Predicting Potential Distribution of Tibetan Spruce (Picea smithiana) in Qomolangma (Mount Everest) National Nature Preserve Using Maximum Entropy Niche-based Model

Tibetan spruce (Picea smithiana) is an endemic species of the Himalayas,and it distributes only in a re-stricted area with very low number.To address the lack of detailed distributional information,we used maximum en-tropy (Maxent) niche-based model to predict the species' potential distribution from limited occurrence-only records.The location data of P.smithiana,relative bioclimatic variables,vegetation data,digital elevation model (DEM),and the derived data were analyzed in Maxent.The receiver operating characteristic (ROC) curve was applied to assess the prediction accuracy.The Maxent jackknife test was performed to quantify the training gains from data layers and the response of P.smithiana distribution to four typical environmental variables was analyzed.Results show that the model performs well at the regional scale.There is a potential for continued expansion of P.smithiana population numbers and distribution in China.P.smithiana potentially distributes in the lower reaches of Gyirong Zangbo and Poiqu rivers in Gyirong and Nyalam counties in Qomolangma (Mount Everest) National Nature Preserve (QNNP),China.The species prefers warm temperate climate in mountain area and mainly distributes in needle-leaved evergreen closed to open forest and mixed forest along the river valley at relatively low altitudes of about 2000-3000 m.Model simulations suggest that distribution patterns of rare species with few species numbers can be well predicted by Max-ent.

Spatial-temporal Evolution and Determinants of the Belt and Road Initiative: A Maximum Entropy Gravity Model Approach

The spatial interaction model is an effective way to explore the geographical disparities inherent in the Belt and Road Initiative(BRI) by simulating spatial flows. The traditional gravity model implies the hypothesis of equilibrium points without any reference to when or how to achieve it. In this paper, a dynamic gravity model was established based on the Maximum Entropy(MaxEnt) theory to estimate and monitor the interconnection intensity and dynamic characters of bilateral relations. In order to detect the determinants of interconnection intensity, a Geodetector method was applied to identify and evaluate the determinants of spatial networks in five dimensions. The empirical study clearly demonstrates a heterogeneous and non-circular spatial structure. The main driving forces of spatial-temporal evolution are foreign direct investment, tourism and railway infrastructure construction, while determinants in different sub-regions show obvious spatial differentiation. Southeast Asian countries are typically multi-island area where aviation infrastructure plays a more important role. North and Central Asian countries regard oil as a pillar industry where power and port facilities have a greater impact on the interconnection. While Western Asian countries are mostly influenced by the railway infrastructure, Eastern European countries already have relatively robust infrastructure where tariff policies provide a greater impetus.

Droplets diameter distribution using maximum entropy formulation combined with a new energy-based sub-model

The maximum entropy principle(MEP) is one of the first methods which have been used to predict droplet size and velocity distributions of liquid sprays. This method needs a mean droplets diameter as an input to predict the droplet size distribution. This paper presents a new sub-model based on the deterministic aspects of liquid atomization process independent of the experimental data to provide the mean droplets diameter for using in the maximum entropy formulation(MEF). For this purpose, a theoretical model based on the approach of energy conservation law entitled energy-based model(EBM) is presented. Based on this approach, atomization occurs due to the kinetic energy loss. Prediction of the combined model(MEF/EBM) is in good agreement with the available experimental data. The energy-based model can be used as a fast and reliable enough model to obtain a good estimation of the mean droplets diameter of a spray and the combined model(MEF/EBM) can be used to well predict the droplet size distribution at the primary breakup.

A MAXIMUM ENTROPY CHUNKING MODEL WITH N-FOLD TEMPLATE CORRECTION

This letter presents a new chunking method based on Maximum Entropy (ME) model with N-fold template correction model.First two types of machine learning models are described.Based on the analysis of the two models,then the chunking model which combines the profits of conditional probability model and rule based model is proposed.The selection of features and rule templates in the chunking model is discussed.Experimental results for the CoNLL-2000 corpus show that this approach achieves impressive accuracy in terms of the F-score:92.93%.Compared with the ME model and ME Markov model,the new chunking model achieves better performance.

