Search

BACKGROUND: The Information Technology and Information Technology Enabled Services (IT and ITES) industry has been the backbone of the Indian economy. The sector is characterized by long working hours, strict deadlines, night shift, constant usage of computers, etc. Hence, the industry and nature of the job are influencing the psychological risk factors of the employees. PURPOSE: The purpose of this paper is to explore the psychological risk factors (job stressors) of technocrat millennials and critically analyze them from the socio-demographic perspective. METHODS: A single cross-sectional study with snowball sampling was collected from 300 technocrat millennials in Ahmedabad city (India). Job stressors are examined as psychological risk factors. MANOVA and independent-sample t-test has been put to use for establishing the relationship between psychological risk factors and socio-demographic variables. RESULTS: The study highlighted that the experience, current position, size of family, number of children, gender, and family type had an impact on psychological risk factors of technocrat millennials. CONCLUSIONS: The study contributes to the literature on the psychological risk factors and its association with demographic variables, and specifically in Ahmedabad city (India). Income did not have an impact on psychological risk factor; whereas education was the only demographic variable affecting the responsibility for persons. The experience influenced the role overload, role ambiguity, poor peer relations, and intrinsic impoverishment. The study explained each socio-demographic variable’s impact on 12 psychological risk factors separately. The result of the paper will give insight to HR managers in the recruitment and selection of employees.

Human activity recognition is a key technology in intelligent video surveillance and an important research direction in the field of computer vision. However, the complexity of human interaction features and the differences in motion characteristics at different time periods have always existed. In this paper, a human interaction recognition algorithm based on parallel multi-feature fusion network is proposed. First of all, in view of the different amount of information provided by the different time periods of action, an improved time-phased video down sampling method based on Gaussian model is proposed. Second, the Inception module uses different scale convolution kernels for feature extraction. It can improve network performance and reduce the amount of network parameters at the same time. The ResNet module mitigates degradation problem due to increased depth of neural networks and achieves higher classification accuracy. The amount of information provided in the motion video in different stages of motion time is also different. Therefore, we combine the advantages of the Inception network and ResNet to extract feature information, and then we integrate the extracted features. After the extracted features are merged, the training is continued to realize parallel connection of the multi-feature neural network. In this paper, experiments are carried out on the UT dataset. Compared with the traditional activity recognition algorithm, this method can accomplish the recognition tasks of six kinds of interactive actions in a better way, and its accuracy rate reaches 88.9%.

In modern day Psychiatric analysis, Epileptic Seizures are considered as one of the most dreadful disorders of the human brain that drastically affects the neurological activity of the brain for a short duration of time. Thus, seizure detection before its actual occurrence is quintessential to ensure that the right kind of preventive treatment is given to the patient. The predictive analysis is carried out in the preictal state of the Epileptic Seizure that corresponds to the state that commences a couple of minutes before the onset of the seizure. In this paper, the average value of prediction time is restricted to 23.4 minutes for a total of 23 subjects. This paper intends to compare the accuracy of three different predictive models, namely – Logistic Regression, Decision Trees and XGBoost Classifier based on the study of Electroencephalogram (EEG) signals and determine which model has the highest rate of detection of Epileptic Seizure.

E-Rickshaw is an E-vehicle that has three wheels, a rechargeable battery driven electric motor as engine. E-rickshaw has become very popular due to low operating cost, low maintenance cost, eco-friendliness and ease of driving. It is perfect for small distance transport. As a last mile connector, it has transformed the public transport system in India. The low cost electric vehicle carries enough people to make a decent income and hence has become a source of livelihood for many. For considering the issues in this paper, detailed attributes of E-rickshaw are studied and Analytical Hierarchy Process (AHP) has been applied to calculate criteria weights for the sorted attributes. Subsequently, Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), a Multi Criteria Decision Making (MCDM) technique has been applied for the selection of best E-Rickshaw. In this paper, sensitivity analysis and comparative analysis have been conducted for further insight.

It is of great significance to recognize the metallurgical entity relations in order to construct the Knowledge graph of Metallurgical Literature and to further understand the metallurgical literature. However, there are few researches on the textual entity relations in metallurgical fields either few marked Corpora. The syntactic structure of the same entity relationship category is relatively simple and has strong domain characteristics. The traditional entity relationship model can not identify the domain entity relationship well. Meanwhile the syntactic structure of the same entity relations class is relatively simple, and the syntactic structure is relatively simple in the recognition of entity relations in metallurgy field. Furthermore, the entities with similar syntactic structure often have the same entity relations and the different words in the sentence have different contribution to the entity relations. In order to solve the mentioned problems, this paper will combine the algorithm that can highlight the syntactic structure in sentences and improve the accuracy of the model with the Algorithm that can highlight the contribution of words in sentences and the loss function level integration is carried out in the framework of small sample prototype network, so as to maximize the advantages of each algorithm and improve the accuracy –firstly, in the coding layer of the prototype network, we use the CNN algorithm which can highlight the important words in the sentences and the TreeLSTM algorithm which can parse the sentences in the text so that the syntactic relations between the words in the sentences can be acted on in the relation recognition, the sentences are coded together by two algorithms, then, the EUCLIDEAN distance loss is calculated by using this high quality coding and the prototype coding, finally, the traditional entity relation recognition model with Attention Mechanism is integrated into the loss function, further highlighting the decisive role of important words in text sentences in relation recognition and improving the generalization of the model. The results showed that compared with the traditional methods such as CNN, RNN, PCNN and Bi-LSTM, the proposed method in this paper has better performance in the case of small sample data set.

