基于多通道Transformer的交通量预测方法

目前,我国高速公路拥堵程度居高不下,而交通流预测作为实现智能交通系统的重要一环,若能对其实现高精度的预测,那么将能够高效地管理交通,从而缓解拥堵。针对该问题,提出了一种考虑时空关联的多通道交通流预测方法（MCST-Transformer）。首先,将Transformer结构用于不同数据的内在规律提取,然后引入空间关联模块对不同数据间的关联特征进行挖掘,最后,借助通道注意力整合优化全局信息。采用广东省高速公路数据,实现了两小时内92个收费站的高精度流量预测。结果表明：MCST-Transformer优于传统机器学习方法以及部分基于注意力机制的时间序列模型,在120 min预测跨度下,相比贝叶斯回归,MAPE降低了5.1%;对比Seq2Seq-Att以及Seq2Seq这些深度学习算法,所提方法的总体MAPE也能降低0.5%,说明通过多通道的方式能够区分不同数据的特性,进而更好地预测。     

基于集成学习的交通拥堵预测模型

The current navigation software has obvious inaccurate speed assessment when facing some serious traffic congestion, and cannot accurately predict the duration of the traffic congestion. Therefore, we propose a traffic congestion prediction model to accu- rately predict the congestion time in the face of most congestion situations through the prediction of speed. Regarding the speed pre- diction model, we select high-similarity samples based on the KNN algorithm. The prediction speed model is divided into two main models, KNN-VA and KNN-RBF, and we use an integrated learning method to fuse these two models to obtain more accurate aver- age speed prediction. Then, the congestion time can be predicted. In order to determine the congestion time, we use the RBF speed prediction method and the sampling method in a fixed area to verify. The results show that the model has high reliability for conges- tion time prediction.     

Anomaly Detection in ICS Datasets with Machine Learning Algorithms

An Intrusion Detection System (IDS) provides a front-line defense mechanism for the Industrial Control System (ICS) dedicated to keeping the process operations running continuously for 24 hours in a day and 7 days in a week. A well-known ICS is the Supervisory Control and Data Acquisition (SCADA) system. It supervises the physical process from sensor data and performs remote monitoring control and diagnostic functions in critical infrastructures. The ICS cyber threats are growing at an alarming rate on industrial automation applications. Detection techniques with machine learning algorithms on public datasets, suitable for intrusion detection of cyber-attacks in SCADA systems, as the first line of defense, have been detailed. The machine learning algorithms have been performed with labeled output for prediction classification. The activity traffic between ICS components is analyzed and packet inspection of the dataset is performed for the ICS network. The features of flow-based network traffic are extracted for behavior analysis with port-wise profiling based on the data baseline, and anomaly detection classification and prediction using machine learning algorithms are performed.     

Urban traffic congestion estimation and prediction based on floating car trajectory data

Traffic flow prediction is an important precondition to alleviate traffic congestion in large-scale urban areas. Recently, some estimation and prediction methods have been proposed to predict the traffic congestion with respect to different metrics such as accuracy, instantaneity and stability. Nevertheless, there is a lack of unified method to address the three performance aspects systematically. In this paper, we propose a novel approach to estimate and predict the urban traffic congestion using floating car trajectory data efficiently. In this method, floating cars are regarded as mobile sensors, which can probe a large scale of urban traffic flows in real time. In order to estimate the traffic congestion, we make use of a new fuzzy comprehensive evaluation method in which the weights of multi-indexes are assigned according to the traffic flows. To predict the traffic congestion, an innovative traffic flow prediction method using particle swarm optimization algorithm is responsible for calculating the traffic flow parameters. Then, a congestion state fuzzy division module is applied to convert the predicted flow parameters to citizens’ cognitive congestion state. Experimental results show that our proposed method has advantage in terms of accuracy, instantaneity and stability.     

基于集成聚类的流量分类架构

流量分类是优化网络服务质量的基础与关键.机器学习算法利用数据流统计特征分类流量,对于识别加密私有协议流量具有重要意义.然而,特征偏置和类别不平衡是基于机器学习的流量分类研究所面临的两大挑战.特征偏置是指一些数据流统计特征在提高部分应用识别准确率的同时也降低了另外一部分应用识别的准确率.类别不平衡是指机器学习流量分类器对样本数较少的应用识别的准确率较低.为解决上述问题,提出了基于集成聚类的流量分类架构（traffic classification framework based on ensemble clustering,简称TCFEC）.TCFEC由多个基于不同特征子空间聚类的基分类器和一个最优决策部件构成,能够提高流量分类的准确率.具体而言,与传统的机器学习流量分类器相比,TCFEC的平均流准确率最高提升5%,字节准确率最高提升6%.     

