Intelligent correlation method of typical business data in power communication operation management by Guilong WU, Zhimin YANG, Yu HUANG

“…In the process of power communication operation management, various independent business data, such as trouble tickets, duty logs, maintenance tickets, and inspection records, are generated and stored.These business data provide important support for the operation management of the power communication network.At present, the statistical process of most business data is relatively independent, and there is less manual correlation in the later stage.Two typical business data of duty log and trouble ticket in power communication operation management were selected, text mining technology was used to build a machine learning model combining unsupervised recall and supervised classification, and the intelligent association method between duty log and trouble ticket was proposed.Besides, the relevant historical business data in the electric power communication operation management system was used to do the experimental verification of the intelligent association method.The results show that it can achieve positive effect.…”
Get full text

Inside the Black Box: Detecting and Mitigating Algorithmic Bias Across Racialized Groups in College Student-Success Prediction by Denisa Gándara, Hadis Anahideh, Matthew P. Ison, Lorenzo Picchiarini

“…Using nationally representative data from the Education Longitudinal Study of 2002 and various machine learning modeling approaches, we demonstrate how models incorporating commonly used features to predict college-student success are less accurate when predicting success for racially minoritized students. …”
Get full text

Unbiased Isotonic Regression Tree for Discovering Hidden Heterogeneity in Monotonicity Constraints by Doowon Choi

“…Integrating domain knowledge is increasingly recognized as vital for improving the relevance and reliability of machine learning models. This integration is often implemented through specific types of constraints that reflect real-world conditions or theoretical insights. …”
Get full text

Substructure correlation adaptation transfer learning method based on K-means clustering by Haoshuang LIU, Yong ZHANG, Yingbo CAO

“…Domain drifts severely affect the performance of traditional machine learning methods, and existing domain adaptive methods are mainly represented by adaptive adjustment cross-domain through global, class-level, or sample-level distribution adaptation.However, too coarse global matching and class-level matching can lead to insufficient adaptation, and sample-level adaptation to noise can lead to excessive adaptation.A substructure correlation adaptation (SCOAD) transfer learning algorithm based on K-means clustering was proposed.Firstly, multiple subdomains of the source domain and the target domain were obtained by K-means clustering.Then, the matching of the second-order statistics of the subdomain center was sought.Finally, the target domain samples were classified by using the subdomain structure.The proposed method approach further improves the performance of knowledge transfer between the source and target domains on top of the traditional approach.Experimental results on common transfer learning datasets show the effectiveness of the proposed method.…”
Get full text

Research progress of KPI anomaly detection in intelligent operation and maintenance by Su WANG, Hua LU, Shuo WANG, Lei CAI, Tao HUANG

“…Existing network monitoring and fault repair mostly rely on rule systems or manual processing.However, the increase in network scale and the diversification of services make this approach difficult to deal with.With the rapid development of technology such as machine learning and deep learning, intelligent operation and maintenance theory has also made great progress, using artificial intelligence technology to enhance the intelligent ability of network operation and maintenance.KPI (key performance indicator) anomaly detection is an underlying core technology of intelligent operation and maintenance.A survey on the KPI anomaly detection technology was given.Firstly, the KPI data and KPI anomalies were described.Then the research progress of single-dimensional KPI and multi-dimensional KPI anomaly detection were introduced.Then, the deployment and application problems of KPI anomaly detection were analyzed.Finally, future research directions were discussed.…”
Get full text

PREDICTING STOCK PRICE DIRECTION OF EUROZONE BANKS: CAN DEEP LEARNING TECHNIQUES OUTPERFORM TRADITIONAL MODELS? by Bogdan Ionuț ANGHEL

“…The findings suggest that traditional machine learning models are more effective than advanced deep learning models for predicting stock price direction in the Eurozone banking sector. …”
Get full text

Smart health-care systems for rheumatology by Suneeta Mohanty, Prasant Kumar Pattnaik

“…Wireless sensors, radio-frequency identification technology, Internet of things, and machine learning (ML) algorithms are the underlying technology to implement smart health care. …”
Get full text

Design of federated routing mechanism in cross-domain scenario by Peizhuang CONG, Yuchao ZHANG, Ye TIAN, Wendong WANG, Dan LI

“…With the development of multi-network integration,how to ensure efficient interconnections among multiple independent network domains is becoming a key problem.Traditional interdomain routing protocol fails due to the limitation of domain information privacy,where each autonomous domain doesn’t share any specific intra-domain information.A machine learning-based federated routing mechanism was proposed to overcome the existing shortcomings.Each autonomous domain shares intra-domain information implicitly via neural network models and parameters.It not only breaks data islands problems but also greatly reduces the amount of transmitted data shared between domains,then decreases convergence delay of entire network information.Based on the federated routing mechanism,border routers can formulate global optimal routing strategies according to the status of entire network.…”
Get full text

Spectral optimization of supercontinuum shaping using metaheuristic algorithms, a comparative study by Mathilde Hary, Teemu Koivisto, Sara Lukasik, John M. Dudley, Goëry Genty

“…Abstract Supercontinuum generation in optical fiber involves complex nonlinear dynamics, making optimization challenging, and typically relying on trial-and-error or extensive numerical simulations. Machine learning and metaheuristic algorithms offer more efficient optimization approaches. …”
Get full text

Event Forecasting in Organizational Networks: A Discrete Dynamical System Approach by Piotr Śliwa

“…The network perspective is believed to successfully model most of the socioeconomic phenomena, which, in combination with the prospects of continuously advancing tools for automated data mining and machine learning, gives a tempting promise to effectively forecast socioeconomic events occurring in our societies and businesses. …”
Get full text

Restricted Isometry Property of Principal Component Pursuit with Reduced Linear Measurements by Qingshan You, Qun Wan, Haiwen Xu