The purpose of this paper is to evolve a novel area-based Pythagorean fuzzy decision model via an approach-oriented measure and an avoidance-oriented measure in support of multiple criteria decision analysis involving intricate uncertainty of Pythagorean fuzziness. Pythagorean membership grades embedded in a Pythagorean fuzzy set is featured by tensible functions of membership, non-membership, indeterminacy, strength, and direction, which delivers flexibility and adaptability in manipulating higher-order uncertainties. However, a well-defined ordered structure is never popular in real-life issues, seldom seen in Pythagorean fuzzy circumstances. Consider that point operators can make a systematic allocation of the indeterminacy composition contained in Pythagorean fuzzy information. This paper exploits the codomains of the point operations (i.e., the quantities that express the extents of point operators) to launch new measurements of approach orientation and avoidance orientation for performance ratings. This paper employs such measurements to develop an area-based performance index and an area-based comprehensive index for conducting a decision analysis. The applications and comparative analyses of the advanced area-based approach to some decision-making problems concerning sustainable recycling partner selection, company investment decisions, stock investment decisions, and working capital financing decisions give support to methodological advantages and practical effectiveness.

In order to improve the security of the Web-based ideological and political education resource information system, this paper analyzes the current privacy protection research and the privacy protection mechanism of Web services, and constructs a service framework of the ideological and political education resource information system based on the AMP module. Moreover, this paper explains the design and implementation of the overall framework, and then focuses on the design and implementation of AMP based on Agent, connection pool and sleep pool. In addition, this paper calculates the basic parameters related to the model, and describes the configuration and function of the parameters in detail. Finally, this paper applies AMP to the practice of the Web-based ideological and political education resource information system, and analyzes the system performance through experimental research. The results show that the system constructed in this paper has achieved a relatively perfect effect.

In order to better promote the development of circular economy, this paper introduces the current international eco-efficiency indicator system used to measure the development of circular economy, and points out that the eco-efficiency indicator system will be an important method for analyzing environmental efficiency in social economic activities. Moreover, this paper combines the actual needs to construct an ecological economy development efficiency evaluation system based on the intelligent DAE model. In addition, this paper combines the specific situation of regional circular economy development, further points out the main indicators that the eco-efficiency indicator system should include, and introduces the experience of establishing circular economy indicators in developed countries. Finally, this paper uses the eco-efficiency index system to evaluate the development of Qingdao’s circular economy, and design experiments to evaluate the efficiency of the system constructed in this paper. The research results show that the evaluation system of the intelligent ecological economy development efficiency constructed in this paper has a certain effect and has certain significance for the development of ecological economy.

In order to improve the effect of traditional cultural innovation, this paper proposes a cultural algorithm with dual knowledge, and improves the effect of the algorithm to obtain a cultural algorithm with dual knowledge. Each individual corresponds to its unique dual knowledge, so that the individual’s evolution can move towards the current optimal solution. This paper constructs a traditional cultural innovation system architecture based on artificial intelligence, analyzes its functional modules, and constructs the system structure from the perspectives of cultural classification and cultural innovation. After constructing the system, this paper designs experiments to verify the system performance. The research results show that the system constructed in this paper performs well in traditional cultural analysis and traditional cultural innovation, and can provide references for related research.

In order to solve the security problems of the ideological and political teaching system itself, the ideological and political teaching information management technology needs to be further improved. According to actual needs, based on artificial intelligence and data security models, this paper studies and implements a network security information management system based on artificial intelligence and security models. The system can effectively manage the contents of safe work and increase the ability of information sharing and collaborative work. According to the actual needs of most current systems, with data mining, data recognition, and security management as the goals, this paper builds the structure of the functional modules and adopts the function cascade to finally realize the safety information management of this system. In addition, this article designs experiments to verify the performance of the model constructed in this article. The research results show that the model has good performance and meets actual needs.

In order to improve the acquisition and recommendation effect of English education resources, this paper proposes a Rank algorithm of web English educational resources based on fuzzy sets and RSS, and deeply studies the basic principles of the algorithm and introduces several keyword extraction techniques. The user’s browsing behavior and user interest acquisition methods are classified. Researchers can plan to further explore the page ranking algorithm to improve the performance of the scheme based on the damping factor. In addition, this paper uses Web technology to acquire English education resources and build a recommendation model, and uses crawler technology to build an overall system model. Finally, this paper designs experiments to verify the performance of the algorithm model constructed in this paper, and analyses the experimental results by mathematical statistics. The research results show that the algorithm model proposed in this paper has significant effects and is of great significance to the acquisition and recommendation of English education resources.

In order to improve the economic history education effect of Shanxi merchants, this paper combines fuzzy control and quantum evolution algorithm to construct Shanxi merchant economic history education system. The purpose of this paper to construct the Shanxi merchant economic history education system is to establish a learning platform on the Internet or local area network that allows students to learn outside the classroom. This system will consist of multiple sub-modules, and it will provide knowledge points and networks, problem sets, student assignments, teacher-student interaction links, and teaching resource management for Shanxi merchant economic history teaching. Moreover, this system will be designed as an open network-assisted teaching system. In addition, this paper designs experiments to verify the performance of the algorithm constructed in this paper. The research shows that the Shanxi merchant economic history education system based on fuzzy control and quantum evolution algorithm constructed in this paper performs well in data mining and also has good performance in practical education.

In order to improve the effect of literary works education, this paper combines intelligent machine learning and reader scoring criteria factors to construct an intelligent education model, and proposes a collaborative filtering recommendation algorithm based on item proportion factors and time decay. When calculating the user similarity, this paper adds the scale factor of the intersection of common scoring items to all the scoring items, and considers the non-intersection part of the user scoring items. Secondly, when predicting the project score, this paper adds a time decay function, combines the forgetting curve law to modify the score prediction method, and combines the actual needs to construct the basic framework of the education model. In addition, this paper designs experiments to verify the performance of the literary work education model constructed in this paper. The research results show that the literary work education model constructed in this paper based on intelligent machine learning and reader rating criteria factors has a certain role in promoting the effect of literary education.

