最新刊期
卷 21 ， 期 11 ， 2022
-
An Analysis Method for Multi-source Cigarette Audit Data Based on Improved Long-Short Term Memory AI导读
摘要：The audit data in the cigarette sales platform have recorded the data evolution during the whole data life cycle. Analyzing the audit data is favorable for identifying the potentially risky behaviors in the cigarette sales over Internet. The cigarette sales audit data have the obvious spatio-temporal correlation character. There still is lack of an ideal analysis method for the cigarette sales audit data. In this paper， we utilize the complementary ensemble empirical mode decomposition （CEEMD） to decompose the cigarette audit data for enhancing the time sequence character. Moreover， we propose an improved long-short term memory （LSTM） recurrent neural network to improve the feature extraction capability for the cigarette audit data. The experimental results over the Hubei cigarette audit database illustrate that the proposed method improved the recall rate for 50 percent and the precision rate for 20 percent than the existing machine learning methods.关键词：complementary ensemble empirical mode decomposition;long-short term memory;cigarette audit data- 116
- |
- 36
- |
- 0
发布时间：2022-12-27 - 摘要：A prediction model of farmland soil CO2 emission was established based on RBF neural network algorithm to solve the difficult problem of farmland soil CO2 emission prediction. The soil moisture content， temperature， organic carbon， ammonium nitrogen， nitrate nitrogen content as the input signal， the maize growth period of soil CO2 flux for the output signal， based on RBF neural network algorithm of farmland soil CO2 emissions prediction model， and select multiple linear and nonlinear regression model， to evaluate the effectiveness of the proposed prediction model. The results showed that the RBF neural network structure of 5-46-1 could better predict soil CO2 emission. The measured value of CO2 emission flux was 0.903（kg/m2）， and the predicted value was 0.854（kg/m2）. The correlation coefficient （R2=0.975） of RBF neural network prediction model is higher than that of other models （linear and nonlinear regression models）， and the mean square error （RMSE=0.091） and mean absolute error （MAE=0.048） of RBF neural network prediction model are lower than that of other models. The prediction performance of RBF neural network algorithm is significantly better than other prediction models， and its accuracy is good， and it can better predict soil CO2 emission flux.关键词：RBF neural network;prediction model;regression model;CO2 emissions;agricultural informatization
- 44
- |
- 18
- |
- 0
发布时间：2022-12-27 - 摘要：In natural language processing， entity and relation extraction is an indispensable part of knowledge graph construction， question answering system design， semantic analysis and other tasks. Most of the information in the field of TCM is stored in the form of unstructured texts. The extraction of key information in TCM texts plays an important role in mining the experience of famous TCM practitioners. However， traditional Chinese medicine texts often have the problems of imbalanced samples and multiple words and one meaning in entity relationship， such as multiple diagnosis results pointing to the same syndrome. To solve these problems， constructed a relationship extraction model based on SimBERT under the semi-supervised learning framework to extract entity relations of traditional Chinese medicine texts. The similar text generation function of SimBERT is used to enhance the text to solve the problem of unbalanced samples. The similar sentence retrieval function of SimBERT solves the problem of multiple words with one meaning. The experimental results show that the SimBERT model based on semi-supervised learning framework can extract entity relations from TCM texts more accurately on the TCM medical case data set constructed in this paper.关键词：relational extraction;SimBERT;cases of traditional Chinese medicine
- 22
- |
- 8
- |
- 0
发布时间：2022-12-27 - 摘要：Aiming at the problem of incomplete knowledge map caused by sparse entities and limited number of entity pairs in knowledge map， a meta learning knowledge map completion model based on small samples is proposed. The important information is transmitted through the relation element， and the gradient element improves the learning efficiency， so as to quickly obtain the triples. The effectiveness of this method is verified on the link prediction task. The results show that the small sample knowledge map completion algorithm based on meta learning improves 41.4% compared with TransE and 7% compared with FSRL on the data set FB15K； On the data set Wordnet18， MeanRank is 37.9% higher than DistMult， and hits＠10 is 18.8% higher than ComplEx； On the data set NELL-995， meanrank increased by 41.4% compared with TransE， and hits＠10 increased by 18.8% compared with FSRL. The proposed method can not only better learn knowledge， but also significantly improve the prediction efficiency of entities and relationships.关键词：knowledge graph;embedded model;few-shot;link prediction;triad classification
- 39
- |
- 16
- |
- 0
发布时间：2022-12-27 - 摘要：In order to improve the accuracy of subway passenger flow forecast， an optimization model combining seasonal difference autoregressive moving average model （SARIMA） and support vector machine （SVM） is proposed by combining the periodicity and nonlinearity of subway passenger flow on different types of dates. The model adopts SARIMA to carry out linear modeling on the time series data of subway passenger flow， and uses SVM to carry out nonlinear modeling on the residual value of SARIMA model input. The prediction results of SARIMA model， SVM model and SARIMA-SVM model on subway passenger flow on weekdays and weekends are compared respectively. The experimental results show that the prediction accuracy of SARIMA-SVM model is higher than that of single model， and the accuracy is improved by 12.24% compared with the combined model without considering the date type. The SARIMA-SVM combined model considering the date type can capture the subway passenger flow law， meet the prediction requirements of subway passenger flow， and provide decision-making basis for subway operation.关键词：SARIMA;SVM;ARIMA;MAPE;passenger flow prediction;combined model