A holistic multi-objective optimization design procedure for ensemble member generation and selection

In the last few years, machine learning techniques have been successfully applied to solve engineering problems. However, owing to certain complexities found in real-world problems, such as class imbalance, classical learning algorithms may not reach a prescribed performance. There can be situations where a good result on different conflicting objectives is desirable, such as true positive and true negative ratios, or it is important to balance model’s complexity and prediction score. To solve such issues, the application of multi-objective optimization design procedures can be used to analyze various trade-offs and build more robust machine learning models. Thus, the creation of ensembles of predictive models using such procedures is addressed in this work. First, a set of diverse predictive models is built by employing a multi-objective evolutionary algorithm. Next, a second multi-objective optimization step selects the previous models as ensemble members, resulting on several non-dominated solutions. A final multi-criteria decision making stage is applied to rank and visualize the resulting ensembles. To analyze the proposed methodology, two different experiments are conducted for binary classification. The first case study is a famous classification problem through which the proposed procedure is illustrated. The second one is a challenging real-world problem related to water quality monitoring, where the proposed procedure is compared to four classical ensemble learning algorithms. Results on this second experiment show that the proposed technique is able to create robust ensembles that can outperform other ensemble methods. Overall, the authors conclude that the proposed methodology for ensemble generation creates competitive models for real-world engineering problems.     

基于侧信道特征的安全代理流量分类方法

安全代理被越来越多的互联网用户用于规避网络审查和访问受限资源,因此安全代理流量的分类对于网络安全和网络管理具有重要意义。为弥补深度包检测技术在过滤和识别不良信息上的不足,提高防火墙流量探测能力,提出一种安全代理流量分类方法。提取用于安全代理流量分类的侧信道特征,包括有效载荷长度序列、信号序列等,使用机器学习和深度学习算法对Shadowsocks、V2Ray、Freegate、Ultrasurf 4种被广泛使用的安全代理流量进行识别。实验结果表明,通过提取与有效载荷内容无关的侧信道特征进行分类,与MLP、LSMP等算法相比,该方法在准确率、F1值等性能方面均有提升。     

基于小波核LS—SVM的网络流量预测

网络流量预测对大规模网络管理、规划、设计具有重要意义。支持向量机方法是近年来发展起来的新型机器学习算法,用于解决高度非线性分类及回归问题。介绍了基于小波核最小二乘支持向量机的网络流量预测方法,利用小波核函数的多分辨特性提高了支持向量机的非线性建模能力。通过对实测网络流量数据的学习,对未来网络流量进行预测。实验结果表明,取得了较好的预测效果。     

Parallelizing Feature Selection

Classification is a key problem in machine learning/data mining. Algorithms for classification have the ability to predict the class of a new instance after having been trained on data representing past experience in classifying instances. However, the presence of a large number of features in training data can hurt the classification capacity of a machine learning algorithm. The Feature Selection problem involves discovering a subset of features such that a classifier built only with this subset would attain predictive accuracy no worse than a classifier built from the entire set of features. Several algorithms have been proposed to solve this problem. In this paper we discuss how parallelism can be used to improve the performance of feature selection algorithms. In particular, we present, discuss and evaluate a coarse-grained parallel version of the feature selection algorithm FortalFS. This algorithm performs well compared with other solutions and it has certain characteristics that makes it a good candidate for parallelization. Our parallel design is based on the master--slave design pattern. Promising results show that this approach is able to achieve near optimum speedups in the context of Amdahl's Law.     