Addressing environmental issues in policy making requires recognising these issues as part of a complex socio-ecological system. The evidence base for such policies and associated monitoring and implementation measures, as well as related official indicators, statistics and environmental accounts are receiving increasing attention. This paper explores the potential of citizen science as a non-traditional source of data to complement the current data production process for evidence-based policy-making, using pollution of surface waters and its effect on associated ecosystems as an example. The paper develops a framework that helps to explore the official data production process in relation to different purposes of environmental policies. This highlights different challenges that the current official data production process sees itself confronted with in relation to the different purposes of the policies and associated monitoring regimes. These questions are explored with reference to the case of evidence-based policy making on water quality of surface freshwater in the EU, with a focus on Luxembourg. The analysis is based on extensive documentary analysis and literature review, as well as a series of interviews and participatory workshops with various stakeholders, and first results of a pilot project work with engaged citizen volunteers to solicit data on water quality with a focus on its nutrient content. On this basis, this paper argues that citizen science has the clear potential to meaningfully contribute both to the evidence base for policy and practice, as well as to an improved governance process.

The development trend of foreign education is affected by many factors, so its future development trend is difficult to judge. Therefore, it is necessary to simulate and analyze the development trend of foreign education through artificial intelligence. According to actual needs, based on artificial intelligence algorithms, this paper builds artificial intelligence simulation analysis model to realize the simulation analysis of foreign education development. Moreover, starting from the overall design architecture of the online education platform, this paper builds functional modules, uses the machine learning constructed in this paper for data training and data prediction, and outputs prediction results. In order to study the performance reliability of the model, we predict and judge the development trend of foreign education and determine the model reliability through empirical judgment. The research results show that the model constructed in this paper has a certain effect.

In the context of information education, English teaching needs to match the development of artificial intelligence to improve the intelligence of English teaching. Based on the artificial intelligence matching model, this paper constructs an English teaching reform model based on artificial intelligence algorithms. Moreover, based on the FISST multi-target tracking method, this paper firstly models the target state and measurement as RFS, and then uses the Bayesian filtering method to recursively calculate the target posterior PDF, which can estimate the number and state of targets in real time and make up for the shortcomings of traditional tracking methods. In addition, the system proposed in this article can be applied to online English teaching. Through this system, teachers can realize one-to-one matching of students, identify the status of students in time, and give corresponding English teaching methods to different students. Finally, this paper designs a controlled experiment to analyze the performance of the algorithm proposed in this paper. The research results show that the model constructed in this paper has certain practical effects.

At present, with the continuous improvement and rapid development of the socialist market economy in China, especially the sudden rise of electronic information technology, the new communication tools are changing with each passing day. In this process, new media information technology, with its unique advantages, “broke” into the vision of modern people, participated in and influenced people’s way of life, behavior habits and social operation, and brought the development of human society into a new stage, the new media age. In today’s era, the new media has been developing rapidly in terms of technology and content, and its content is timeliness and interesting. It matches the characteristics of contemporary college students in pursuit of freshness and individuality, so that it has won the general favor of college students. However, when the new media information technology affects and permeates all aspects of the University, the ideological and political education of college students does not seem to keep pace with the times. Because of the lack of experience, many colleges and universities have not integrated the new media information technology with the ideological and political education of college students, but simply turn the original paper into a web page, ignore the readability and dissemination of the content, resulting in the poor practical effect of education, and then a series of real problems urgently to be solved. Therefore, this paper makes a detailed summary of the problems and Countermeasures of the current new media and college students’ Ideological and political teaching, aiming at arousing people’s attention to the political and ideological education of college students in the new media era, and providing suggestions for the use of new media information technology to promote the development of College Students’ Ideological and Political Education under the information age.

In the face of the arrival of artificial intelligence era, English classroom teaching is bound to change in organizational form, activity design, teacher-student relationship and technology integration. This paper combines SPOC and rain classroom, from knowledge infusion to ability improvement, individual teaching to team teaching, static resources to dynamic resources, artificial evaluation to intelligent evaluation, which brings new development opportunities for education. This paper uses the combination of Internet and artificial intelligence to analyze the teaching mixed teaching mode based on SPOC+and WeChat Rain classroom on the Internet. Based on the analysis results, we novelty propose the mixed English teaching mode based on such platforms to reveal the combination of Internet and artificial intelligence. On the second basis, we propose that future English teachers should continue to develop in the integration of teaching resources, integration of educational technology, design of in-depth learning activities, improvement of professional quality and self-cultivation of personality charm, and actively adapt to the era of artificial intelligence for College English classroom teaching.

To improve the effectiveness of digital inclusive finance risk prevention, this paper constructs a digital inclusive finance risk prevention system based on machine learning and neural network algorithms and performs special data pre-processing for convolutional neural networks. Moreover, based on the index data arrangement with the shortest double Euclidean distance, this paper uses the principle of combining qualitative analysis and quantitative analysis to process the data set. Aiming at the characteristics of different links in the whole process of digital inclusive finance, this paper has formed a preliminary digital inclusive finance risk factor assessment of influencing factors. Besides, this paper combines the needs of digital inclusive finance risk methods to construct a digital inclusive finance risk prevention model, and design functional modules based on process analysis. Finally, this paper designs experiments to verify the performance of the digital inclusive finance risk prevention model constructed in this paper. The experimental research results show that the model constructed in this paper has a certain effect.