- 12
- |
- 5
- |
- 0
发布时间：2022-12-27 - 摘要：In order to solve the problem of low resolution and blocked pedestrian targets， an anchor-free multispectral pedestrian detection method based on feature selection is proposed.This method takes the dual channel CenterNet as the basic network architecture， applies the feature selection attention network between the two convolution modules， filters the semantic information and detailed features， finally obtains the information that is more conducive to detecting low resolution and occluding pedestrian targets.The experimental results show that compared with the traditional channel fusion multispectral pedestrian detection method， this method reduces the miss-rate of pedestrian detection by 3%， and reduces the miss-rate of small pedestrian targets and seriously occluded pedestrian targets by 15% and 9% respectively， which has a certain application value.关键词：multispectral pedestrian detection;feature selection;anchor-free;CenterNet network
- 25
- |
- 6
- |
- 0
发布时间：2022-12-27 - 摘要：The transmission of massive data between the cloud center and end devices causes network load and delay data transmission. To handle this problem， propose a sortable data compression and transmission mechanism oriented to efficient edge computing， and uses self-coding network to realize efficient homomorphic data compression. It also provides an effective guarantee for the efficient and accurate data classification on edge network nodes. Additionally， we proposed a data feature reconstruction mechanism to alleviate the loss of classification accuracy which caused by data compression. In detail， the feature reconstruction is based on the coding mechanism to reduce the data scale. We verified the performance of the proposed data compression mechanism. The experiments show that the proposed compression mechanism can reduce the data scale and improves the data processing efficiency.关键词：edge computing;feature reconstruction;self-encoding network;sortable data compression
- 33
- |
- 7
- |
- 0
发布时间：2022-12-27 - 摘要：For 4-R（2-SS） parallel mechanism， to solve the problems of the inter-chain coupling and the model uncertainty with unknown upper bound caused by motor parameter drift with temperature change， load change， external disturbance and so on， a double-gain adaptive sliding mode kinematic control method for the parallel mechanism is proposed. The kinematic model of each branch， which considers the inter-chain couple is established， and a sliding mode control strategy is introduced. To make the system have better tracking performance and effectively reduce the chattering of sliding mode. A double-gain adaptive rule is introduced in the sliding mode control， which can adjust the sliding mode switching gain near the sliding mode quickly and effectively， and can avoid the overestimation of sliding mode switching gain. Then a double-gain adaptive sliding mode kinematic control algorithm without the information of unknown upper bound of the system is designed， which can overcome the inter-chain couple. Lyapunov theory is used to prove the stability of the proposed control method. The effectiveness of the proposed control method is verified by MATLAB simulation experiment and prototype system experiment.关键词：parallel mechanism;inertia matrix;inter-chain couple;sliding mode control;double-gain;adaptive rule
- 17
- |
- 3
- |
- 0
发布时间：2022-12-27 - 摘要：To improve the problems of poor accuracy， low efficiency and poor repeatability of traditional random circle detection algorithm， a new detection algorithm based on curve fitting is proposed. The algorithm is based on a probabilistic algorithm to curve-fit the extracted edges of the image， and combines curvature features to segment and filter the edges for classification， and finally uses the random circle detection algorithm for circle recognition. The experimental results show that compared with the traditional random circle detection algorithm， the random circle detection algorithm based on curve fitting can accurately identify the circle contour in the image and improve the execution efficiency by more than 59.1%. The curve fitting-based random circle detection algorithm not only improves the accuracy of recognizing circle contours in images， but also improves the algorithm execution efficiency and anti-interference.关键词：edge extraction;curve fitting;randomized algorithm;circle detection
- 22
- |
- 7
- |
- 0
发布时间：2022-12-27 - 摘要：In Leibo county， Liangshan Yi autonomous prefecture， Sichuan province， the debris flow susceptibility was evaluated， first of all， considering factors such as the natural geography and climate characteristics of Leibo， by ArcGIS to obtain data and draw the corresponding evaluation factor in Leibo county administrative areas classification diagram， use frequency ratio method analysis of evaluation factors on the sensitivity of the debris flow disaster； Then， use SMOTE algorithm to increase a few samples， make positive sample and negative sample reach equilibrium. Finally， the FNN-SGD model was proposed and applied to the assessment of debris flow susceptibility in Leibo county. According to the results of the model， the distribution map of debris flow susceptibility was drawn and the ROC curve was used as the evaluation index. Compared with the feedforward neural network， logistic regression and random forest model， the results showed that： FNN-SGD model has higher accuracy and stability， and is more suitable for the assessment of debris flow susceptibility in Leibo county.关键词：debris flow;GIS;feedforward neural network;logistic regression;random forest;FNN-SGD
- 179
- |
- 78
- |
- 0