A New Detection Approach Based on the Maximum Entropy Model

The maximum entropy model was introduced and a new intrusion detection approach based on the maximum entropy model was proposed. The vector space model was adopted for data presentation. The minimal entropy partitioning method was utilized for attribute diseretization. Experiments on the KDD CUP 1999 standard data set were designed and the experimental results were shown. The receiver operating eharaeteristie（ROC） curve analysis approach was utilized to analyze the experimental results. The analysis results show that the proposed approach is comparable to those based on support vector maehine（SVM） and outperforms those based on C4.5 and Naive Bayes classifiers. According to the overall evaluation result, the proposed approach is a little better than those based on SVM.

RESEARCH OF PINYIN-TO-CHARACTER CONVERSION BASED ON MAXIMUM ENTROPY MODEL

This paper applied Maximum Entropy (ME) model to Pinyin-To-Character (PTC) conversion in-stead of Hidden Markov Model (HMM) that could not include complicated and long-distance lexical informa-tion. Two ME models were built based on simple and complex templates respectively, and the complex one gave better conversion result. Furthermore, conversion trigger pair of y A → y B cBwas proposed to extract the long-distance constrain feature from the corpus; and then Average Mutual Information (AMI) was used to se-lect conversion trigger pair features which were added to the ME model. The experiment shows that conver-sion error of the ME with conversion trigger pairs is reduced by 4% on a small training corpus, comparing with HMM smoothed by absolute smoothing.

Prediction of the global potential geographical distribution of Hylurgus ligniperda using a maximum entropy model

Background: Hylurgus ligniperda(Fabricius) is native to Europe but has established populations in many countries and regions. H. ligniperda mainly infests Pinus species, and can cause severe weakness and even death of the host through its boring activity;it can also be a vector of various pathogenic fungi. This study was conducted to investigate the environmental variables limiting the distribution of H. ligniperda and the change trend of its suitable areas under climate change.Results: We used a maximum entropy model to predict the potential geographical distribution of H. ligniperda on a global scale under near current and future climatic scenarios using its occurrence data and environmental variables. The result shows that the areas surrounding the Mediterranean region, the eastern coastal areas of Asia, and the southeastern part of Oceania are highly suitable for H. ligniperda. The environmental variables with the greatest effect on the distribution of H. ligniperda were determined using the jackknife method and Pearson’s correlation analysis and included the monthly average maximum temperature in April, precipitation of driest quarter, the monthly average minimum temperature in December, precipitation of coldest quarter, mean temperature of driest quarter and mean diurnal range.Conclusions: Excessive precipitation in winter and low temperatures in spring had a great effect on the distribution of H. ligniperda. The potential geographical distribution of H. ligniperda was predicted to change under future climatic conditions compared with near current climate conditions. Highly suitable areas, moderately suitable areas and low suitable areas were predicted to increase by 59.99%, 44.43% and 22.92%, respectively, under the2081–2100 ssp245 scenario.

Video segmentation using Maximum Entropy Model

Detecting objects of interest from a video sequence is a fundamental and critical task in automated visual surveillance. Most current approaches only focus on discriminating moving objects by background subtraction whether or not the objects of interest can be moving or stationary. In this paper, we propose layers segmentation to detect both moving and stationary target objects from surveillance video. We extend the Maximum Entropy (ME) statistical model to segment layers with features, which are collected by constructing a codebook with a set of codewords for each pixel. We also indicate how the training models are used for the discrimination of target objects in surveillance video. Our experimental results are presented in terms of the success rate and the segmenting precision.

Maximum Entropy Distribution Function and Uncertainty Evaluation Criteria

Marine environmental design parameter extrapolation has important applications in marine engineering and coastal disaster prevention.The distribution models used for environmental design parameter usually pass the hypothesis tests in statistical analysis,but the calculation results of different distribution models often vary largely.In this paper,based on the information entropy,the overall uncertainty test criteria were studied for commonly used distributions including Gumbel,Weibull,and Pearson-III distribution.An improved method for parameter estimation of the maximum entropy distribution model is proposed on the basis of moment estimation.The study in this paper shows that the number of sample data and the degree of dispersion are proportional to the information entropy,and the overall uncertainty of the maximum entropy distribution model is minimal compared with other models.