In general, Robotics is the area concerned with the linking of perception to action, and AI must have a central role in Robotics if the association is to be intelligent. Skiing and Snowboarding are famous winter games worldwide, enjoyed by participants of all ages and skill levels. Leg dominance has been recounted as a probable risk factor in downhill skiers for lower-limb injuries. Furthermore, snowboarders are more likely to injure their ankles than alpine skiers. To overcome these issues, in this paper, the Artificial Intelligence assisted Statistical model (AIASM) has been proposed to the smart robotic supporting leg for skiers and snowboarders. This paper introduces the concept and study of a robotic modular leg (RML) system with a reduced degree of freedom (DOF). The RML gives a perspective on physics that uses dynamic skiing methods and strategies to produce functional ski movements. Kinematic and dynamic models for the leg system are developed and used for modeling tendency, angle, and measurement, unweighting technique to create balanced and realistic curvature turns and peaks. The experimental results show that the suggested system has a performance rate of 95.31% with different ski movements at various intervals, curves, diameters, and peak shapes for tracking the desired footpath.

In recent years, China’s colleges have made gratifying achievements in the funding work of poor students, but there are still some problems. In order to improve the accuracy of the funding work, the performance of the poor students in colleges should be evaluated effectively. This paper uses the design idea based on the whole process, and the fuzzy comprehensive evaluation method and the hierarchical analysis method, and constructs the performance evaluation index system of the poor students in colleges. Then, taking the performance evaluation of poor students’ support in Jiangxi University of Technology as an example, according to China’s national conditions, the empirical analysis shows that the poverty students’ support work in Jiangxi University of Technology is at the general level, and can be improved from four aspects: perfecting the mechanism of identifying poor students, broadening the funding channels, perfecting the supervision mechanism of financial aid for poor students, and combining financial aid with mental support. The research of this paper is of great significance to improve the management level of the funding of poor students in colleges and universities.

Activity recognition and classification are emerging fields of research that enable many human-centric applications in the sports domain. One of the most critical and challenged aspects of coaching is improving the performance of athletes. Hence, in this paper, the Adaptive Evolutionary Neuro-Fuzzy Inference System (AENFIS) has been proposed for sports person activity classification based on the biomedical signal, trial accelerator data and video surveillance. This paper obtains movement data and heart rate from the developed sensor module. This small sensor is patched onto the user’s chest to get physiological information. Based on the time and frequency domain features, this paper defines the fuzzy sets and assess the natural grouping of data via expectation-maximization of the probabilities. Sensor data feature selection and classification algorithms are applied, and a majority voting is utilized to choose the most representative features. The experimental results show that the proposed AENFIS model enhances accuracy ratio of 98.9%, prediction ratio of 98.5%, the precision ratio of 95.4, recall ratio of 96.7%, the performance ratio of 97.8%, an efficiency ratio of 98.1% and reduces the error rate of 10.2%, execution time 8.9% compared to other existing models.

The increasing maturity of Internet information technology has led to the rapid rise of block-chain technology. In essence, the block-chain is a shared database. In recent years, block-chain technology has attracted the attention of the public and society. This paper mainly conducts systematic research and analysis on the integration and optimization of big data in sports training by block-chain, and achieves large-scale consensus in the case of decentralization, therefore, the information of the network will not be tampered with and forged, thus crossing the boundary of the bottom layer of the information system. This paper is based on the integration and optimization of sports training is big data. With the help of block-chain technology, each module is designed and implemented according to the needs. At the same time, the pow algorithm inside the block-chain is improved. The research results show that it is based on the block-chain. The integration and optimization of technical sports training big data can reduce the number of sports training injuries of sports school students by 10% -20%, which is very conducive to the healthy growth and development of sports school students.

Stress is indeed a life aspect that influences everyone, even though athletes seem to suffer from it one step ahead of others because of the extent they are expected to balance between coursework, workouts, and competitions, along with everyday life and family stress. Therefore, an efficient psychological health analysis for sportspersons is crucial in sports training. This paper introduces a Fuzzy-assisted Neural Network model for Psychological Health Analysis (FNN-PHA) to assess mental stress by monitoring the Electro Cardio Gram signal (ECG), Electroencephalogram (EEG), and Pulse rate. This paper integrates the fuzzy assisted Petri nets, fuzzy assisted k-complex detector, and fuzzy assisted transient time analyzer to ensure the psychological health analysis neural network model’s adaptive performance. The strength of the proposed fuzzy model demonstrates interpretability against the accuracy of different criteria. The simulation analysis shows that the FNN-PHA model enhances the prediction ratio of 98.7%, emotional stability of 96.7%, personal growth of 95.7%, physical fitness level of 97.8%, and depression ratio of 12.5% when compared to other existing models.

With increasing demand from the research community for more frequent and unrestricted access to data, national statistical offices (NSOs) are adopting the 5 safes framework to manage the disclosure risk for releasing such data. In this paper, under some mild conditions, we show that the probability of disclosure, given the controls in the 5 safes, is not greater than the product of the smallest conditional disclosure probability amongst the 5 controls and the Risk Ratios of the remaining four safe controls. By computing the disclosure probabilities of all possible configurations of the controls in each of the 5 dimensions of the framework, one can select the set which has the least control on data, but which also meet the confidentiality and privacy requirements of the NSO. Where the required assumption of unconditional independence of the safes cannot be met, the paper proposes a merger of some of the controls to overcome the violation.

This paper is an attempt to contribute to the discussion on multinational enterprises shifting certain economic activities to special purpose entities abroad for reasons of tax optimisation. The authors argue that a transfer of production to a special purpose entity abroad permitted by tax law is not necessarily a transfer of production in an economic sense. Special purpose vehicles can be involved in production processes from a legal point of view without producing any goods or services in reality – a phenomenon the authors call ‘virtual production’. Thus, simply mirroring the legal transactions in national accounts may result in a distorted representation of economic reality. Unfortunately, the System of National Accounts in its current version as well as other existing guidelines lack clear guidance in order to distinguish virtual from real economic activity in the context of special purpose vehicles. This paper offers a proposal for improvement of existing definitions and concepts.