发布时间：2022-12-27 - 摘要：Because ferromagnetic substances have a strong interference to MRI examination， traditional magnetic anomaly detection methods can not cope with complex and changeable environmental background， and even reflect different performances for different characteristic targets. Therefore， it is of great significance to stably and accurately detect ferromagnetic substances on MRI examiners in the field of medical safety. At present， magnetic detection technology can not effectively extract magnetic anomaly signals from the noise of complex magnetic field environment， so EMD algorithm is applied to detect weak magnetic anomaly signals in MRI strong magnetic noise environment. This method not only does not need any prior information of the target signal， but also does not need any assumptions about the background noise. It can also detect the target signal adaptively. Firstly， EMD is used to decompose the detected signal to obtain some eigen modal components， and then the reconstructed signal is filtered to detect the magnetic anomaly signal in the reconstructed signal using an energy detector. It can be seen from the simulation and experimental data that the EMD algorithm can effectively remove high-frequency noise， and the signal to noise ratio of the processed signal is more than 16dB. The CSEMD method can suppress the mode aliasing problem in the EMD method， improve the detection efficiency of ferromagnetic materials， and detect weak magnetic anomaly signals in complex environmental backgrounds， in order to provide a new solution for the detection of ferromagnetic materials.关键词：empirical mode decomposition;weak signal detection;magnetic anomaly detection;energy detector
- 9
- |
- 3
- |
- 0
发布时间：2022-12-27 - 摘要：A hybrid recommendation algorithm based on multi-source data clustering and singular value decomposition is proposed to solve the problems of traditional single recommendation algorithm， such as user cold start， high-dimensional sparse data， algorithm accuracy and scalability. Firstly， the algorithm uses TF-IDF formula to process the user item scoring matrix and item characteristic matrix， and generates the user item preference matrix； Second， combined with the user characteristic matrix and scoring matrix as the algorithm input， the improved k-means clustering algorithm is used to divide user clusters； Third， the bissvd algorithm with time decay function is used to decompose and reduce the dimension of the score matrix corresponding to each user class cluster， and the random gradient descent method is used to re predict the score and fill the score matrix； Finally， the user's prediction score vector is sorted from high to low to generate a recommendation list. The experimental results on movielens dataset show that the accuracy and recall of this algorithm are improved by 5.4% and 6.8% respectively compared with the traditional collaborative filtering recommendation algorithm based on SVD， showing better accuracy and recommendation performance， and improving the cold start problem of users. The proposed method has a certain reference for the current hybrid recommendation algorithm.关键词：hybrid recommendation;clustering;singular value decomposition;multi-source data;collaborative filtering;recommended algorithm
- 13
- |
- 4
- |
- 0
发布时间：2022-12-27 - 摘要：Aiming at the problem of too large data scale of the knowledge graph and too long loading time when querying， a compression algorithm CER （Compression based on Equivalence Relation） is proposed from the compression direction of the knowledge graph. This method first judges the edge label of the knowledge graph Whether they are the same， then determine the attribute value of the node label， compress the edge label and divide the vertex set according to different situations. The compression algorithm reduces the spatial scale of the knowledge graph， thereby improving the query efficiency to a certain extent. Finally， a comparative experiment of four compression algorithms is carried out for three public knowledge graph data sets， and the effect of this algorithm is verified from the compression ratio of edges and nodes， compression time， and query time based on the PLL index algorithm，which proved the proposed algorithm has better compression effect.关键词：knowledge graph;compression algorithm;edge label;node label;PLL index algorithm
- 13
- |
- 4
- |
- 0
发布时间：2022-12-27 - 摘要：Aiming at the impact of external factors in the Buck switching power and the uncertainty in the mathematical model on performance， an integral terminal sliding mode controller with an extreme learning machine observer is designed. The system uses integral terminal sliding surface， The initial state of the movement point of the system is located on the sliding surface， which shortens the time of approaching movement and optimizes the response speed. Use extreme learning machin technology to fit the error term of the system， and act on the Buck switching power through feedforward compensation， so that the system has strong robustness. Build a simulation model in the PSIM simulation software， and compare it with the sliding mode controller to verify that the designed it has a certain anti-interference effect， keeping the Buck switching power with high-speed response and small overshoot under various environments performance.关键词：Buck switching power;sliding mode controller;extreme learning machine;observer
- 70
- |
- 6
- |
- 0
发布时间：2022-12-27 - 摘要：In order to prolong the service life of the battery and improve the safety performance of the battery， a battery management system that integrates signal acquisition， automatic control and remote monitoring functions is designed based on the Raspberry Pi. The system uses Raspberry Pi as the core to build a server for users to access， and uses HTTP （Hyper Text Transfer Protocol） protocol， combined with WiFi， NAT-DDNS （Network Address Translation-Dynamic Domain Name Server） technology to achieve remote monitoring and control of the system. The data acquisition of the battery system is realized through the signal acquisition module. Compared with the traditional single system， the system has multiple communication protocol functions. Through the remote management of the battery， the controllability of the battery system is enhanced to meet the convenience and functional requirements of battery management.关键词：Raspberry Pi;battery management system;server;WiFi;NAT-DDNS
