应用生长、分级的自组织映射模型进行意识任务分类

提出一种使用生长、分级的自组织映射(growing hierarchical self-organizing map,GHSOM)模型进行基于EEG信号的意识任务分类来实现脑机接口技术的方法.GHSOM模型是自组织映射(self-organizing map,SOM)的一种变体,由多层的SOM组成,具有一定的分级结构,能够表达数据中不同层次的信息.同时研究了使用平均量化误差(mean quantization error,mqe)和量化误差(quantization error,qe)两种方法实现的GHSOM模型对意识任务分类的作用.结果表明,GHSOM模型对于意识任务的可分性能够提供可视化的信息,并且发现使用量化误差方法实现的GHSOM模型提供较多的数据信息和较高的分类精度.使用GHSOM模型进行了5类意识任务的分类,平均分类精度可达80%.     

基于SOM的野生秤锤树群落的空间分布和环境解释

本文采用自组织特征映射网络(self-organizing map, SOM)对南京老山野生秤锤树(Sinojackia xylocarpa)群落进行数量分类和排序, 分析了其与环境因子之间的关系。结果表明: (1) SOM将秤锤树野生群落的100个样方划分为5个群丛类型, 分类结果在空间上反映了秤锤树野生群落的演替变化趋势, 各群丛的群落结构和物种组成存在差异且群丛界限明显, 可较好地进行排序与分类的环境解释。(2)通过环境因子梯度的可视化方法, 确定了海拔、坡位和土壤厚度是影响该地区秤锤树生长和分布的主要因子, 同时也揭示了以不同优势种为代表的各群丛和环境因子的关系。(3) SOM可以摆脱许多定量技术的限制性假设, 使得神经网络对于群落生态特征及探索群落和环境相互关系具有良好展现力; SOM群落生态数据具有更高的映射能力, 进行群落分类以及较少程度的排序的潜力, 将有利于不同群落类型的分类和管理, 对于濒危植物保护具有重要意义。     

基于SOFM网络的生态水文区划

生态水文区划对缓解区域水资源开发利用和生态环境保护之间矛盾起到了重要作用。本文基于自组织特征映射(self-organizing feature map,SOFM)人工神经网络建立了生态水文区划模型。首先运用主成分分析法从众多生态水文指标中提取出能代表绝大部分信息的主成分指标;其次依据提取的主成分指标,利用系统聚类得到区域聚类谱系图;而后构建SOFM神经网络,依据网络结果和系统聚类谱系图,划分合理的生态水文区。以福建省泉州市为例进行了生态水文区划研究,将研究区划分为4类区域,各区域具有明显的生态水文特征,针对不同区域特征,提出了常规、加强、较为严格和最为严格4种生态环境保护和水资源开发策略。     

庞泉沟自然保护区华北落叶松林的自组织特征映射网络分类与排序

自组织特征映射网络（SOM）是新近引入植物生态学的分析方法,对复杂问题和非线性问题具有较强的分析和求解功能。本研究应用SOM分类和排序研究了庞泉沟自然保护区华北落叶松林。研究结果表明,SOM将120个样方分为7个植物群落类型,分类结果具有明确的生态意义;样方和物种在SOM训练图上呈现一定规律的分布;7个群落类型各有其分布范围和界限,揭示了群落间的生态关系。在此基础上,通过引入一种在SOM训练图上可视化环境因子梯度的方法,能够较好地完成样方、物种和环境因子相互关系的分析,揭示了海拔是影响该区华北落叶松林生长和分布的最主要因子。生态分析表明SOM分类和排序是一种有效的梯度分析方法,适用于表征生态特征和探索群落和环境相互关系的研究。     

基于优化核参数支持向量机的意识任务分类

根据支持向量机的基本原理,给出一种推广误差上界估计判据,并利用该判据进行最优核参数的自动选取。对三种不同意识任务的脑电信号进行多变量自回归模型参数估计,作为意识任务的特征向量,利用支持向量机进行训练和分类测试。分类结果表明,优化核参数的支持向量机分类器取得了最佳的分类效果,分类正确率明显高于径向基函数神经网络。     