Recommendation algorithm is not only widely used in entertainment media, but also plays an important role in national strategy, such as the recommendation algorithm of byte beating company. This paper studies the personalized recommendation algorithm based on representation learning. The data in social network is complex, and the data mainly exists in various platforms. This paper introduces AI (Artificial Intelligence) algorithm to guide the algorithm of representation learning, and integrates the algorithm steps of representation learning, to realize the implementation of personalized recommendation algorithm in social network, and compares the representation learning algorithm. Finally, this paper designs a method based on heat conduction and text mining to provide users with webpage recommendations and help users better mine interesting popular webpages. Research shows that the performance of IMF is better than that of PMF because it overcomes the sparsity of data by pre-filling. The accuracy of IMF is 3.69% higher than that of PMF on the epinions data set, and 6.24% higher than that of PMF on the double data set. Rtcf, socialmf, tcars, CSIT, isrec, and hesmf have better performance than PMF and IMF. Among them, rtcf, socialmf, tcars, CSIT, isrec, and hesmf improve the MAE performance of PMF by 7.6%, 6.3%, 8.8%, 7.9%, 9.5% and 14.2%, respectively.

This paper presents an experimental study on the non-dispersive infrared (NDIR) detection technology and dark channel dehazing technology. Based on the analysis of Beer-Lambert Law and differential carbon dioxide detection principle, this paper proposes an atmospheric light value estimation algorithm based on NDIR detection technology. First, the change characteristics of the gas concentration in indoor smoky environment are collected and analyzed. Then appropriate weighting coefficients are chosen based on the gas characteristics to estimate the atmospheric light value. Finally, the digital image dehazing technology through dark channel prior is used for calculation to obtain a haze-free image with high quality and high resolution. The experiment in this paper proves the feasibility of combining NDIR detection technology with dehazing technology, and its ability to improve image quality and achieve better restoration effect.

The important strategic position of military UAVs and the wide application of civil UAVs in many fields, they all mark the arrival of the era of unmanned aerial vehicles. At present, in the field of image research, recognition and real-time tracking of specific objects in images has been a technology that many scholars continue to study in depth and need to be further tackled. Image recognition and real-time tracking technology has been widely used in UAV aerial photography. Through the analysis of convolution neural network algorithm and the comparison of image recognition technology, the convolution neural network algorithm is improved to improve the image recognition effect. In this paper, a target detection technique based on improved Faster R-CNN is proposed. The algorithm model is implemented and the classification accuracy is improved through Faster R-CNN network optimization. Aiming at the problem of small target error detection and scale difference in aerial data sets, this paper designs the network structure of RPN and the optimization scheme of related algorithms. The structure of Faster R-CNN is adjusted by improving the embedding of CNN and OHEM algorithm, the accuracy of small target and multitarget detection is improved as a whole. The experimental results show that: compared with LENET-5, the recognition accuracy of the proposed algorithm is significantly improved. And with the increase of the number of samples, the accuracy of this algorithm is 98.9%.

For this reason, the key to solve the problem of rural financial development is to solve the problem of rural financial development. At present, the state has given important instructions to improve the rural financial ecological environment, but the relevant research on the evaluation of rural financial ecological environment in China is still insufficient. In view of this situation, this paper puts forward a BP neural network model for the comprehensive evaluation of rural financial ecological environment. First of all, this paper studies the relevant basic theory of financial ecology and ecological environment comprehensive evaluation. Through the research, this paper believes that the construction of rural financial ecological environment involves many factors, and each factor has a mutual influence. It is difficult to determine the influence of a single factor on the final result. Therefore, in view of this complex situation, this paper establishes a set of multi factor evaluation index systems including economy, policy, law, culture, etc. And these complex factors are trained by BP neural network. The training results were normalized to quantify the specific impact of each index on the rural financial ecological environment. Finally, in order to verify the actual evaluation effect of this model, a number of comparative experiments including validity verification, stability analysis, comparison and verification of different model error rates are carried out. Through the analysis of experimental data, we can see that the BP neural network evaluation model in this paper has good comprehensive performance and significantly improves the calculation accuracy compared with the traditional analytic hierarchy process.

At present, the relevant test data and training indicators of athletes during rehabilitation training lack screening and analysis, so it is impossible to establish a long-term longitudinal tracking research system and evaluation system. In order to improve the practical effect of sports rehabilitation activities, this paper successively introduces the matrix normal mixed model and the fuzzy clustering algorithm based on the K-L information entropy regularization and the matrix normal mixed model. Moreover, this paper uses the expectation maximization algorithm to estimate the parameters of the model, discusses the framework, key technologies and core services of the development platform, and conducts certain research on the related technologies of the three-tier architecture. At the same time, according to the actual needs of sports rehabilitation training, this paper designs the functions required for exercise detection and prescription formulation. In addition, this paper analyzes and designs the database structure involved in each subsystem. Finally, this paper designs experiments to verify the performance of the model constructed in this paper. The research results show that the performance of the model constructed in this paper meets the expectations of model construction, so it can be applied to practice.

Nowadays, wearable technology and the Internet of Things (IoT) are transforming the healthcare sector by refining the way how devices, applications, and people connect and interact with each other. IoT applications in sports are tremendously useful to monitor health and reduce the risk factor. The battery life of wearable and accurate monitoring has been considered a significant challenge in sports medicine. Hence, in this paper, Energy Efficient IoT based Improved Health Monitoring system (EEIoT-IHMS) has been proposed for accurate and continuous sports person’s health monitoring system. This paper determines the optimal set of clusters based on sensor features, in which power usage has been minimized by duty cycling with optimized prediction accuracy. The experimental results demonstrate that the proposed (EEIoT-IHMS) enhances accuracy ratio, improves battery life, and reduces energy consumption compared to other popular methods.