- 29
- |
- 20
- |
- 0
发布时间：2022-12-27 - 摘要：In order to explore how to scientifically， reasonably and effectively improve the energy use efficiency of colleges and Universities under the new situation， it is proposed to build a smart energy management platform for colleges and universities based on 5G private network. The platform is based on the cloud pipe end architecture， combined with 5G technology and the construction method of university smart campus， to carry out intelligent management of university energy. Moreover， with the help of the 5G private network， the platform realizes the reverse query function to collect and manage the collected energy consumption data in an all-round， all-time， high-precision and high-density manner， so as to achieve the ultimate goal of energy conservation and emission reduction. The experimental results show that the platform can accurately locate the problems and causes in the construction of the platform through the reverse query function， and significantly improve the reliability of the system. The proposed method can provide reference for colleges and universities that are about to carry out the construction of intelligent energy management platform.关键词：5G;dual domain fusion;smart energy;smart campus
- 20
- |
- 5
- |
- 0
发布时间：2022-12-27 - 摘要：Large traffic attacks of application layer DDoS are gradually showing a high incidence trend. The existing attack detection methods can only respond according to the detection results when large-scale traffic enters the target host， but the large traffic in terabytes may have downtime before the detection system alerts. To this end， an application layer DDoS attack detection method based on near attack source deployment is proposed. The detection node is deployed forward to the near attack source， and normal traffic information is fused to the target through timing and reinforcement learning to intercept attack traffic while detecting attacks. The experimental results show that the system has good detection rate and interception rate， and can weaken the harm of attack traffic to the target host， so as to propose a new solution to solve the large traffic attack of application layer DDoS.关键词：distributed denial of service;application-layer DDoS attack detection;information fusion;timing mechanism;attack detection
- 18
- |
- 6
- |
- 0
发布时间：2022-12-27 - 摘要：With the regulation of virtual currency "mining" activities， some active mining behaviors such as using graphics cards and hard disk mining have been effectively curbed， but the passive mining behaviors that achieve the purpose of mining through web mining hijacking attacks are difficult to be detected by users because of their easy invasion and stealthy characteristics. In order to automatically judge the passive mining beharior， take a website embedded with a mining program as an example， proposes a feature analysis method based on Chrome DevTools Protocol to analyze and judge the websocket traffic and export CPU performance profiling report respectively. After verification， This method can effectively detect and judge web mining hijacking attacks， it has certain reference value.关键词：virtual currency;cryptojacking attacks;Web mining;characteristic analysis
- 9
- |
- 4
- |
- 0
发布时间：2022-12-27 - 摘要：Target detection methods based on deep learning are more and more widely used. For specific detection problems， public dataset has been difficult to meet the requirements. In order to improve the production efficiency of the custom target detection data set， taking the production of a custom helmet wearing detection data set as an example， based on the Scrapy crawler framework， the OpenCV library and the YOLOv5 target detection algorithm， a method of collecting and semi-automatic labeling of custom target detection data sets is designed. The image datas are obtained through Scrapy crawling and OpenCV acquisition of video frames， and the image deduplication is performed through the histogram method. The public dataset is used to train the YOLOv5s model， and the trained model is used to predict the target frame and category of the collected pictures to generate the corresponding annotation file. Filter out high-quality data through annotated files， and use LableImg software to manually correct incorrect annotations. Through experiments， a total of 21 069 pictures were obtained， and 5 053 pictures were deduplicated and filtered. After generating annotations and manual corrections， 10 967 datas were finally obtained as a dataset. Experimental results show that this method can avoid the inefficiency of manual data collection and labeling， greatly save the time of making a custom target detection dataset， and obtain a dataset with large amount of datas quickly.关键词：Scrapy;target detection;dataset;annotation;OpenCV
- 29
- |
- 4
- |
- 0
发布时间：2022-12-27 - 摘要：Due to the overlap of intensity ranges between different object regions， it is difficult to segment the images in the presence of intensity inhomogeneity. To solve this problem， we propose a new multiphase image segmentation model based on fuzzy membership functions and Retinex theory. The method allows that each pixel can be in several regions， which reflect the uncertainty of the images. Moreover， image segmentation is achieved by minimizing the energy functional. We apply the alternating minimization method and design an effective algorithm to solve the solution of our model. Finally， the numerical experimental results are provided to verify the better performance of the proposed model for segmenting the real images and MR images with intensity inhomogeneity than other test methods， the average values of SA index and Dice index reach 0.950 6 and 0.914 1， respectively. Compared with the related representative algorithms， the Dice value is improved by 0.002 7~0.010 7， and the