Maximum entropy spectral characteristics of seismic activity for great earthquakes in China

The maximum entropy spectral characteristics of seismicity in the seismic enhanced region of 11 great earthquakes is analysed in this paper to seek the difference of seismic period spectral structure between the normal and the abnormal stage of seismic activity in this paper. The results show that, during decades or even one hundred years before great earthquakes, only short periods with 6.5～24.3 years appear, and long ones disappear. Otherwise, long periods with 18.5～38.5 years exist chiefly within the normal stages. Decades years after great earthquakes, the period spectra of seismicity are generally about several or ten years. Then the characteristics of great earthquakes is explained physically by applying the strong body seismogenic model, so a method of studying and predicting great earthquakes is offered.

Subjectivity in Application of the Principle of Maximum Entropy

Complete prior statistical information is currently required in the majority of statistical evaluations of complex models. The principle of maximum entropy is often utilized in this context to fill in the missing pieces of available information and is normally claimed to be fair and objective. A rarely discussed aspect is that it relies upon testable information, which is never known but estimated, i.e. results from processing of raw data. The subjective choice of this processing strongly affects the result. Less conventional posterior completion of information is equally accurate but is computationally superior to prior, as much less information enters the analysis. Our recently proposed methods of lean deterministic sampling are examples of very few approaches that actively promote the use of minimal incomplete prior information. The inherited subjective character of maximum entropy distributions and the often critical implications of prior and posterior completion of information are here discussed and illustrated, from a novel perspective of consistency, rationality, computational efficiency and realism.

Fourth-Order Predictive Modelling: I. General-Purpose Closed-Form Fourth-Order Moments-Constrained MaxEnt Distribution

This work (in two parts) will present a novel predictive modeling methodology aimed at obtaining “best-estimate results with reduced uncertainties” for the first four moments (mean values, covariance, skewness and kurtosis) of the optimally predicted distribution of model results and calibrated model parameters, by combining fourth-order experimental and computational information, including fourth (and higher) order sensitivities of computed model responses to model parameters. Underlying the construction of this fourth-order predictive modeling methodology is the “maximum entropy principle” which is initially used to obtain a novel closed-form expression of the (moments-constrained) fourth-order Maximum Entropy (MaxEnt) probability distribution constructed from the first four moments (means, covariances, skewness, kurtosis), which are assumed to be known, of an otherwise unknown distribution of a high-dimensional multivariate uncertain quantity of interest. This fourth-order MaxEnt distribution provides optimal compatibility of the available information while simultaneously ensuring minimal spurious information content, yielding an estimate of a probability density with the highest uncertainty among all densities satisfying the known moment constraints. Since this novel generic fourth-order MaxEnt distribution is of interest in its own right for applications in addition to predictive modeling, its construction is presented separately, in this first part of a two-part work. The fourth-order predictive modeling methodology that will be constructed by particularizing this generic fourth-order MaxEnt distribution will be presented in the accompanying work (Part-2).

基于CAN和REST物联网技术的智能矿山安全检测系统研发

针对矿井安全生产检测数据传输效率低下和共享性差的特点,综合考虑开发成本与工作环境要求,基于CAN(Controller Area Network)和REST(Representational State Transfer)物联网技术提出了智能矿山安全检测方法,设计了矿山安全检测判别程序,采用最大熵模型算法开发了数据信息预警程序。结合CAN总线技术,将多传感器信息进行有机融合并进行安全数据检测,将井下传感器设备相关信息经过判断分析后传输至总机。将所提安全检测方法进行了系统开发,并在淮北某矿进行了应用。结果表明:基于CAN和REST物联网技术的安全检测方法能够实现多点测量,并可随机增减检测设备,可实现数据实时传输和共享,有助于实现矿山安全实时检测。

气候变化对中国沿海红树林潜在分布格局的影响

探明气候变化对红树林分布的影响对红树林的保护与恢复具有重要意义。研究设置了3种气候变化情景(SSP1_2.6、SSP2_4.5和SSP5_8.5),运用MaxEnt模型预测当前时段(1970—2014年)与未来时期(2090s)的气候条件下红树林的潜在分布区,探究红树林分布格局对气候变化的响应。结果表明:MaxEnt模型预测效果较好,测试集的AUC值均大于0.90,TTS值超过0.75;当前气候条件下,红树林的潜在分布区约占中国沿海区域的3.16%,主要集中在福建、台湾地区、广东、海南和广西东南部的盐沼滩涂和河口湿地;制约红树林分布的关键因子为年平均气温、平均日较差、最冷月最低温、海拔和土壤电导率,5个变量的累积贡献率达到了80%以上;未来3种气候变化情景下,红树林的潜在分布区不断向高纬度方向扩展,最北端可能达到浙江中部到江苏南部的海岸带,质心位置也呈现出地带性转移。但3种气候变化情景下的红树林潜在分布面积发生了变化,其中,在SSP5_8.5情景下,红树林分布面积表现为负增长,由当前的3.16%下降到2.76%。研究结果可为中国沿海红树林北向迁移提供一定科学依据。

