检索结果-内蒙古大学图书馆

2nd International Conference on Management Engineering, Software Engineering and Service Sciences (ICMSS)

作者： Zhang, MinYi Zhang, Pingjian South China Univ Technol Sch Software Engn Guangzhou Guangdong Peoples R China

ISBN: (纸本)9781450354318

Academic circle are groups of similar scholars in academic networks. The academic circle also called community circle. A reliable and efficient community discovery method can play an important role in the communication of academic field. At present, there is a lack of universal judgment index and authoritative division of academic community in the academic field, and it is difficult to judge the reliability of the algorithm. This paper improves the classic DBSCAN (Density-Based Spatial clustering of Applications with Noise) clustering algorithm by according to the characteristics of the data set itself to determine the input parameters of the algorithm, and propose a new algorithm called improved DBSCAN. The new algorithm is applied to community networks and has been demonstrated superior effectiveness over DBSCAN.

关键词： Academic circle community circle DBSCAN improved DBSCAN clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

An Integrated Model for Crime Prediction Using Temporal and Spatial Factors 18

An Integrated Model for Crime Prediction Using Temporal and ...

引用

18th IEEE International Conference on Data Mining Workshops (ICDMW)

作者： Yi, Fei Yu, Zhiwen Zhuang, Fuzhen Zhang, Xiao Xiong, Hui Northwestern Polytech Univ Xian Shaanxi Peoples R China Chinese Acad Sci Key Lab Intelligent Informat Proc Inst Comp Technol Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China Najing Univ Nanjing Jiangsu Peoples R China Rutgers State Univ New Brunswick NJ USA

ISBN: (纸本)9781538691595

Given its importance, crime prediction has attracted a lot of attention in the literature, and several methods have been proposed to discover different aspects of characteristics for crime prediction. In this paper, we propose a Clustered Continuous Conditional Random Field (Clustered-CCRF) model which is able to effectively exploit both spatial and temporal factors for crime prediction in an integrated way. In particular, we observe that the crime number at one specific area is not only conditioned on its own historical records but also has high correlation to crime records from similar areas. Therefore, we propose two factors: an auto-regressed temporal correlation and a feature-based inter-area spatial correlation, to measure such patterns for crime prediction. Further, we present a tree-structured clustering algorithm to discover high similar areas based on spatial characteristics to improve the performance of our proposed model. Experiments on real-world crime dataset demonstrate the superiority of our proposed model over the state-of-the-art methods.

关键词： Crime Prediction Temporal and Spatial Factors Continuous Conditional Random Field clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

Application of Deep Learning Method in Short-term Load Forecasting of Characteristic Enterprises 18

Application of Deep Learning Method in Short-term Load Forec...

引用

International Conference on Artificial Intelligence and Cloud Computing (AICCC)

作者： Dou, Yuchen Zhang, Xinman Wu, Zhihui Zhang, Hang Xi An Jiao Tong Univ Sch Elect & Informat Engn Xian 710049 Shaanxi Peoples R China Guangdong Xian Jiaotong Univ Acad 3 Daliangdesheng East Rd Foshan 528000 Guangdong Peoples R China

ISBN: (纸本)9781450366236

Short-term load forecasting is an important basic work for the normal operation and control of power systems. The results of power load forecasting have a great impact on dispatching operation of the power system and the production operation of the enterprise. Accurate load forecasting would help improve the safety and stability of power system and save the cost of enterprise. In order to extract the effective information contained in the data and improve the accuracy of short-term load forecasting, this paper proposes a long-short term memory neural network model (LSTM) with deep learning ability for short-term load forecasting combined with clustering algorithm. Deep learning is in line with the trend of big data and has a strong ability to learn and summarize large amounts of data. Through the research on the characteristics and influencing factors of the characteristic enterprises, the collected samples are clustered to establish similar day sets. This paper also studies the impact of different types of load data on prediction and the actual problem of input training sample selection. The LSTM prediction model is built with subdividing and clustering the input load sample set. Compared with other traditional methods, the results prove that LSTM proposed has higher accuracy and applicability.

关键词： Short-term Load Forecasting Deep Learning Long-short Term Memory Neural Network Model clustering algorithm Similar Day Set

来源：评论

学校读者我要写书评

暂无评论

IOT Enabled Security System for Android users

IOT Enabled Security System for Android users

引用

International Conference on Recent Innovations in Electrical, Electronics and Communication Engineering (ICRIEECE)

作者： Roy, Sulochana Ramaiah Inst Technol Dept Comp Sci & Engn Bangalore Karnataka India