- 15
- |
- 5
- |
- 0
发布时间：2022-12-27 - 摘要：Aiming at the problems of incomplete segmentation and missing edge details of DeeplabV3+ network in land classification， a land use classification model based on improved DeeplabV3+ unmanned aerial vehicle （UAV） image was proposed， taking the residential area near Yueceng Reservoir as the study area. Firstly， obtain the UAV image data in the study area and establish the corresponding data set； Secondly， the lightweight network MobilenetV2 is introduced to replace the backbone feature extraction network of DeeplabV3+， which greatly reduces the parameters of the model and improves the calculation speed of the model； Finally， Coordinate Attention （CA） mechanism is added to reduce the loss of detail and improve the segmentation accuracy. The experimental results showed that the improved model not only has better segmentation performance than the original DeeplabV3+ model， but also effectively solves the problems of road disconnection and incomplete segmentation， improves the segmentation accuracy of ground objects， and greatly improves the segmentation efficiency.关键词：DeeplabV3+ model;ground object classification;MobileNetV2;attention mechanism;semantic segmentation network;unmanned aerial vehicle
- 20
- |
- 2
- |
- 0
发布时间：2022-12-27 -
Research on Image Recognition of Agricultural Pests Based on Transfer Learning and GhostNet Model AI导读
摘要：Aiming at the image recognition of crop pests， selected 10 common crop pests as the detection targets. After the samples were screened through the IP102 data set and augmented by data， a crop pest data set Cron Insect containing 3 413 images was constructed. Use the lightweight convolutional neural network GhostNet to build a recognition model， and use the transfer learning method to train the GhostNet model to build a lightweight crop pest detection network. The recognition accuracy of GhostNet network based on transfer learning training is 93.64%， and the recognition accuracy of GhostNet network without transfer learning is 92.33%. The confusion matrix shows that the GhostNet model based on transfer learning training can effectively improve the recognition accuracy of pest images. The experimental results show that the GhostNet model based on transfer learning is more suitable for the identification of small sample crop pests， and the model can be further applied to agricultural pest control.关键词：crop pest recognition;GhostNet;transfer learning;convolutional neural network;deep image recognition- 25
- |
- 6
- |
- 0
发布时间：2022-12-27 - 摘要：The study of automatic detection and identification of crop pests and diseases is essential to promote healthy and sustainable agriculture. To this end， a lightweight recognition model PP-LCNet is designed and applied to the recognition problem of alfalfa pests， and the model structure is adjusted to enhance its ability to extract channel features while ensuring a good recognition correct rate； then the activation function of the model is optimized to make its differentiation boundaries more obvious in the classification problem； finally， a depth-separable convolution is added to effectively reduce the number of model parameters and lower the The final addition of depth-separable convolution effectively reduces the number of model parameters and lowers the computational cost. The performance of PP-LCNet model was compared with MobileNetV2 and EfficientNetB0， and it was found that the recognition accuracy of PP-LCNet increased by 2.97% and 1.74%， the network latency decreased by 42.25% and 91.84%， and the number of parameters decreased by 14.28% and 43.39%， respectively. the performance of PP-LCNet model for The PP-LCNet model has good effect on the recognition and classification of alfalfa pests， and is more suitable for deployment in mobile and embedded devices，and showed wider application range.关键词：convolutional neural network;pest classification;depth separable convolution;SE-layers
- 13
- |
- 2
- |
- 0
发布时间：2022-12-27 - 摘要：Since the normalization of epidemic prevention and control， online teaching has enjoyed a strong momentum of development. The embedded system practice courses are combined with a variety of Internet information teaching platforms and the application of pocket experimental instruments， giving full play to the advantages of the courses， and improving students' engineering literacy， software design and innovation ability with the data storage， course playback and course data analysis functions of the Internet platform. The practice results show that compared with traditional offline teaching， this method has stronger advantages in pre class preview， classroom interaction， review and guidance after class， and is worthy of further research， promotion and use， with a view to providing reference and reference for other colleges to carry out online embedded system courses.关键词：online teaching;pocket instrument;embedded system;Internet platform
- 16
- |
- 9
- |
- 0
发布时间：2022-12-27 - 摘要：In order to speed up the cultivation of innovative talents， through decades of unremitting efforts， a set of computer innovative talents cultivation mechanism has been summarized through exploration and practice. Through the rich teaching design in teaching， scientific research， competition， application and other aspects， the mechanism is committed to broadening students' vision in the professional field， cultivating their keen insight and innovative ability to be brave in innovation and practice， and making a beneficial supplement to traditional teaching. This mechanism has the following three contributions： ① integrating teaching and competition to improve the application ability of professional knowledge； ② Integrate teaching and scientific research to improve professional research ability； ③ Use graduation design to improve the comprehensive innovation ability of solving complex problems. This innovative talent training mechanism has been put into use in the teaching process， and has achieved outstanding results in computer discipline competition， scientific research paper publishing， national invention patent application， undergraduate graduation design， etc. At the same time， more than 200 excellent undergraduates have been trained， and a number of innovative technical backbones and researchers have been continuously delivered to first-class IT enterprises and well-known universities at home and abroad.关键词：first-class university;computer major;innovative talents;cultivation mechanism