气候变化情景下大花杓兰在中国的适生区预测

大花杓兰(Cypripedium macranthos)隶属兰科杓兰属,是国家二级重点保护野生植物,与大多数杓兰属植物分布在我国西南山区不同,主要分布于我国的华北、东北和台湾等地区。多年来,过渡采挖等导致了大花杓兰种群数量和个体数目急剧下降。鉴于大花杓兰特殊的分布格局和濒危现状,选择过去、当前和未来8个气候情景,利用MaxEnt物种分布模型结合38个环境变量及来源于数据库和最新实地调查的80个分布位点进行建模,分析了影响大花杓兰分布的关键环境变量,预测了其在当前、过去和未来气候情景下的适生区及其分布中心和迁移趋势。结果表明:当前情景下,大花杓兰适生区主要分布在我国东北和华北地区。影响其分布的5个关键环境变量分别是:UV-B最强月份均值(UV-B3,贡献率:54.0%)、森林覆盖率(FOR,贡献率:14.3%)、降水量季节性变化(BIO15,贡献率:7.4%)、温度季节性变动系数(BIO4,贡献率:6.8%)和草/灌木/林地(GRS,贡献率:4.6%)。其中,紫外辐射相关变量是首次被运用在杓兰属植物的适生区分布预测中,并被证实对大花杓兰的分布具有重要影响。过去3个气候情景下大花杓兰总适生区面积受冰期影响呈现先减少后增加的状态,未来4个气候情景与当前情景相比有增加的趋势。大花杓兰适生区分布中心除末次盛冰期位于河北省外,其余7个气候情景下均位于与辽宁省相近的内蒙古自治区东南部区域。气候变化背景下,采取对预测的适生区全面资源调查、生境和生长监测、科普宣教等策略对大花杓兰的保护至关重要。本研究的预测结果能够为大花杓兰甚至杓兰属植物的保育策略的制定提供理论性参考。

敏感设备电压暂降故障样本增广与概率评估最大熵建模

敏感设备电压暂降故障概率评估面临样本小和先验知识不足两大难题。本文基于自编码技术和最大熵原理,提出一种适用于小样本的设备故障概率评估随机建模方法。首先,考虑敏感设备主要对暂降幅值和持续时间特征敏感,电压耐受曲线(VTC)存在不确定性的实际情况,利用自适应Kmeans聚类算法对样本暂降幅值、持续时间聚类,在稀疏自编码器(SAE)损失函数中添加VTC不确定约束进行样本特征学习,提出基于SAE–自适应Kmeans的故障样本增广方法。其次,针对先验知识不足问题,提出基于增广样本的设备故障概率评估最大熵建模方法。最后,以个人计算机为例,在VTC概率密度函数服从均匀、正态和不同指数分布且样本数仅为5的情况下进行验证,与传统最大熵法、未引入自适应Kmeans聚类进行VTC不确定区域约束的SAE样本增广进行比较,同时与先验知识不足情况下基于主观假设的评估方法进行比较。结果表明,所提方法适用于小样本和不同分布,评估结果误差低于传统最大熵法与基于主观假设的方法,验证了稀疏自编码样本增广和最大熵建模方法对于小样本设备故障概率评估的有效性、合理性和可行性。

云南烤烟适宜种植区对气候变化的响应

为揭示云南省烤烟适宜种植区对气候变化的响应情况,以烤烟在云南省的实测分布数据以及相关环境变量数据为基础,运用最大熵(MaxEnt)模型,研究历史时期云南省烤烟适宜种植区的分布特征,并分析未来气候变化条件下云南烤烟适宜种植区的变化情况。结果显示:(1)按环境变量对云南烤烟种植的贡献率大小排序,影响云南烤烟种植的主导环境变量依次为5月平均降水量、最热季节平均温度、土壤质地分类、土壤碳酸盐含量,4项的累计贡献率达到80.9%,且总体上气候变量对云南烤烟种植的影响高于土壤变量。(2)历史时期,云南烤烟适宜种植区主要分布在曲靖、昆明、楚雄、昭通、大理、保山、临沧、玉溪东部、红河中北部、文山西北部等地区,其余丽江南部、普洱中部亦有零星分布,适宜种植区的总面积达23.82×10^(6)hm^(2),占全省面积的60.44%。(3)在RCP4.5和RCP8.5排放路径下,烤烟适宜种植区面积呈现不同程度的增加趋势,分别增加到28.28×10^(6)hm^(2)和26.09×10^(6)hm^(2),分别占全省面积的71.76%和66.19%,总适宜种植区的质心(几何中心)沿偏东偏北方向迁移,且这种迁移主要是由高适宜种植区质心的迁移引起。