自组织特征人工神经网络在庞泉沟自然保护区植物群落分类中的应用

人工神经网络是较新的数学分析工具,其中的自组织特征映射网络（SOFM）具有较强的聚类功能。应用SOFM网络对庞泉沟自然保护区植物群落进行了分类研究。在讨论了SOFM网络的数学原理、聚类方法和步骤的前提下,分类过程在MATLAB（6.5）神经网络工具箱（NNTool）中编程实现。结果将89个样方分为13个植物群落类型。分类结果符合植被实际,生态意义明确,表明SOFM网络可以很好地反映植物群落的生态关系,是非常有效的植物群落数量分类方法。     

基于SOM聚类方法在临床检验知识发现中的应用

目的:探讨基于自组织特征映射网络的知识发现方法在建立应用临床检验项目的门诊患者聚类模型中的适用性,为临床检验决策支持系统的建立提供参考.方法:以2009-2011年西安市两所最大的综合医院的内科门诊患者就诊资料为训练样本,运用MatlabR2009b软件包中的SOM Tool Box工具箱建立应用临床检验项目的门诊患者的自组织映射网络聚类模型.用Excel软件绘制柱状图描述各组门诊患者的疾病特征.结果:自组织特征映射网络训练学习后,得出聚类结果,输出结果与临床检验决策真实值具有一定一致性,训练结果表明患者的聚类结果有一定的临床意义.患者性别、年龄、3年累计临床检验项目数、疾病特征对聚类模型的贡献较大.结论:自组织映射方法对应用临床检验项目的门诊患者聚类效果较好,具有较高的应用和推广价值,可为临床检验决策支持系统的研究提供方法学依据.     

长白落叶松林生物量的模拟估测

利用样木收获法收集了34个样地中长白落叶松林分地上部分生物量信息,选取其中29个样地生物量信息分别与样地林分因子信息和TM遥感影像信息拟合建立生物量模型,利用其余5个样地的生物量信息进行模型精度检验和误差分析.结果表明：长白落叶松地上部分生物量均可用林分因子和遥感因子进行线性拟合;林分因子线性模型对长白落叶松中幼林地上生物量的估测精度较高（林分P=94.33%,遥感P=92.32%）,且检验误差较小（林分MRE=6%,遥感MRE=31%）,模型模拟效果较好;若只考虑长白落叶松中龄林,这2种模型的估测效果相当（林分模型和遥感模型的误差分别为329.9和313.6 t）.整体而言,林分因子模型估测长白落叶松树皮、干材和总生物量的效果优于遥感因子模型,对于中龄林来说,遥感模型估测叶花果、树枝和树冠生物量的效果较好.     

基于卷积神经网络-长短期记忆神经网络模型利用光学体积描记术重建动脉血压波信号

目的 直接动脉血压（arterial blood pressure,ABP）连续监测是侵入式的,传统袖带式的间接血压测量法无法实现连续监测。既往利用光学体积描记术（photoplethysmography,PPG）实现了连续无创血压监测,但其为收缩压和舒张压的离散值,而非ABP波的连续值,本研究期望基于卷积神经网络-长短期记忆神经网络（CNN-LSTM）利用PPG信号波重建ABP波信号,实现连续无创血压监测。方法 构建CNN-LSTM混合神经网络模型,利用重症监护医学信息集（medical information mart for intensive care,MIMIC）中的PPG与ABP波同步记录信号数据,将PPG信号波经预处理降噪、归一化、滑窗分割后输入该模型,重建与之同步对应的ABP波信号。结果 使用窗口长度312的CNN-LSTM神经网络时,重建ABP值与实际ABP值间误差最小,平均绝对误差（mean absolute error,MAE）和均方根误差（root mean square error,RMSE）分别为2.79 mmHg和4.24 mmHg,余弦相似度最大,重建ABP值与实际ABP值一致性和相关性情况良好,符合美国医疗器械促进协会（Association for the Advancement of Medical Instrumentation,AAMI）标准。结论 CNN-LSTM混合神经网络可利用PPG信号波重建ABP波信号,实现连续无创血压监测。     

