intensity control device, watering control device and real-time monitoring device. The growth and flowering period of jasmine was taken as the experimental object to verify the plant effect. The intelligent plant cabinet controlled the temperature deviation within ±0.5 °C, the air humidity deviation within ±1.5% RH, the light intensity deviation within ±23 Lx, the soil humidity deviation within ±4.1% RH, and the flowering period was prolonged by 4-5 days. The plant cabinet can intelligently control environmental factors according to the growth habits of jasmine, and the system has fast response and stable operation, which can be extended to intelligent cultivation of various potted plants.
Intelligent voice recognition systems play an essential role in the development of intelligent museums. This paper aims to provide an overview of the fundamental concepts, functions, and implementation methods of intelligent voice recognition technology. Additionally, this paper analyzes the current status of utilizing intelligent voice recognition systems in museums across various regions globally, including North America, Europe, and Asia. Finally, the potential future developments and trends of intelligent voice recognition technology in museums are also discussed.
This article aims to explore the method of utilizing the win32com module to interact with application API for achieving automation operations. Taking the AutoCAD application as an example, it introduces the automation operations of common tasks (such as file opening, property retrieval, etc.) by invoking the API interface. By studying the API documentation of AutoCAD and combining it with the win32com. client. Dispatch function, sample code is written to demonstrate the application of this technique in automation operations.
Artificial intelligence is a technology and method that utilizes computers and algorithms to simulate and implement human intelligence. It can learn and optimize through a large number of data and algorithms to achieve various intelligent tasks and functions, such as natural language processing, image recognition, intelligent decision-making, etc. Artificial intelligence technology has been widely applied in fields such as healthcare, finance, transportation, and manufacturing, becoming an important force driving economic and social development. Artificial intelligence is closely related to the human brain. This article analyzes the relationship between artificial intelligence and the human brain, including their connections and differences, imitation mechanisms, neural networks, and deep learning.
Due to the difficulty in collecting injury cases in the actual life of intelligent service robots, it is impossible to use existing product injury information to construct injury scenarios required for risk assessment. Based on a large number of consumer and expert questionnaires, potential hazards of intelligent service robots have been preliminary identified, mainly including small component water bombs, laser radiation, and kinetic energy hazards. Using virtual reality (VR) experiments to verify and confirm the sources of harm, in order to enhance the scientific determination of product safety injury probability and severity, the weight method is used to assign different weights to the probability and severity of injury occurrence. VR experiments account for 0.4, consumer research accounts for 0.3, and expert research accounts for 0.3. Combined with matrix method, the safety risk level of the product is calculated as serious risk.
Car insurance fraud is a high-risk area in insurance and it accounts for as much as 80% of all insurance fraud, with a significant portion of it stemming from the risk of parts leakage. Current anti-leakage techniques in auto insurance mainly rely on the analysis of individual parts data such as vehicle accident records and parts damage lists, which neglects the consideration of the correlation among parts and leads to difficulty in identifying leaked parts effectively. This study proposes a method based on the co-occurrence relationship and density clustering of auto parts to detect parts leakage. In this research, an undisclosed dataset on car insurance fraud was utilized to conduct experiments, and the detection results of parts leakage were obtained. This method takes into account the correlation among auto parts, and possesses higher accuracy and practicality.
This paper studies scheduling activities with stochastic duration of activities, and constructs a multi-objective optimization model, aiming at the balance between makespan and resource leveling. In this paper, a multi-objective optimization genetic algorithm based on fast non-dominated sorting (NSGA-II), is selected for solving the problem. In order to satisfy precedence constraints, an encoding scheme based on logical relationships, a crossover operator where sets are crossover units and mutation which is performed within the set are designed. Finally, the effectiveness of the proposed NSGA-II algorithm is illustrated by comparing the mutual coverage of the solution set of the improved with the original. The mutual coverage solution of sets refers to the mutual dominance between the Pareto optimal solutions produced by different algorithms.
This article aims to introduce the overall design and specific implementation of the agricultural cultural heritage restoration project based on VR technology. We will expound on the overall design of agricultural culture restoration, including content planning, software selection, and technical routes for system implementation, and provide a detailed introduction to the agricultural element modeling based on Blender software, including the establishment of the basic scene and model elements of farming, the production of scene elements for farming display mode, the production of scene elements for farming interaction mode, and the design and model export of interactive farming tools. Finally, we introduce the implementation of virtual reality scenes, including the construction of virtual reality environments in the Unreal Engine5, the implementation of model import, lighting, reflection, and post-effects, the implementation of roaming and interactive functions, and the publication as an EXE file.
In complex environments where narrow passageways exist, there are problems such as low success rate of path finding in the path planning of mobile robots based on the traditional fast expanding random tree (RRT) algorithm. To address the above problems, an RRT algorithm for autonomous narrow channel finding is proposed, using an entrance finding algorithm to identify the entrance of the channel and a bias strategy to rationalize the sampling point selection to improve the success rate of path planning. In addition, a greedy algorithm is introduced to optimize the initial path and improve the quality of the planned path. Through experiments, it is shown that the proposed algorithm improves 59.6%, 56%, 9.3%, and 40% in four aspects compared with the RRT algorithm with bias in narrow channel environment in terms of iteration time, number of iterations, path length, and path planning success rate, respectively.
In this paper, we address the difficult problems in pipeline leakage diagnosis and investigate the leakage diagnosis algorithm based on vibration signal analysis with pipeline vibration signal as the basis. The SSA-VMD algorithm is used to decompose the pipeline vibration signal and conduct comparative experiments. The results of the comparison experiments show the superiority of the SSA-VMD algorithm. Then the decomposed IMF components are feature extracted to construct single feature vectors and then combined feature vectors. Finally, the support vector machine is optimized using the improved grid search method, and comparative experiments are conducted for the single feature vector and the combined feature vector, respectively. The experimental results show that the support vector machine optimized by the improved grid search method has a higher recognition accuracy.
To solve the constrained clustering problem, this paper improves the K-means and proposes a constrained K-means algorithm (CK-means). CK-means algorithm takes into account both clustering analysis and constraints, and can effectively deal with clustering problems with constraints, such as distribution center location problem with warehouse capacity constraints, vehicle routing problem with capacity constraints, etc. It has higher practical value and a wider range of applications. There are two core innovations of the CK-means algorithm: firstly, incorporating constraints into the K-means. The second is a search strategy based on sample weights. In addition, this paper also applies the CK-means algorithm to the location problem of distribution stations at the end of JD Logistics’ supply chain. The experimental results show that the CK-means can solve the clustering problem with constraints with effect.