ISBN: (纸本)9781538659953

Staying connected 24 X 7 with our near and dear ones and providing them with continuous added security services, especially when women and children step out of their houses to meet their daily needs, is one of the major challenges faced in today's society. In this project we aim to propose the design of an Internet of Things - based safety device using the Raspberry Pi 3 microcontroller, along with a pair of Android applications that would assist the primary user in notifying the required faculty instantaneously during any sort of emergency condition. Not only would this help the guardian to track the current location of the primary user i.e. their ward on receiving an emergency notification with the timestamp or allow the primary user to forward emergency calls to the nearest located emergency service provider, like ambulance, police, car workshops, cab services, which is mapped by the application of a clustering algorithm but also helps in capturing the allied emergency data like images or voice clips, storing the links to those in the Firebase cloud and then streaming it live over the Android application, once the guardian tries to read the notification forwarded to him/her by the primary user. These real time data being constantly stored in the Firebase storage, acts as a concrete source of reliable information, that can be of great help to the users in future for any sort of criminal or other investigations. Also the proposed system has been so designed that both the Android application and the IOT-based safety device can work independently of each other, with each one connected to the same or different Wifi networks, as per the availability, in order to minimize the chances of failure of this safety package when applied to real life scenario's.

关键词： Internet of Things Raspberry Pi 3 Android clustering algorithm Firebase Cloud fingerprint authentication nearest located emergency service provider ambulance police car workshops cab services images or voice clips criminal or other investigations real time data

来源：评论

学校读者我要写书评

暂无评论

A new validity index for evaluating the clustering results by partitional clustering algorithms

引用

SOFT COMPUTING 2016年第3期20卷 1127-1138页

作者： Yue, Shihong Wang, Jianpei Wang, Jeenshing Bao, Xiujuan Tianjin Univ Sch Elect Engn & Automat Tianjin 300072 Peoples R China Natl Cheng Kung Univ Dept Elect Engn Tainan 701 Taiwan

Partitional clustering algorithms are the most widely used approach in clustering problems. However, how to evaluate the clustering performance of these algorithms remains unanswered due to the lack of an efficient measure for accurately representing the separation among partitioned clusters. In this paper, based on twomost commonly used partitional clustering algorithms, c-means and fuzzy c-means, and their variants, we developed a new measure, called as dual center, to represent the separation among clusters. The new measure can efficiently represent the separation among various clusters. According to the defined measure, a new validity index is proposed for evaluating the clustering performance of partitional algorithms. Two groups of benchmark datasets with different characteristics were used to validate the effectiveness of the proposed validity index. Experimental results provide evidence that the proposed validity index outperforms some existing representative validity indexes in the two groups of typical and representative datasets.

关键词： Validity index Dual center Separation measure clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

Video Anomaly Detection using Ensemble One-class Classifiers 37

Video Anomaly Detection using Ensemble One-class Classifiers

引用

37th Chinese Control Conference (CCC)

作者： Li, Gang Feng, Zuren Lv, Na Xi An Jiao Tong Univ Syst Engn Inst Xian 710049 Shaanxi Peoples R China Inner Mongolia Univ Sci & Technol Baotou 014010 Peoples R China

ISBN: (纸本)9789881563958

In this paper we present a novel algorithm for video anomaly detection. It is based on multiple local cells, which are acquired by splitting entire monitor scene. At each local cell, we group all feature vectors with clustering algorithm based on minimum spanning tree, and further model all groups using improved one-class SVM to build ensemble classifiers. For any new features at each local node in incoming video clips, we use the corresponding learned ensemble classifiers to estimate maximum abnormality degree. The proposed approach has been tested on publicly available datasets with frame-level and pixel-level criteria, and outperforms other state-of-the-art approaches.

关键词： anomaly detection clustering algorithm minimum spanning tree one-class classifier

来源：评论

学校读者我要写书评

暂无评论

A Synergistic Cloud Service Approach for Cold Start Problems 4

A Synergistic Cloud Service Approach for Cold Start Problems

引用

IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computing, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI)

作者： Qi, Rui-dong Zhou, Jian-tao Song, Xiaoyu Inner Mongolia Univ Coll Comp Sci Hohhot Peoples R China Portland State Univ Dept Elect & Comp Engn Portland OR 97207 USA

ISBN: (纸本)9781538693803

More and more new cloud users use the personalised cloud service combination strategy (CSCS). Solving the Cold Start problem of the cloud environment becomes intractable. The paper represents a novel method of choosing the most optimal combinatorial features based on the attenuation function to cluster, and integrating multi level sampling method to cope with the pure Cold Start for cloud users. By means of every clustering process with different combinatorial features, then using the relatively stable number of clusters for every clustering obtains the optimal combinatorial features, which presents the tendency of the whole society of cloud users who use the CSCS. Meanwhile, we propose the function of periodic attenuation that enhances the degree of recommendation for CSCSs which have been issued recently. We harness the vectors of preference and disfavour to calculate the similarity of cloud users. An improved cluster algorithm of CFSFDP is employed. Moreover, it is worth selecting the most representative features to cluster which demonstrates effectively. In addition, the attenuation function can increase the probability of recommendation of recent CSCS, and the multi level sampling method has been used to heighten the diversity of recommendations. The method of ours can enhance the effectiveness and intelligence of recommendation for the pure Cold Start problem.