一种青海湖流域消除植被光谱对土壤光谱影响的土壤有机质含量遥感估算方法

基于遥感技术手段快速测定区域尺度土壤有机质含量(SOM), 对气候、陆地生态系统和农业等领域具有重要的作用和意义。但现有的多光谱遥感影像因其波段宽度较窄, 包含的土壤有机质信息有限, 导致其估算结果的可靠性与精度较低。为此, 以青海湖流域为实证试验区, 将2016 年9 月底(此时, 青海湖流域牧草等植被停止生长, 土壤有机质积累达到全年最高)地面采集并测定的土壤有机质含量数据与同时期MODIS 黑空BRDF/Albedo 产品的宽、窄波段进行了对比与检验。发现: BRDF/Albedo 宽波段的相关性(近红外、短波波段相关系数分别为0.704 和 0.670)高于窄波段相关性(第2, 5, 6 波段的相关系数分别是0.583、0.631 和0.625), 证实了宽波段含有更加丰富、完整的土壤有机质含量信息。为了进一步提高SOM 估算的精度, 基于梯形方法构建了宽波段近红外反照率/植被覆盖度梯形特征空间,从宽波段近红外反照率(包含植被、土壤混合光谱)中成功分离出裸土反照率, 并分别构建了SOM 遥感估算模型。经验证,消除了植被对土壤光谱影响的裸土反照率模型精度(均方根误差为16.87、平均绝对百分比误差为30.0%, 希尔不等系数为0.22)高于宽波段近红外反照率模型精度(均方根误差为20.12、平均绝对百分比误差为31.0%, 希尔不等系数为0.27)。该方法简单易操作, 不仅有助于提高表层土壤有机质含量遥感估算的精度, 也可为土壤其他属性如N, P等元素含量的遥感估算提供了新思路。     

不同思维作业电信号的高准确度模式分类

本文采用独立分量(ICA)分析对不同思维作业的脑电(EEG)信号进行预处理,再用自回归(AR)参数模型提取EEG信号特征,最后利用BP网络完成对特征样本集的训练和分类。实验结果表明,所采用的方法提高了脑电思维模式作业的准确度,对两种到五种不同思维作业未经训练的数据的平均分类准确度达到79％以上,超过现有文献报道的结果。     

Using hierarchical text classification to investigate the utility of machine learning in automating online analyses of wildlife exploitation

Expanding digital data sources, including social media, online news articles and blogs, provide an opportunity to understand better the context and intensity of human-nature interactions, such as wildlife exploitation. However, online searches encompassing large taxonomic groups can generate vast datasets, which can be overwhelming to filter for relevant content without the use of automated tools. The variety of machine learning models available to researchers, and the need for manually labelled training data with an even balance of labels, can make applying these tools challenging. Here, we implement and evaluate a hierarchical text classification pipeline which brings together three binary classification tasks with increasingly specific relevancy criteria. Crucially, the hierarchical approach facilitates the filtering and structuring of a large dataset, of which relevant sources make up a small proportion. Using this pipeline, we also investigate how the accuracy with which text classifiers identify relevant and irrelevant texts is influenced by the use of different models, training datasets, and the classification task. To evaluate our methods, we collected data from Facebook, Twitter, Google and Bing search engines, with the aim of identifying sources documenting the hunting and persecution of bats (Chiroptera). Overall, the ‘state-of-the-art’ transformer-based models were able to identify relevant texts with an average accuracy of 90%, with some classifiers achieving accuracy of >95%. Whilst this demonstrates that application of more advanced models can lead to improved accuracy, comparable performance was achieved by simpler models when applied to longer documents and less ambiguous classification tasks. Hence, the benefits from using more computationally expensive models are dependent on the classification context. We also found that stratification of training data, according to the presence of key search terms, improved classification accuracy for less frequent topics within datasets, and therefore improves the applicability of classifiers to future data collection. Overall, whilst our findings reinforce the usefulness of automated tools for facilitating online analyses in conservation and ecology, they also highlight that the effectiveness and appropriateness of such tools is determined by the nature and volume of data collected, the complexity of the classification task, and the computational resources available to researchers.     

