Machine learning algorithms are an important measure with which to perform landslide susceptibility assessments, but most studies use GIS-based classification methods to conduct susceptibility zonation.This study pres...Machine learning algorithms are an important measure with which to perform landslide susceptibility assessments, but most studies use GIS-based classification methods to conduct susceptibility zonation.This study presents a machine learning approach based on the C5.0 decision tree(DT) model and the K-means cluster algorithm to produce a regional landslide susceptibility map. Yanchang County, a typical landslide-prone area located in northwestern China, was taken as the area of interest to introduce the proposed application procedure. A landslide inventory containing 82 landslides was prepared and subsequently randomly partitioned into two subsets: training data(70% landslide pixels) and validation data(30% landslide pixels). Fourteen landslide influencing factors were considered in the input dataset and were used to calculate the landslide occurrence probability based on the C5.0 decision tree model.Susceptibility zonation was implemented according to the cut-off values calculated by the K-means cluster algorithm. The validation results of the model performance analysis showed that the AUC(area under the receiver operating characteristic(ROC) curve) of the proposed model was the highest, reaching 0.88,compared with traditional models(support vector machine(SVM) = 0.85, Bayesian network(BN) = 0.81,frequency ratio(FR) = 0.75, weight of evidence(WOE) = 0.76). The landslide frequency ratio and frequency density of the high susceptibility zones were 6.76/km^(2) and 0.88/km^(2), respectively, which were much higher than those of the low susceptibility zones. The top 20% interval of landslide occurrence probability contained 89% of the historical landslides but only accounted for 10.3% of the total area.Our results indicate that the distribution of high susceptibility zones was more focused without containing more " stable" pixels. Therefore, the obtained susceptibility map is suitable for application to landslide risk management practices.展开更多
In this work, a hardware intrusion detection system （IDS） model and its implementation are introduced to perform online real-time traffic monitoring and analysis. The introduced system gathers some advantages of man...In this work, a hardware intrusion detection system （IDS） model and its implementation are introduced to perform online real-time traffic monitoring and analysis. The introduced system gathers some advantages of many IDSs： hardware based from implementation point of view, network based from system type point of view, and anomaly detection from detection approach point of view. In addition, it can detect most of network attacks, such as denial of services （DOS）, leakage, etc. from detection behavior point of view and can detect both internal and external intruders from intruder type point of view. Gathering these features in one IDS system gives lots of strengths and advantages of the work. The system is implemented by using field programmable gate array （FPGA）, giving a more advantages to the system. A C5.0 decision tree classifier is used as inference engine to the system and gives a high detection ratio of 99.93%.展开更多
A fuzzy clustering analysis based phonetic tied-mixture HMM(FPTM) was presented to decrease parameter size and improve robustness of parameter training. FPTM was synthesized from state-tied HMMs by a modified fuzzy C-...A fuzzy clustering analysis based phonetic tied-mixture HMM(FPTM) was presented to decrease parameter size and improve robustness of parameter training. FPTM was synthesized from state-tied HMMs by a modified fuzzy C-means clustering algorithm. Each Gaussian codebook of FPTM was built from Gaussian components within the same root node in phonetic decision tree. The experimental results on large vocabulary Mandarin speech recognition show that compared with conventional phonetic tied-mixture HMM and state-tied HMM with approximately the same number of Gaussian mixtures, FPTM achieves word error rate reductions by 4.84% and 13.02% respectively. Combining the two schemes of mixing weights pruning and Gaussian centers fuzzy merging, a significantly parameter size reduction was achieved with little impact on recognition accuracy.展开更多