关键词： Cold start problem cloud service combination clustering algorithm multi level sampling recommendation

来源：评论

学校读者我要写书评

暂无评论

An Energy-Efficient Adaptive Overlapping clustering Method for Dynamic Continuous Monitoring in WSNs

引用

IEEE SENSORS JOURNAL 2017年第3期17卷 834-847页

作者： Hu, Yuan Niu, Yugang Lam, James Shu, Zhan East China Univ Sci & Technol Minist Educ Key Lab Adv Control & Optimizat Chem Proc Shanghai 200237 Peoples R China Univ Hong Kong Dept Mech Engn Hong Kong Hong Kong Peoples R China Univ Southampton Fac Engn & Environm Electromech Engn Grp Southampton SO17 1BJ Hants England

clustering is a key technique to improve energy efficiency in wireless sensor networks (WSNs). In continuous monitoring applications, the clusters should be formed dynamically according to the event development for energy-efficient data gathering. In this paper, an energy-efficient adaptive overlapping clustering (EEAOC) method is proposed in WSNs for continuous monitoring applications. In EEAOC, a 2-logical-coverage overlapping clustering topology is established such that the adjacent sensors in the event area can be grouped into the same cluster for data fusion and the cluster migration operation can be processed without changing the overlapping structure among clusters. Moreover, to further reduce energy consumption, a hybrid data reporting strategy that switches between time-driven and event-driven schemes is introduced based on the QoS requirements in continuous monitoring applications. Simulation results show that EEAOC achieves a longer network lifetime cycle.

关键词： clustering algorithm event-driven clustering event tracking wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

Slice_OP: Selecting Initial Cluster Centers Using Observation Points 1

引用

14th International Conference on Advanced Data Mining and Applications (ADMA)

作者： Masud, Md Abdul Huang, Joshua Zhexue Zhong, Ming Fu, Xianghua Mahmud, Mohammad Sultan Shenzhen Univ Coll Comp Sci & Software Engn Big Data Inst Shenzhen 518060 Peoples R China

ISBN: (数字)9783030050900

ISBN: (纸本)9783030050900;9783030050894

This paper proposes a new algorithm, Slice_OP, which selects the initial cluster centers on high-dimensional data. A set of observation points is allocated to transform the high-dimensional data into one-dimensional distance data. Multiple Gamma models are built on distance data, which are fitted with the expectation-maximization algorithm. The best-fitted model is selected with the second-order Akaike information criterion. We estimate the candidate initial centers from the objects in each component of the best-fitted model. A cluster tree is built based on the distance matrix of candidate initial centers and the cluster tree is divided into K branches. Objects in each branch are analyzed with k-nearest neighbor algorithm to select initial cluster centers. The experimental results show that the Slice OP algorithm outperformed the state-of-the-art Kmeans++ algorithm and random center initialization in the k-means algorithm on synthetic and real-world datasets.

关键词： Initial cluster center clustering algorithm Center initialization Observation point

来源：评论

学校读者我要写书评

暂无评论

Mining Anomaly Communication Patterns for Industrial Control Systems

Mining Anomaly Communication Patterns for Industrial Control...

引用

Australasian Universities Power Engineering Conference (AUPEC)

作者： Yu, Tsung-Chiao Huang, Jyun-Yao Liao, I-En Kao, Kuo-Fong NCHU Taiwan Informat Secur Ctr Dept Comp Sci & Engn Taichung Taiwan Hsiuping Univ Sci & Technol Taiwan Informat Secur Ctr NCHU Dept Informat Networking Technol Taichung Taiwan

ISBN: (纸本)9781538684740

The attacks on industrial control systems ( ICS) have been exemplified by the malwares Stuxnet, Industroyer, and Triton that targeted nuclear facilities of Iran in 2010, power grid of Ukraine in 2016, and Safety Instrumented System ( SIS) controllers of a Middle East country in 2017, respectively. As a result, the issues concerning Critical Infrastructure Information Protection ( CIIP) have drawn much attention among academia, industry, and government in many countries. In this paper, we propose an anomaly detection method for ICS networks. The main idea of the proposed method is to model the normal behavior patterns of TCP and UDP payloads as frequent patterns and non-frequent pattern clusters. The normal behavior payloads are first processed by sequential pattern mining algorithm to extract frequent patterns, and then the payloads are projected against frequent patterns. After projection, the projected payloads are clustered using hierarchical agglomerative clustering algorithm to find representative variations in normal behaviors. The experimental results show that the proposed method has very good performance in terms of the metrics such as accuracy, recall, precision, false alarm, and false dismissal for the ICS networks that use Modbus/ TCP or BACnet protocols. The proposed system model can also leverage honeypots deployed in ICS networks to generate attack signatures, which can be helpful in filtering out known attacks.

关键词： clustering algorithm Industrial control Network security Power system protection SCADA systems

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：