- 12
- |
- 10
- |
- 0
发布时间：2022-12-27 - 摘要：With the advent of the era of big data and artificial intelligence， NoSQL databases have an increasingly profound impact on the industry and academia， and how to cultivate students' NoSQL practical application ability has become the key. However， colleges and universities do not pay enough attention to relevant courses， and students lack relevant NoSQL database practical ability. The gap between them and the big data ability required by the industry is widening. To solve this problem， introduce the team's experience in implementing NoSQL database practice teaching based on two-tier learning mode in the past five years. With the method of Benchmarking and according to students’ course evaluation， good practical teaching effect has been achieved.关键词：NoSQL;practice teaching;ability enhancement;benchmarking
- 8
- |
- 4
- |
- 0
发布时间：2022-12-27 - 摘要：The vigorous development of big data and artificial intelligence has brought changes to people's production and life. With the third wave of artificial intelligence， the reform of higher education based on emerging technologies is also unstoppable. On the basis of the investigation of the current situation of artificial intelligence education， it is analyzed that the current development of artificial intelligence education still has the problems of weak human environment， insufficient data volume and low security. Based on the current limitations， combined with the future development trend of artificial intelligence， combined with the three aspects of "teaching"， "learning" and "evaluation" in the traditional education model， analyzes the effect and feasibility of higher education enabled by artificial intelligence， puts forward some suggestions of artificial intelligence enabled precise teaching， so that artificial intelligence can optimize higher education teaching quality as a whole.关键词：artificial intelligence;higher education;online education;reform of educational mode
- 10
- |
- 7
- |
- 0
发布时间：2022-12-27 - 摘要：The database industry has entered a new stage. Domestic databases have replaced foreign databases， and the scale of user database records is very large. Therefore， new requirements are put forward for the teaching and practice of database principles course： first， the practice environment needs to use domestic databases； second， students must make good use of domestic databases； third， a group of students must be cultivated to master the core technology of databases in order to develop domestic databases. Introduce the specific methods of teaching and practice of database principle course in the School of Computer Science， Wuhan University： integrated teaching， integrating some functions and principles of database into classroom teaching； according to the characteristics of students， purposefully increase the depth of teaching； mine the cases suitable for teaching in life and learning， and improve the participation of students in the classroom； multi-level curriculum practice activities； and ideological and political elements in teaching. Through the comprehensive application of these methods， the teaching quality of database principle course can be effectively improved.关键词：integrated teaching;multi-level teaching;database core;curriculum ideological political teaching
- 7
- |
- 2
- |
- 0
发布时间：2022-12-27 - 摘要：Operating system course is an important professional basic course. However， the traditional course teaching is guided by the teaching content and teaching process， ignoring the students' teaching subject status. Therefore， the proposed agile teaching considers the individuality and differences of students， pays attention to the flexibility， integrity and iteration of the teaching system， and stimulates students' interest in learning. This paper discusses the connotation and practice of agile teaching system by taking the teaching concept of agile course as an example. Practice shows that the teaching mode realizes the interactive feedback between students and teaching effect， improves the teaching effect of operating system， and significantly improves the students' curriculum objectives. The proposed method provides a new teaching mode for the teaching of operating system course， which can be used for reference by other colleges and universities.关键词：agile teaching;teaching practice;operating system course;tracking and response;iteration
- 22
- |
- 6
- |
- 0
发布时间：2022-12-27 - 摘要：Curriculum ideology and politics is an important strategic measure to implement the fundamental task of building morality and cultivating people. According to the characteristics of the cyberspace security technology course， the ideological and political points are explored from the five levels of "country - nation - society - profession - individual"， value shaping goals are added to the syllabus， such as building the "Three Stresses" culture and cultivating students' spirit of "Honker "， etc. At the same time， the ideological and political elements in the frontier theoretical knowledge and practical skills are explored，and the course content system has been reshaped. The course ideological and political case base is built， the organic integration of ideological and political education and teaching content is achieved and the seamless connection between the course knowledge and the job skill needs is realized. In the course of teaching， the teaching process is organized and implemented by using the "two-subject and three-stage" teaching mode， and the specific implementation process of ideological and political teaching is introduced in detail with the typical ideological and political cases excavated. Finally， the effectiveness of ideological and political education is summarized and the improvement ideas for the continuous construction of Ideological and political courses are put forward.关键词：cyberspace security technology;curriculum ideology and politics;teaching practice