基于多类别支持向量机的城市交通网络模式分类研究

在网络层次上进行区域交通信号控制、交通分配和路径诱导是缓解交通堵塞的有效途径之一。为进一步提高城市交通网络分类检测的准确性,将支持向量机（Support Vector Machine）应用于交通事件的模式分类研究。通过提出一种基于多类别支持向量机的交通模式分类方法,设计了适合该检测系统的网络结构。仿真结果表明：相对于其他算法,城市交通网络的状态可分为数量有限且不同类型的模式,并且这些模式不断重复出现,当系统识别出网络处于某种模式时,就可参照事先确定的优化参数及策略进行交通控制和诱导,以缓解交通拥塞,提高交通系统的运行效率。该网络结构对于小样本数据具有检测率高、误报率低的优点,完全适用于城市交通的模式分类,同时也存在不足之处,指出了今后进一步研究的方向。     

Predicting protein subchloroplast locations: the 10th anniversary

Chloroplast is a type of subcellular organelle in green plants and algae. It is the main subcellular organelle for conducting photosynthetic process. The proteins, which localize within the chloroplast, are responsible for the photosynthetic process at molecular level. The chloroplast can be further divided into several compartments. Proteins in different compartments are related to different steps in the photosynthetic process. Since the molecular function of a protein is highly correlated to the exact cellular localization, pinpointing the subchloroplast location of a chloroplast protein is an important step towards the understanding of its role in the photosynthetic process. Experimental process for determining protein subchloroplast location is always costly and time consuming. Therefore, computational approaches were developed to predict the protein subchloroplast locations from the primary sequences. Over the last decades, more than a dozen studies have tried to predict protein subchloroplast locations with machine learning methods. Various sequence features and various machine learning algorithms have been introduced in this research topic. In this review, we collected the comprehensive information of all existing studies regarding the prediction of protein subchloroplast locations. We compare these studies in the aspects of benchmarking datasets, sequence features, machine learning algorithms, predictive performances, and the implementation availability. We summarized the progress and current status in this special research topic. We also try to figure out the most possible future works in predicting protein subchloroplast locations. We hope this review not only list all existing works, but also serve the readers as a useful resource for quickly grasping the big picture of this research topic.We also hope this review work can be a starting point of future methodology studies regarding the prediction of protein subchloroplast locations.     

A combinational incremental ensemble of classifiers as a technique for predicting students’ performance in distance education

The ability to predict a student’s performance could be useful in a great number of different ways associated with university-level distance learning. Students’ marks in a few written assignments can constitute the training set for a supervised machine learning algorithm. Along with the explosive increase of data and information, incremental learning ability has become more and more important for machine learning approaches. The online algorithms try to forget irrelevant information instead of synthesizing all available information (as opposed to classic batch learning algorithms). Nowadays, combining classifiers is proposed as a new direction for the improvement of the classification accuracy. However, most ensemble algorithms operate in batch mode. Therefore a better proposal is an online ensemble of classifiers that combines an incremental version of Naive Bayes, the 1-NN and the WINNOW algorithms using the voting methodology. Among other significant conclusions it was found that the proposed algorithm is the most appropriate to be used for the construction of a software support tool.     

Mobile Apps identification based on network flows

Network operators and mobile carriers are facing serious security challenges caused by an increasing number of services provided by smartphone Apps. For example, Android OS has more than 1 million Apps in stores. Hence, network administrators tend to adopt strict policies to secure their infrastructure. The aim of this study is to propose an efficient framework that has a classification component based on traffic analysis of Android Apps. The framework differs from other proposed studies by focusing on identifying Apps traffic from a network perspective without introducing any overhead on subscribers smartphones. Additionally, it involves a technique for pre-processing network flows generated by Apps to acquire a set of features that are used to build an identification model using machine learning algorithms. The classification model is built using classification ensembles. A group of chosen users contribute in training the classification model, which learns the normal behavior of selected Apps. Eventually, the model should be able to detect abnormal behavior of similar Apps across the network. A 93.78% classification accuracy is achieved with a low false positive rate under 0.5%. In addition, the framework is able to detect abnormal flows of unknown classes by implementing an outlier detection mechanism and reported a 94% accuracy.     

Decay-weighted extreme learning machine for balance and optimization learning

The original extreme learning machine (ELM) was designed for the balanced data, and it balanced misclassification cost of every sample to get the solution. Weighted extreme learning machine assumed that the balance can be achieved through the equality of misclassification costs. This paper improves previous weighted ELM with decay-weight matrix setting for balance and optimization learning. The decay-weight matrix is based on the sample number of each class, but the weight sum values of each class are not necessarily equal. When the number of samples is reduced, the weight sum is also reduced. By adjusting the decaying velocity, classifier could achieve more appropriate boundary position. From the experimental results, the decay-weighted ELM obtains the better effects in solving the imbalance classification tasks, particularly in multiclass tasks. This method was successfully applied to build the prediction model in the urban traffic congestion prediction system.     