In order to improve the simulation effect of complex traffic conditions, based on machine learning algorithms, this paper builds a simulation model. Starting from the macroscopic traffic flow LWR theory, this paper introduces the process of establishing the original CTM mathematical model, and combines it with machine learning algorithms to improve it, and establishes the variable cell transmission model VCTM ordinary transmission, split transmission, and combined transmission mathematical expressions. Moreover, this paper establishes a road network simulation model to calibrate related simulation parameters. In addition, this paper combines the actual needs of complex traffic conditions analysis to construct a complex traffic simulation control model based on machine learning, and designs a hybrid microscopic traffic simulation system architecture to simulate all relevant factors of complex road conditions. Finally, this paper designs experiments to verify the performance of the simulation model. The research results show that the simulation control model of complex traffic conditions constructed in this paper has certain practical effects.

Tuberculosis, as a more common infectious disease with serious physical damage to humans, has been relatively vacant in predictive model research. In order to improve the accuracy of pulmonary tuberculosis, this study combined the incidence of tuberculosis, collected data using data collection methods, used a single data model for predictive analysis, and compared with the actual situation. At the same time, through the comparative analysis, the paper draws the shortcomings of the traditional single model algorithm, constructs a combined model for the prediction of tuberculosis, and collects the incidence of tuberculosis. In addition, this paper draws it into a statistical chart, and analyzes its pathological characteristics and the dynamic trend of the onset. Through experimental research, it can be seen that the prediction accuracy of the combined model of this study is high, which can provide theoretical reference for subsequent related research.

The development of hardware technology and information technology has promoted the development of image recognition technology. Today, image recognition technology has been applied to many national defense technologies; especially target image recognition technology is widely used in the field of air threat prevention. However, nowadays, the air target recognition technology has the disadvantage of high misjudgment rate. The main reason is that the sky is too large and the distance gap makes it difficult to distinguish the target image from other noise images. This paper takes the neural network as the classification tool, through image preprocessing and contour extraction, establishes the recognition model of the target image. The simulation results of 10 data sets show that the method used in this paper is more than 85% accurate, but the error rate is only 0.7%. The simulation results show that the model designed in this paper can achieve air target recognition very well.

In the process of the continuous development of e-commerce, the main body of e-commerce activities gradually formed the value chain through a variety of ways, and the e-commerce ecosystem was born. The e-commerce ecosystem model based on value chain is established, and the subject members are divided into core population, key population, assistant population and dependent population, and the hierarchical structure. And related characteristics of the ecosystem are studied. The evolution mechanism of e-commerce ecosystem based on value chain and the role of value chain in the evolution of e-commerce ecosystem are analyzed. This paper summarizes the evolution path of e-commerce ecosystem based on value chain, including the characteristics of each evolution stage and the countermeasures that core e-commerce enterprises should take. In this paper, based on the whole value chain theory of lean management, by implementing the lean logistics improvement of the whole value chain, implementing fine management and control could be over the whole process of inventory formation, highlighting the planning, implementation and assessment of inventory. The results indicates in the whole value chain, through the effective building of internal and external supply chains, the inventory turnover rate continues to rise, which effectively improves the pressure drop level of the company’s “two funds”. And it can promote the sustainable and healthy development of the company.

The network provides a convenient mechanism for publishing and obtaining documents, and has now become a gathering place for all kinds of information. In the network, the amount of information increases exponentially, and how to dig useful patterns or knowledge from the massive network culture has become a hot topic for scholars. In data mining, in order to enable readers to quickly obtain the content of interest, research text classification, and automatically classify text data according to a certain classification model. Internet cultural text data has the characteristics of unstructured, subjective, high-dimensional, etc., which makes it difficult for text mining algorithms to extract effective and easy-to-understand classification rules, and the computational complexity is too high. This paper proposes a feature selection method based on robust features, using sample deviation and variance as the criteria for feature attributes to rank the importance of feature attributes, and select the best feature attribute subset. The experimental results show that the classification accuracy of the feature selection method based on sample deviation and variance proposed in this paper is higher than the traditional word frequency as the feature selection method, which proves the feasibility and superiority of the feature selection method proposed in this paper.

At present, many exciting results have been achieved in the application of deep learning to image recognition. However, there are still many problems to be overcome before deep learning is used in practical applications such as image retrieval, image annotation, and image-text conversion. This paper studies the structure of deep learning, improves the commonly used training algorithms, and proposes two new neural network models for different application scenarios. This paper uses Support Vector Machine (SVM) as the main classifier for Internet of Things image recognition and uses the database of this paper to train SVM and CNN. At the same time, the effectiveness of the two for image recognition is tested, and the trained classifier is used for image recognition. The result surface: In the labeled data set, the rank-1 accuracy of CNN is 85.77%, which is higher than 90.28% of the SVM method. In the detection data, CNN’s rank-1 accuracy rate is 83.11%, which also exceeds SVM’s 80.22%. SVM+CNN has a rank 1 value of 84.69% for the detection data set. This shows that deep learning can map the feature representation of the image and the feature representation of the word to the same space, making the calculation of the similarity and correlation between the image and the text easier and more straightforward.