Objective: According to RFM model theory of customer relationship management, data mining technology was used to group the chronic infectious disease patients to explore the effect of customer segmentation on the mana...Objective: According to RFM model theory of customer relationship management, data mining technology was used to group the chronic infectious disease patients to explore the effect of customer segmentation on the management of patients with different characteristics. Methods: 170,246 outpatient data was extracted from the hospital management information system (HIS) during January 2016 to July 2016, 43,448 data was formed after the data cleaning. K-Means clustering algorithm was used to classify patients with chronic infectious diseases, and then C5.0 decision tree algorithm was used to predict the situation of patients with chronic infectious diseases. Results: Male patients accounted for 58.7%, patients living in Shanghai accounted for 85.6%. The average age of patients is 45.88 years old, the high incidence age is 25 to 65 years old. Patients was gathered into three categories: 1) Clusters 1—Important patients (4786 people, 11.72%, R = 2.89, F = 11.72, M = 84,302.95);2) Clustering 2—Major patients (23,103, 53.2%, R = 5.22, F = 3.45, M = 9146.39);3) Cluster 3—Potential patients (15,559 people, 35.8%, R = 19.77, F = 1.55, M = 1739.09). C5.0 decision tree algorithm was used to predict the treatment situation of patients with chronic infectious diseases, the final treatment time (weeks) is an important predictor, the accuracy rate is 99.94% verified by the confusion model. Conclusion: Medical institutions should strengthen the adherence education for patients with chronic infectious diseases, establish the chronic infectious diseases and customer relationship management database, take the initiative to help them improve treatment adherence. Chinese governments at all levels should speed up the construction of hospital information, establish the chronic infectious disease database, strengthen the blocking of mother-to-child transmission, to effectively curb chronic infectious diseases, reduce disease burden and mortality.展开更多
目的构建男男同性性行为人群(men who have sex with men,MSM)丙型肝炎病毒(hepatitis C virus,HCV)感染高风险行为评价工具。方法本研究组首先开发HCV感染高风险行为评价工具,2019年12月20日—2020年1月14日利用社交软件平台,通过在线...目的构建男男同性性行为人群(men who have sex with men,MSM)丙型肝炎病毒(hepatitis C virus,HCV)感染高风险行为评价工具。方法本研究组首先开发HCV感染高风险行为评价工具,2019年12月20日—2020年1月14日利用社交软件平台,通过在线调查收集目标人群相关信息,对该工具进行评估,采用决策树模型进行数据分析。结果HCV感染高风险行为评价工具包含的6个条目并全部纳入树模型,树模型包括5层,27个节点,模型Risk估计量为0.085,模型预测正确率为91.52%,树模型索引图和收益图显示模型拟合良好。重要性评价结果显示,对MSM人群HCV感染风险影响由大至小的条目依次为:HIV结果、毒品使用、性病或相关症状、安全套使用、群交和创伤性操作。结论本研究开发的HCV感染高风险行为评价工具简单、易操作,可用于评价MSM人群的HCV感染高风险行为,为精准行为干预提供科学依据。展开更多
基金This research is funded by the National Natural Science Foundation of China(Grant Nos.41807285 and 51679117)Key Project of the State Key Laboratory of Geohazard Prevention and Geoenvironment Protection(SKLGP2019Z002)+3 种基金the National Science Foundation of Jiangxi Province,China(20192BAB216034)the China Postdoctoral Science Foundation(2019M652287 and 2020T130274)the Jiangxi Provincial Postdoctoral Science Foundation(2019KY08)Fundamental Research Funds for National Universities,China University of Geosciences(Wuhan)。
文摘Machine learning algorithms are an important measure with which to perform landslide susceptibility assessments, but most studies use GIS-based classification methods to conduct susceptibility zonation.This study presents a machine learning approach based on the C5.0 decision tree(DT) model and the K-means cluster algorithm to produce a regional landslide susceptibility map. Yanchang County, a typical landslide-prone area located in northwestern China, was taken as the area of interest to introduce the proposed application procedure. A landslide inventory containing 82 landslides was prepared and subsequently randomly partitioned into two subsets: training data(70% landslide pixels) and validation data(30% landslide pixels). Fourteen landslide influencing factors were considered in the input dataset and were used to calculate the landslide occurrence probability based on the C5.0 decision tree model.Susceptibility zonation was implemented according to the cut-off values calculated by the K-means cluster algorithm. The validation results of the model performance analysis showed that the AUC(area under the receiver operating characteristic(ROC) curve) of the proposed model was the highest, reaching 0.88,compared with traditional models(support vector machine(SVM) = 0.85, Bayesian network(BN) = 0.81,frequency ratio(FR) = 0.75, weight of evidence(WOE) = 0.76). The landslide frequency ratio and frequency density of the high susceptibility zones were 6.76/km^(2) and 0.88/km^(2), respectively, which were much higher than those of the low susceptibility zones. The top 20% interval of landslide occurrence probability contained 89% of the historical landslides but only accounted for 10.3% of the total area.Our results indicate that the distribution of high susceptibility zones was more focused without containing more " stable" pixels. Therefore, the obtained susceptibility map is suitable for application to landslide risk management practices.