Classifying Response Correctness across Different Task Sets: A Machine Learning Approach

Erroneous behavior usually elicits a distinct pattern in neural waveforms. In particular, inspection of the concurrent recorded electroencephalograms (EEG) typically reveals a negative potential at fronto-central electrodes shortly following a response error (Ne or ERN) as well as an error-awareness-related positivity (Pe). Seemingly, the brain signal contains information about the occurrence of an error. Assuming a general error evaluation system, the question arises whether this information can be utilized in order to classify behavioral performance within or even across different cognitive tasks. In the present study, a machine learning approach was employed to investigate the outlined issue. Ne as well as Pe were extracted from the single-trial EEG signals of participants conducting a flanker and a mental rotation task and subjected to a machine learning classification scheme (via a support vector machine, SVM). Overall, individual performance in the flanker task was classified more accurately, with accuracy rates of above 85%. Most importantly, it was even feasible to classify responses across both tasks. In particular, an SVM trained on the flanker task could identify erroneous behavior with almost 70% accuracy in the EEG data recorded during the rotation task, and vice versa. Summed up, we replicate that the response-related EEG signal can be used to identify erroneous behavior within a particular task. Going beyond this, it was possible to classify response types across functionally different tasks. Therefore, the outlined methodological approach appears promising with respect to future applications.     

Fine hyperspectral classification of rice varieties based on self-attention mechanism

The accurate identification of rice varieties using rapid and nondestructive hyperspectral technology is of practical significance for rice cultivation and agricultural production. This paper proposes a convolutional neural network classification model based on a self-attention mechanism (self-attention-1D-CNN) to improve accuracy in distinguishing between crop species in fields using canopy spectral information. After experimental materials were planted in the research area, portable equipment was used to collect the canopy hyperspectral data for rice during the booting stage. Five preprocessing methods and three extraction methods were used to process the data. A comparison of the classification accuracy of different classification models showed that the self-attention-1D-CNN proposed in this study achieved the best classification with an accuracy of 99.93%. The research demonstrated the feasibility of using hyperspectral technology for the fine classification of rice varieties, and the feasibility of using the CNN model as a potential classification method for near-ground crop monitoring and classification.     

Classification methods for ongoing EEG and MEG signals

Classification algorithms help predict the qualitative properties of a subject's mental state by extracting useful information from the highly multivariate non-invasive recordings of his brain activity. In particular, applying them to Magneto-encephalography (MEG) and electro-encephalography (EEG) is a challenging and promising task with prominent practical applications to e.g. Brain Computer Interface (BCI). In this paper, we first review the principles of the major classification techniques and discuss their application to MEG and EEG data classification. Next, we investigate the behavior of classification methods using real data recorded during a MEG visuomotor experiment. In particular, we study the influence of the classification algorithm, of the quantitative functional variables used in this classifier, and of the validation method. In addition, our findings suggest that by investigating the distribution of classifier coefficients, it is possible to infer knowledge and construct functional interpretations of the underlying neural mechanisms of the performed tasks. Finally, the promising results reported here (up to 97% classification accuracy on 1-second time windows) reflect the considerable potential of MEG for the continuous classification of mental states.     