The ability to perceive students’ emotions in real time is related to whether the description of the student’s state is accurate, and it is also related to whether the goal of achieving the students’ individual learning needs can be achieved. This paper studies the students’ boring emotion in English learning, and builds the recognition system of students’ boring emotion in English learning based on fuzzy neural network. Moreover, this paper combines the actual needs to construct the system function modules, and carries out the algorithm analysis and framework construction, and uses the mathematical modeling method to add the emotional factor to the English learner state modeling. In addition, according to the actual needs of the system constructed in this paper, the boring emotion of English learners is recognized. In addition, this paper designs experiments to verify the performance of the model, and analyze the system reliability from the theoretical perspective and the practical perspective. The experimental research results show that the model constructed in this paper meets the expected goals.

With the advancement of information construction, as an important part of smart city, the construction of smart society is invested. As a new mode of social management innovation under the new situation, the advantage of smart society is that it can complete the full integration of existing public resources and infrastructure, realize the knowledge mining of appearance information, so as to make social services and management more information-based and intelligent based on the common needs of smart social management, this paper analyzes the overall function of social management system in detail, puts forward the architecture and simulation of social management service system based on internet of things information model, and realizes the smart social management system, so as to make social service more high-quality and efficient, and social management more standardized and scientific. This paper describes how to promote social management services, and the specific needs of social information as the research basis of this paper, realize the application of social information integrated management service platform, and then give play to the advantages of information technology, realize the transformation of social management work mode, facilitate the social management work, improve the overall quality of life of the community, and reduce the cost of management.

This paper adopts the decentralized model of e-commerce model through edge computing for in-depth analysis and discussion, and designs and implements the decentralized e-commerce model. Based on the background of the e-commerce environment formed by the development and application of Internet technology, the research starts from the core role of innovation management in business model innovation; based on analyzing the mechanism and path of business model innovation, the research is conducted and theoretical support is provided for business model innovation management, aiming to provide practical and feasible operational solutions for commercial enterprises to adapt to the development of e-commerce and carry out business model innovation. Based on clarifying the idea and composition of the evaluation of business model innovation capability of commercial enterprises based on e-commerce environment, the paper analyzes and constructs the evaluation index system of business model innovation capability of commercial enterprises, and proposes the evaluation method based on gray theory and fuzzy evaluation theory.

In the decision-making system of sports assistant teaching and training, the performance of such system is not robust to different situations with a low accuracy. To solve the problems in the decision-making system, we proposed a decision-making method combining association rules and support vector machine (SVM) in this paper. First of all, we give a computer-aided decision support system for sports assistant learning and teaching training, which is fully elaborated from three aspects: virtual reality (VR) technology, VR based sports assistant learning and teaching and situational cognition, and VR based sports assistant learning and teaching mode. After that, the paper gives the feature extraction of sports auxiliary teaching training through association rules and the decision-making of the extracted association rules by SVM. We have done two different experiments for both association rules mining and SVM on both experiment group and control group of databases. Experimental results have shown that the training characteristics of sports auxiliary teaching very well. In the decision support of association rules, compared with the existing BP neural network, linear discriminant analysis and naive Bayes and other methods, the SVM method has better effect of action recognition in decision support system of sports assistant teaching and training. The robustness is the best for the application of SVM. We provide a new perspective for the decision support of sports auxiliary teaching training by using association rules and SVM. Through the method of this paper, we can obtain better decision-making effect and more robust process of sports auxiliary teaching and training.

The wide application of artificial intelligence technology in various fields has accelerated the pace of people exploring the hidden information behind large amounts of data. People hope to use data mining methods to conduct effective research on higher education management, and decision tree classification algorithm as a data analysis method in data mining technology, high-precision classification accuracy, intuitive decision results, and high generalization ability make it become a more ideal method of higher education management. Aiming at the sensitivity of data processing and decision tree classification to noisy data, this paper proposes corresponding improvements, and proposes a variable precision rough set attribute selection standard based on scale function, which considers both the weighted approximation accuracy and attribute value of the attribute. The number improves the anti-interference ability of noise data, reduces the bias in attribute selection, and improves the classification accuracy. At the same time, the suppression factor threshold, support and confidence are introduced in the tree pre-pruning process, which simplifies the tree structure. The comparative experiments on standard data sets show that the improved algorithm proposed in this paper is better than other decision tree algorithms and can effectively realize the differentiated classification of higher education management.

With the development of space technology in recent years, various spacecraft with different sensors have been launched one after another, and there are more and more satellite remote sensing images in different situations. How to obtain better quality images has become the main research direction in the field of image fusion. Image fusion is an important branch and main research object of information fusion. Generally speaking, with the rapid development of the information society, people have higher and higher quality requirements for a variety of images, so There are serious challenges in storing, transmission and signal sampling. Nowadays, with the development of compressed sensing (CS) theory, a new sampling method has been extensively studied by many scholars. Compared with traditional Nyquist sampling theorem, Compression sensing can respond to the original signal with fewer sampling points. This paper is introduced the fusion algorithm of satellite remote sensing image based on compressed sensing. Firstly, the basic theory of compressed perceptual discretization signal is introduced. Secondly, a sparse CS remote sensing image fusion algorithm based on wavelet transform is proposed. Finally, through simulation verification, comparing the widely used IHS fusion and PCA fusion image methods, the method in this paper can get higher correlation coefficient and lower interaction entropy and spectral distortion after fusion. Compared with other methods in this paper, the fusion image can carry more space information and the original image is more similar.

How to correctly use the spatial measurement model to analyze the differences, causes, and influencing factors of regional finance, and to establish corresponding development strategies to keep the regional financial differences in an ideal state is one of the problems that the current regional financial development needs to face. Based on the spatial Dubin model (SEM), spatial Dubin error model (SDEM), generalized spatial Dubin model (GSAM), etc. in the spatial measurement method, this paper compares the actual effects of different models, and combines the machine learning algorithm to improve the model algorithm, and intelligently analyzes the spillover effects of regional financial space. After constructing the model, this paper takes area A as an example to analyze the spillover effect of regional financial space through the model constructed in this paper. In addition, this paper conducts an empirical analysis on the influencing factors of regional financial development and proposes corresponding strategies to provide support for the construction of regional financial centers.