- 12
- |
- 8
- |
- 0
发布时间：2022-12-27 - 摘要：In recent years， with the vigorous development of artificial intelligence technology， the education field has carried out exploration and research around the topics of education informatization， robotics， in-depth learning and so on. The current application of robots has provided new methods for education at all levels. The deep integration of robots and education will effectively promote the teaching reform in the intelligent era and the improvement of students' learning effect. It explores the general situation of the integration and development of education and robotics in the past 10 years， presents the research hotspots and development status of robot application in education by using the bibliometric method， and finds that the research in this field mainly focuses on hot topics such as special education， medical education， interdisciplinary integration， etc； At the same time， it also considers the integration of education and teaching and robot， and puts forward suggestions on the development path from the three directions of multi-disciplinary integration， teachers' main body and students' main body.关键词：education;robot;artificial intelligence;teaching reform
- 8
- |
- 2
- |
- 0
发布时间：2022-12-27 - 摘要：In order to understand the current situation of the digital competence of the keynote teachers in the online education industry，based on the recruitment information of Internet education enterprises and the existing research results， a digital competency questionnaire for the keynote teachers of Internet education enterprises is prepared by combining the literature research method， job analysis method and questionnaire survey method to obtain data. The data shows that the overall number of teachers is more women than men， the age structure is reasonable， and the educational level is high. However， the requirements for professional background and teacher qualification are not strict， and the digital competency level of practitioners is generally general. Although they perform well in sense of responsibility， patience， team cooperation， etc.， the ability to create a digital environment is insufficient. At the same time， the competency is divided into five dimensions： digital teaching ability， digital technology ability， digital learning and innovation， digital value and pursuit，basic personality traits for in-depth analysis， and development suggestions are put forward for reference in the selection， training and performance evaluation of enterprise teachers.关键词：digital competency;lead teacher;online education;preliminary conceptualization;job analysis method
- 8
- |
- 4
- |
- 0
发布时间：2022-12-27 - 摘要：With the development of artificial intelligence and other technologies driving "Internet + education" into a new era， and the presence of COVID-19 as a regular epidemic， open online courses platform， as a typical place of online learning， bears the responsibility of large-scale teaching. Although online learning courses are presented in the form of large-scale open learning， personalized learning of learners is still essential， and the development of artificial intelligence provides practical possibilities for this. Starting from the relationship among learners， teachers， and the machine， the research explores and puts forward the collaborative mechanism of open online courses platform based on the concept of "human-computer collaboration" in six aspects：personalized learning goals determine， personalized learning content generation， personalized intervention dynamic learning process， and promote the collaborative learning support of personalized knowledge construction， personalized academic evaluation and feedback ， in order to make large-scale personalized online learning possible.关键词：open online courses platform;human-computer collaboration;large scale;personalized learning
- 6
- |
- 2
- |
- 0
发布时间：2022-12-27 - 摘要：In order to cultivate the ability of computer undergraduate students to solve complex engineering problems， the system of cultivating college students' innovation and entrepreneurship ability is constructed， and the innovative practice curriculum system and experimental content suitable for ordinary undergraduate students are designed， with the new engineering professional certification requirements as the starting point， and the ability of computer talents to solve complex engineering problems as the goal. According to the practice results of the innovative practice project of "Two Connections and Six Emphasis"， this paper analyzes the methods and implementation plans for ordinary undergraduate universities to cultivate the ability of computer undergraduates to solve complex engineering problems， and explores the long-term operation mechanism and incentive mechanism for engineering students to learn， so as to effectively improve the learning effectiveness of science and engineering students.关键词：complex engineering problem;Innovative Practice;computer discipline;new engineering
- 7
- |
- 3
- |
- 0