100% Classification Accuracy Considered Harmful: The Normalized Information Transfer Factor Explains the Accuracy Paradox

The most widely spread measure of performance, accuracy, suffers from a paradox: predictive models with a given level of accuracy may have greater predictive power than models with higher accuracy. Despite optimizing classification error rate, high accuracy models may fail to capture crucial information transfer in the classification task. We present evidence of this behavior by means of a combinatorial analysis where every possible contingency matrix of 2, 3 and 4 classes classifiers are depicted on the entropy triangle, a more reliable information-theoretic tool for classification assessment.Motivated by this, we develop from first principles a measure of classification performance that takes into consideration the information learned by classifiers. We are then able to obtain the entropy-modulated accuracy (EMA), a pessimistic estimate of the expected accuracy with the influence of the input distribution factored out, and the normalized information transfer factor (NIT), a measure of how efficient is the transmission of information from the input to the output set of classes.The EMA is a more natural measure of classification performance than accuracy when the heuristic to maximize is the transfer of information through the classifier instead of classification error count. The NIT factor measures the effectiveness of the learning process in classifiers and also makes it harder for them to “cheat” using techniques like specialization, while also promoting the interpretability of results. Their use is demonstrated in a mind reading task competition that aims at decoding the identity of a video stimulus based on magnetoencephalography recordings. We show how the EMA and the NIT factor reject rankings based in accuracy, choosing more meaningful and interpretable classifiers.     

Integrating Local and Global Error Statistics for Multi-Scale RBF Network Training: An Assessment on Remote Sensing Data

Background

This study discusses the theoretical underpinnings of a novel multi-scale radial basis function (MSRBF) neural network along with its application to classification and regression tasks in remote sensing. The novelty of the proposed MSRBF network relies on the integration of both local and global error statistics in the node selection process.

Methodology and Principal Findings

The method was tested on a binary classification task, detection of impervious surfaces using a Landsat satellite image, and a regression problem, simulation of waveform LiDAR data. In the classification scenario, results indicate that the MSRBF is superior to existing radial basis function and back propagation neural networks in terms of obtained classification accuracy and training-testing consistency, especially for smaller datasets. The latter is especially important as reference data acquisition is always an issue in remote sensing applications. In the regression case, MSRBF provided improved accuracy and consistency when contrasted with a multi kernel RBF network.

Conclusion and Significance

Results highlight the potential of a novel training methodology that is not restricted to a specific algorithmic type, therefore significantly advancing machine learning algorithms for classification and regression tasks. The MSRBF is expected to find numerous applications within and outside the remote sensing field.     

Individually Adapted Imagery Improves Brain-Computer Interface Performance in End-Users with Disability

Brain-computer interfaces (BCIs) translate oscillatory electroencephalogram (EEG) patterns into action. Different mental activities modulate spontaneous EEG rhythms in various ways. Non-stationarity and inherent variability of EEG signals, however, make reliable recognition of modulated EEG patterns challenging. Able-bodied individuals who use a BCI for the first time achieve - on average - binary classification performance of about 75%. Performance in users with central nervous system (CNS) tissue damage is typically lower. User training generally enhances reliability of EEG pattern generation and thus also robustness of pattern recognition. In this study, we investigated the impact of mental tasks on binary classification performance in BCI users with central nervous system (CNS) tissue damage such as persons with stroke or spinal cord injury (SCI). Motor imagery (MI), that is the kinesthetic imagination of movement (e.g. squeezing a rubber ball with the right hand), is the "gold standard" and mainly used to modulate EEG patterns. Based on our recent results in able-bodied users, we hypothesized that pair-wise combination of "brain-teaser" (e.g. mental subtraction and mental word association) and "dynamic imagery" (e.g. hand and feet MI) tasks significantly increases classification performance of induced EEG patterns in the selected end-user group. Within-day (How stable is the classification within a day?) and between-day (How well does a model trained on day one perform on unseen data of day two?) analysis of variability of mental task pair classification in nine individuals confirmed the hypothesis. We found that the use of the classical MI task pair hand vs. feed leads to significantly lower classification accuracy - in average up to 15% less - in most users with stroke or SCI. User-specific selection of task pairs was again essential to enhance performance. We expect that the gained evidence will significantly contribute to make imagery-based BCI technology become accessible to a larger population of users including individuals with special needs due to CNS damage.