In the field of intelligent robot engineering, whether it is humanoid, bionic or vehicle robots, the driving forms of standing, moving and walking, and the consciousness discrimination of the environment in which they are located have always been the focus and difficulty of research. Based on such problems, Naive Bayes Classifier (NBC), Support Vector Machine(SVM), k-Nearest-Neighbor (KNN), Decision Tree (DT), Random Forest (RF) and eXtreme Gradient Boosting (XGBoost) were introduced to conduct experiments. The six individual classifiers have an obvious effect on a particular type of ground, but the overall performance is poor. Therefore, the paper proposes a “Novel Hybrid Evolutionary Learning” method (NHEL) which combines every single classifier by means of weighted voting and adopts an improved genetic algorithm (GA) to obtain the optimal weight. According to the fitness function and evolution times, this paper designs the adaptively changing crossover and mutation rate and applies the conjugate gradient (CG) to enhance GA. By making full use of the global search capabilities of GA and the fast local search ability of CG, the convergence speed is accelerated and the search precision is upgraded. The experimental results show that the performance of the proposed model is significantly better than individual machine learning and ensemble classifiers.

Aiming at the problem that the traditional OCR processing method ignores the inherent connection between the text detection task and the text recognition task, This paper propose a novel end-to-end text spotting framework. The framework includes three parts: shared convolutional feature network, text detector and text recognizer. By sharing convolutional feature network, the text detection network and the text recognition network can be jointly optimized at the same time. On the one hand, it can reduce the computational burden; on the other hand, it can effectively use the inherent connection between text detection and text recognition. This model add the TCM (Text Context Module) on the basis of Mask RCNN, which can effectively solve the negative sample problem in text detection tasks. This paper propose a text recognition model based on the SAM-BiLSTM (spatial attention mechanism with BiLSTM), which can more effectively extract the semantic information between characters. This model significantly surpasses state-of-the-art methods on a number of text detection and text spotting benchmarks, including ICDAR 2015, Total-Text.

For deep learning’s insufficient learning ability of a small amount of data in the Chinese named entity recognition based on deep learning, this paper proposes a named entity recognition of local adverse drug reactions based on Adversarial Transfer Learning, and constructs a neural network model ASAIBC consisting of Adversarial Transfer Learning, Self-Attention, independently recurrent neural network (IndRNN), Bi-directional long short-term memory (BiLSTM) and conditional random field (CRF). However, of the task of Chinese named entity recognition (NER), there are only few open labeled data sets. Therefore, this article introduces Adversarial Transfer Learning network to fully utilize the boundary of Chinese word segmentation tasks (CWS) and NER tasks for information sharing. Plus, the specific information in the CWS is also filtered. Combing with Self-Attention mechanism and IndRNN, this feature’s expression ability is enhanced, thus allowing the model to concern the important information of different entities from different levels. Along with better capture of the dependence relations of long sentences, the recognition ability of the model is further strengthened. As all the results gained from WeiBoNER and MSRA data sets by ASAIBC model are better than traditional algorithms, this paper conducts an experiment on the data set of Xinjiang local named entity recognition of adverse drug reactions (XJADRNER) based on manual labeling, with the accuracy, precision, recall and F-Score value being 98.97%, 91.01%, 90.21% and 90.57% respectively. These experimental results have shown that ASAIBC model can significantly improve the NER performance of local adverse drug reactions in Xinjiang.

As a new extended model of fuzzy sets, hesitant fuzzy set theory is a useful tool to process uncertain information in decision making problems. The traditional hesitant fuzzy multi-attribute decision making (MADM) can only choose an optimal strategy, which is not suitable for all of the complex scenarios. Typically, in practical application, decision making problems may be more complicated involving three options of acceptance, non-commitment and rejection decisions. Three-way decisions, which divide universe into three disjoint regions by a pair of thresholds, are more efficient to deal with these problems. Therefore, how to utilize three-way decision theory to process hesitant fuzzy information is an essential issue to be studied. In this paper, from the perspective of hesitant fuzzy distance, a hesitant fuzzy three-way decision model is proposed. First, because hesitant fuzzy element (HFE) is a set of several possible membership degrees, it cannot be compared with thresholds directly. Hence, this paper converts it into the comparison between the distance and the thresholds. Then, to calculate thresholds more reasonably, shadowed set theory is introduced to avoid the subjectivity of threshold acquisition. Furthermore, sequential strategy is adopted to solve the multi-attribute decision making problems. Finally, an example of medical diagnosis and simulation experiments are given to prove the accuracy and efficiency of the proposed hesitant fuzzy three-way decision model.

This paper aims to put forward a hesitant fuzzy multi-attribute group decision making (MAGDM) method based on the weighted power aggregation operators in social network. From the point of view of social network analysis, decision makers (DMs) are interconnected in the process of MAGDM. Furthermore, the dimension of the obtained hesitant fuzzy element (HFE) by original power operators will be greater with the increasing number of attributes and alternatives and DMs, which will lead to the problem of “intermediate expression swell". This paper combines the order operation laws with the power operators to redefine two novel hesitant fuzzy power aggregation operators to simplify the involved calculation and explore new operators’ properties. Meanwhile, when two given elements have different number of values, we use the strength of social ties and social influence to develop an algorithm for extending the HFEs objectively. On the other hand, the PageRank algorithm and the deviation method are used to determine DMs’ combined weights. The feasibility of the proposed hesitant fuzzy MAGDM method based on social network is illustrated by the application to the actual issue of decision making and the comparative analysis with the existing method.