“…The principal component prsuit with reduced linear measurements (PCP_RLM) has gained great attention in applications, such as machine learning, video, and aligning multiple images. …”
Get full text

Survey of encrypted malicious traffic detection based on deep learning by Mingfang ZHAI, Xingming ZHANG, Bo ZHAO

“…With the increasing awareness of network security,encrypted communication dominates and encrypted traffic grows rapidly.Traffic encryption,while protecting privacy,also masks illegal attempts and changes the form of threats.As one of the most important branch of machine learning,deep learning performs well in traffic classification.For several years,research on deep-learning based intrusion detection has been deepened and achieved good results.The steps of encrypted malicious traffic detection were introduced to be a general detection framework model named “six-step method”.Then,discussion and induction of data processing and detection algorithms were carried out combined with this model.Both advantages and disadvantages of various algorithm models were given as well.Finally,future research directions were pointed out with a view to providing assistance for further research.…”
Get full text

Malicious DNS traffic detection based neural networks by SHAN Kangkang, YUAN Shuhong, CHEN Wenzhi, WANG Zhibo

“…To solve the problems of low detection accuracy and speed caused by low efficiency in extracting traffic features using machine learning to detect malicious DNS traffic, a malicious DNS traffic detection method FDS-DL was proposed, which combines frequency domain feature aggregation analysis and neural networks algorithms. …”
Get full text

Research on federated learning approach based on local differential privacy by Haiyan KANG, Yuanrui JI

“…As a type of collaborative machine learning framework, federated learning is capable of preserving private data from participants while training the data into useful models.Nevertheless, from a viewpoint of information theory, it is still vulnerable for a curious server to infer private information from the shared models uploaded by participants.To solve the inference attack problem in federated learning training, a local differential privacy federated learning (LDP-FL) approach was proposed.Firstly, to ensure the federated model training process was protected from inference attacks, a local differential privacy mechanism was designed for transmission of parameters in federated learning.Secondly, a performance loss constraint mechanism for federated learning was proposed and designed to reduce the performance loss of local differential privacy federated model by optimizing the constraint range of the loss function.Finally, the effectiveness of proposed LDP-FL approach was verified by comparative experiments on MNIST and Fashion MNIST datasets.…”
Get full text

Research on 4G/5G voice quality optimization based on data mining and geographic visualization by Jian HU, Lei SUN, Yiyan YIN, Xiaokang YANG, Jingui BAI, Yejiang ZHANG

“…Traditional voice quality optimization relies on field test, case accumulation and expert experience.It is costly and inefficient to analyze problems by manual test.Through the application of data mining, decision tree machine learning algorithm, geographic visualization and other technologies, a voice experience optimization visualization platform based on big data analysis was developed, which could effectively identify the laws in voice big data, and realize the functions of correlation analysis between user voice experience index and wireless network performance index, intelligent recognition of degradation threshold and image analysis of poor quality areas.It is conducive to reduce the skill threshold of network engineers, improve the work efficiency of network optimization, save network operation and maintenance costs, and provide accurate and effective voice experience improvement solutions for the industry.…”
Get full text

Method for Creating Domain-Specific Dataset Ontologies from Text in Uncontrolled English by Minab Shokoufeh Salem, Nazaruka Erika

“…The prominent technologies for implementation of the proposed method are machine learning, including classification algorithms and natural language processing using a large language model. …”
Get full text

Network traffic classification method basing on CNN by Yong WANG, Huiyi ZHOU, Hao FENG, Miao YE, Wenlong KE

“…Since the feature selection process will directly affect the accuracy of the traffic classification based on the traditional machine learning method,a traffic classification algorithm based on convolution neural network was tailored.First,the min-max normalization method was utilized to process the traffic data and map them into gray images,which would be used as the input data of convolution neural network to realize the independent feature learning.Then,an improved structure of the classical convolution neural network was proposed,and the parameters of the feature map and the full connection layer were designed to select the optimal classification model to realize the traffic classification.The tailored method can improve the classification accuracy without the complex operation of the network traffic.A series of simulation test results with the public data sets and real data sets show that compared with the traditional classification methods,the tailored convolution neural network traffic classification method can improve the accuracy and reduce the time of classification.…”
Get full text

High-dimensional outlier detection based on deep belief network and linear one-class SVM by Haoqi LI, Na YING, Chunsheng GUO, Jinhua WANG

“…Aiming at the difficulties in high-dimensional outlier detection at present,an algorithm of high-dimensional outlier detection based on deep belief network and linear one-class SVM was proposed.The algorithm firstly used the deep belief network which had a good performance in the feature extraction to realize the dimensionality reduction of high-dimensional data,and then the outlier detection was achieved based on a one-class SVM with the linear kernel function.High-dimensional data sets in UCI machine learning repository were selected to experiment,result shows that the algorithm has obvious advantages in detection accuracy and computational complexity.Compared with the PCA-SVDD algorithm,the detection accuracy is improved by 4.65%.Compared with the automatic encoder algorithm,its training time and testing time decrease significantly.…”
Get full text

Effects of Organizational Culture on Employer Attractiveness of Hotel Firms: Topic Modeling Approach by Kyoung-Joo Lee, Sun-Yong Choi

“…This study combines an unsupervised machine learning tool for topic modeling (latent Dirichlet allocation) with the coding process of researchers to measure the different cultural attributes of hotel firms. …”
Get full text

Learning the simplicity of scattering amplitudes by Clifford Cheung, Aurélien Dersy, Matthew D. Schwartz

“…This work explores the application of machine learning to a particular facet of this challenge: the task of simplifying scattering amplitudes expressed in terms of spinor-helicity variables. …”
Get full text