气候变化背景下青藏高原特有种唐古特红景天的地理分布格局预测

为探究唐古特红景天(Rhodiola tangutica)在青藏高原的潜在适宜分布及其生态适应性,基于唐古特红景天青藏高原38个分布位点以及当前和未来5个时期7个环境因子,利用MaxEnt模型构建唐古特红景天分布与环境因子的关系模型,模拟和验证当前时期1970—2000年与SSP245情景模式下未来4个时期(2030s、2050s、2070s和2090s)的潜在地理分布格局。结果表明:(1)海拔(Alt)和最干月份降水量(Bio14)是影响唐古特红景天地理分布格局的最主要生态因子,贡献率分别为89.3%和4.9%。(2)当前气候背景下,唐古特红景天在青藏高原地区的适宜分布区总面积为195.21×10^(4) km^(2),主要位于青海南部和东北部、四川西北部及西藏中部等。(3)与当前时期相比,未来4个时期唐古特红景天的适宜分布区总面积变化不太明显,但高适宜区面积却有所增加,呈现出向青藏高原中部集中分布的趋势。

中国水生蔬菜作物野生近缘种多样性热点区域模拟

水生蔬菜作物是长江流域农耕文明驯化作物中的独特类群,其野生近缘种是育种和品种改良的宝贵遗传资源。本研究从中国植物+物种信息系统和中国数字植物标本馆获取了10种水生蔬菜作物的36个野生近缘种的地理分布数据并进行了校对,结合相应的气候因子数据,使用最大熵模型(maximum entropy model,MaxEnt)构建生态位模型,并对物种未来分布变化进行预测。结果表明:水生蔬菜作物野生近缘种多样性热点区域主要位于长江中下游的洞庭湖和鄱阳湖流域,与长江流域农耕文明起源地相重合,表明农耕文明对植物资源的就近利用。在未来升温的气候情景下,水生蔬菜作物野生近缘种适宜分布热点区域可能会进一步扩张,多样性热点区域中心向高纬度地区迁移,这有利于水生蔬菜作物野生近缘种多样性的维持和提升,并为其保护提供了理论依据。

气候变化背景下的天山云杉潜在分布区预测

天山云杉(Picea schrenkiana)是天山地区最主要的树种之一,为天山山地的水土保持和水源涵养发挥着重要的作用。本文基于气候相似性原理以最大熵(MaxEnt)模型为基础结合ArcGIS预测当前(2000—2020年)与2020—2040年、2040—2060年两个时段3种气候情景低温室气体排放条件(SSP1-2.6)、中温室气体排放条件(SSP3-7.0)和高温室气体排放条件(SSP5-8.5)下天山云杉的潜在分布范围,并分析影响天山云杉分布的主要环境因子。结果表明:(1)MaxEnt模型对天山云杉的分布区预测可信度高,所有模型AUC值均大于0.99。温度(等温性、季节性温度变异、年平均温度)与降水(最冷季度的降水量、最湿月的降水量、最干季度的降水量)是影响天山云杉潜在分布的主导因子;其中温度为当前主要的影响因子,最冷季度的降水量为未来的主要影响因子。(2)当前时期天山云杉的适生区主要分布在新疆、青海、内蒙古、西藏、甘肃、宁夏、陕西、四川等地区的山区,总适生区面积为299.17×10^(4)km^(2),高适区面积为49.45×10^(4)km^(2)。未来各情景下天山云杉的潜在适生区面积变化不大且分布仍以这些地区为主,但高适生区较当前均表现增加。除2020—2040年SSP5-8.5情景下天山云杉的适宜分布向东南方向迁移外,其他情景天山云杉的适宜分布有向西扩展的趋势。