文摘In this work, a hardware intrusion detection system （IDS） model and its implementation are introduced to perform online real-time traffic monitoring and analysis. The introduced system gathers some advantages of many IDSs： hardware based from implementation point of view, network based from system type point of view, and anomaly detection from detection approach point of view. In addition, it can detect most of network attacks, such as denial of services （DOS）, leakage, etc. from detection behavior point of view and can detect both internal and external intruders from intruder type point of view. Gathering these features in one IDS system gives lots of strengths and advantages of the work. The system is implemented by using field programmable gate array （FPGA）, giving a more advantages to the system. A C5.0 decision tree classifier is used as inference engine to the system and gives a high detection ratio of 99.93%.
基金Supported by the Science and TechnologyCommittee of Shanghai (0 1JC14 0 3 3 )
文摘A fuzzy clustering analysis based phonetic tied-mixture HMM(FPTM) was presented to decrease parameter size and improve robustness of parameter training. FPTM was synthesized from state-tied HMMs by a modified fuzzy C-means clustering algorithm. Each Gaussian codebook of FPTM was built from Gaussian components within the same root node in phonetic decision tree. The experimental results on large vocabulary Mandarin speech recognition show that compared with conventional phonetic tied-mixture HMM and state-tied HMM with approximately the same number of Gaussian mixtures, FPTM achieves word error rate reductions by 4.84% and 13.02% respectively. Combining the two schemes of mixing weights pruning and Gaussian centers fuzzy merging, a significantly parameter size reduction was achieved with little impact on recognition accuracy.
文摘Objective: According to RFM model theory of customer relationship management, data mining technology was used to group the chronic infectious disease patients to explore the effect of customer segmentation on the management of patients with different characteristics. Methods: 170,246 outpatient data was extracted from the hospital management information system (HIS) during January 2016 to July 2016, 43,448 data was formed after the data cleaning. K-Means clustering algorithm was used to classify patients with chronic infectious diseases, and then C5.0 decision tree algorithm was used to predict the situation of patients with chronic infectious diseases. Results: Male patients accounted for 58.7%, patients living in Shanghai accounted for 85.6%. The average age of patients is 45.88 years old, the high incidence age is 25 to 65 years old. Patients was gathered into three categories: 1) Clusters 1—Important patients (4786 people, 11.72%, R = 2.89, F = 11.72, M = 84,302.95);2) Clustering 2—Major patients (23,103, 53.2%, R = 5.22, F = 3.45, M = 9146.39);3) Cluster 3—Potential patients (15,559 people, 35.8%, R = 19.77, F = 1.55, M = 1739.09). C5.0 decision tree algorithm was used to predict the treatment situation of patients with chronic infectious diseases, the final treatment time (weeks) is an important predictor, the accuracy rate is 99.94% verified by the confusion model. Conclusion: Medical institutions should strengthen the adherence education for patients with chronic infectious diseases, establish the chronic infectious diseases and customer relationship management database, take the initiative to help them improve treatment adherence. Chinese governments at all levels should speed up the construction of hospital information, establish the chronic infectious disease database, strengthen the blocking of mother-to-child transmission, to effectively curb chronic infectious diseases, reduce disease burden and mortality.
基金HIV/HCV No Co-Infection(NoCo)Program(IN-US-987-5557)
文摘目的构建男男同性性行为人群(men who have sex with men,MSM)丙型肝炎病毒(hepatitis C virus,HCV)感染高风险行为评价工具。方法本研究组首先开发HCV感染高风险行为评价工具,2019年12月20日—2020年1月14日利用社交软件平台,通过在线调查收集目标人群相关信息,对该工具进行评估,采用决策树模型进行数据分析。结果HCV感染高风险行为评价工具包含的6个条目并全部纳入树模型,树模型包括5层,27个节点,模型Risk估计量为0.085,模型预测正确率为91.52%,树模型索引图和收益图显示模型拟合良好。重要性评价结果显示,对MSM人群HCV感染风险影响由大至小的条目依次为:HIV结果、毒品使用、性病或相关症状、安全套使用、群交和创伤性操作。结论本研究开发的HCV感染高风险行为评价工具简单、易操作,可用于评价MSM人群的HCV感染高风险行为,为精准行为干预提供科学依据。