发布时间：2022-12-27 - 摘要：In the post epidemic era， many overseas students are temporarily unable to return to university and can only have the online classes with the network platform. In order to improve the initiative of foreign students to study at home and improve the teaching effect， this paper first explores the teaching method of the combination of teaching on online live platform + recording and broadcasting of micro class + social platform. Then， according to the characteristics of Microcomputer Principle and Interface Technology for foreign students， and making full use of the online teaching platform supported by modern network technology， multi-dimensional online teaching links are designed before， during and after class. The teaching practice of CS and EIE majors in many semesters shows that this model can promote the independent learning of foreign students at home， obtain good practical teaching effect， and has good practicality and practical application value. Finally， it puts forward some thoughts on the problems faced by foreign students' online teaching.关键词：foreign students in China;online teaching;post-epidemic era;teaching method;teaching links
- 6
- |
- 4
- |
- 0
发布时间：2022-12-27 - 摘要：In order to effectively improve the innovation ability and engineering practice ability of engineering postgraduates， taking computer vision course as an example， by analyzing the problems existing in the teaching of computer courses for postgraduates， a blended teaching reform based on scientific research empowerment and project driven is implemented. Through the introduction of CDIO engineering education mode and the cultivation of innovation consciousness， moral quality and professional quality throughout the whole process of education and teaching， put forward a new TP （technology， professionalism） + CDIO hybrid teaching method and its evaluation mechanism based on hierarchical project driven. The teaching practice shows that this method can effectively promote the improvement of postgraduates' skill level and the advancement of ability structure.关键词：postgraduate education;innovative ability;CDIO;hybrid teaching
- 8
- |
- 6
- |
- 0
发布时间：2022-12-27 - 摘要：Under the realistic background， the gradual upgrading of network complexity and dynamic characteristics put forward higher requirements for the flexibility and adaptability of complex network research. Role discovery is one of the cores of complex network analysis，and it has good network adaptability. Graph based role discovery method is the earliest research method， but its problems of high computational complexity and poor flexibility are becoming more and more prominent with the increase of network complexity. Feature based role discovery algorithm has attracted even more attention because of its adaptability； multidimensional features have gradually replaced two-dimensional measurement and become the mainstream method of role discovery. First introduces the static role discovery methods， including graph based and feature-based， then summarizes the research status of dynamic role， and finally puts forward the research prospect of role discovery from the theoretical progress and practical needs.关键词：complex network;network feature;role discovery algorithm;dynamic network
- 22
- |
- 11
- |
- 0
发布时间：2022-12-27 - 摘要：Educational evaluation is an important part of education. The development of big data provides new ideas for educational evaluation. In order to grasp the main current situation and development trend of domestic education evaluation research supported by big data， 90 core journals of CNKI were analyzed by using Citespace， Bicomb and SPSS software， this is of great significance to the quantitative analysis of the research hotspot and development trend in this field. Review the types and quantities， authors， institutions， research hotspots and development trends of educational evaluation literature research in China， It is found that the number of theoretical studies is significantly higher than that of empirical studies； Study authors are scattered， Education evaluation mainly revolves around five areas， respectively is： education quality evaluation research， evaluation of the ideological and political education， education， big data research， technology research and education evaluation data driven learning evaluation research.关键词：big data;educational evaluation;CiteSpace;clustering analysis;multidimensional scaling analysis
- 7
- |
- 2
- |
- 0
发布时间：2022-12-27 - 摘要：Virtual assistant provides services that require extensive collection and storage of users' voice data， which has potential security risks for language biological information. With the international journal literature published between 2014 and June 2021 as the main sample， the narrative review method and coding analysis method are used to sort out and analyze the research on international virtual assistant and user language bio information security issues from the aspects of research theme， main achievements and development trend， and predict the future key trends， core challenges and research paths in this field in China based on the current research situation in China， It provides a more comprehensive information reference for the research of bio information security of virtual assistant language.关键词：virtual assistant;language bioinformation;language bioinformation security;information safety
- 13
- |
- 2
- |
- 0
发布时间：2022-12-27 - 摘要：Human posture estimation technology is very popular in the research of computer vision， especially in the fields of human-computer interaction， film production， medical and health care. Openpose algorithm has a far-reaching impact on the research of human posture estimation， and the improvements based on this algorithm emerge one after another in practical application. Therefore， by listing 25 key algorithms， this paper introduces the development process of human posture estimation in recent five years， selects 12 improved algorithms with high feasibility and based on openpose algorithm， and classifies different objects from the perspectives of convolutional neural network and machine learning classifier， so as to look forward to the development route of human posture estimation technology in the future.关键词：human pose estimation;OpenPose;CNN;machine learning classifier
- 74
- |
- 7
- |
- 0
发布时间：2022-12-27
- Technical support is provided by Beijing Founder electronics co., LTD 鄂ICP备12016542号-70
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0