基于深度生成对抗网络的恶意TLS流量识别

恶意加密流量识别公开数据集中存在的类不平衡问题,严重影响着恶意流量预测的性能。本文提出使用深度生成对抗网络DGAN中的生成器和鉴别器,模拟真实数据集生成并扩展小样本数据,形成平衡数据集。此外,针对传统机器学习方法依赖人工特征提取导致分类准确度下降等问题,提出一种基于双向门控循环单元BiGRU与注意力机制相融合的恶意流量识别模型,由深度学习算法自动获取数据集不同时序的重要特征向量,进行恶意流量得识别。实验表明,与常用恶意流量识别算法相比,该模型在精度、召回率、F1等指标上都有较好的提升,能有效实现恶意加密流量的识别。     

基于机器学习的网络流量分类研究进展

机器学习方法不依赖匹配协议端口或解析协议内容,而是利用网络流的各种统计特征识别网络应用,近年来得到了广泛关注和快速发展.本文总结了基于机器学习的网络流量分类方法自2004年来的研究进展,并且按有监督、无监督与半监督的区别进行分类、分析与比较.重点讨论了基于机器学习的网络流量分类研究的挑战与方向,即解决样本标注瓶颈、样本分布不平衡与动态变化、实时与连续分类以及分类算法可扩展性等核心问题.     

Exploring spatial–temporal relations via deep convolutional neural networks for traffic flow prediction with incomplete data

Traffic flow prediction is a fundamental component in intelligent transportation systems. Various computational methods have been applied in this field, among which machine learning based methods are believed to be promising and scalable for big data. In general, most of machine learning based methods encounter three fundamental issues: feature representation of traffic patterns, learning from single location or network, and data quality. In order to address these three issues, in this work we present a deep architecture for traffic flow prediction that learns deep hierarchical feature representation with spatio-temporal relations over the traffic network. Furthermore, we design an ensemble learning strategy via random subspace learning to make the model be able to tolerate incomplete data. Accordingly the contributions of this work are summarized as the three points. First, we transform the time series analysis problem into the task of image-like analysis. Benefitting from the image-like data form, we can jointly explore spatio-temporal relations simultaneously by the two-dimension convolution operator. In addition, the proposed model can tolerate the incomplete data, which is very common in traffic application field. Finally, we propose an improved random search based on uniform design in order to optimize hyper-parameters for deep Convolutional Neural Networks (deep CNN). A large range of experiments with various traffic conditions have been performed on the traffic data originated from the California Freeway Performance Measurement System (PeMS). The experimental results corroborate the effectiveness of the proposed approach compared with the state of the art.     

Empirical analysis of search based algorithms to identify change prone classes of open source software

There are numerous reasons leading to change in software such as changing requirements, changing technology, increasing customer demands, fixing of defects etc. Thus, identifying and analyzing the change-prone classes of the software during software evolution is gaining wide importance in the field of software engineering. This would help software developers to judiciously allocate the resources used for testing and maintenance. Software metrics can be used for constructing various classification models which can be used for timely identification of change prone classes. Search based algorithms which form a subset of machine learning algorithms can be utilized for constructing prediction models to identify change prone classes of software. Search based algorithms use a fitness function to find the best optimal solution among all the possible solutions. In this work, we analyze the effectiveness of hybridized search based algorithms for change prediction. In other words, the aim of this work is to find whether search based algorithms are capable for accurate model construction to predict change prone classes. We have also constructed models using machine learning techniques and compared the performance of these models with the models constructed using Search Based Algorithms. The validation is carried out on two open source Apache projects, Rave and Commons Math. The results prove the effectiveness of hybridized search based algorithms in predicting change prone classes of software. Thus, they can be utilized by the software developers to produce an efficient and better developed software.     

Combining instance-based learning and logistic regression for multilabel classification

Multilabel classification is an extension of conventional classification in which a single instance can be associated with multiple labels. Recent research has shown that, just like for conventional classification, instance-based learning algorithms relying on the nearest neighbor estimation principle can be used quite successfully in this context. However, since hitherto existing algorithms do not take correlations and interdependencies between labels into account, their potential has not yet been fully exploited. In this paper, we propose a new approach to multilabel classification, which is based on a framework that unifies instance-based learning and logistic regression, comprising both methods as special cases. This approach allows one to capture interdependencies between labels and, moreover, to combine model-based and similarity-based inference for multilabel classification. As will be shown by experimental studies, our approach is able to improve predictive accuracy in terms of several evaluation criteria for multilabel prediction.