Classification and regression models of audio and vibration signals for machine state monitoring in precision machining systems

We present a data-driven method for monitoring machine status in manufacturing processes. Audio and vibration data from precision machining are used for inference in two operating scenarios: (a) variable machine health states (anomaly detection); and (b) settings of machine operation (state estimation). Audio and vibration signals are first processed through Fast Fourier Transform and Principal Component Analysis to extract transformed and informative features. These features are then used in the training of classification and regression models for machine state monitoring. Specifically, three classifiers (K-nearest neighbors, convolutional neural networks and support vector machines) and two regressors (support vector regression and neural network regression) were explored, in terms of their accuracy in machine state prediction. It is shown that the audio and vibration signals are sufficiently rich in information about the machine that 100% state classification accuracy could be accomplished. Data fusion was also explored, showing overall superior accuracy of data-driven regression models.     

Applying data mining to learn system dynamics in a biological model

Data mining consists of a set of powerful methods that have been successfully applied to many different application domains, including business, engineering, and bioinformatics. In this paper, we propose an innovative approach that uses genetic algorithms to mine a set of temporal behavior data output by a biological system in order to determine the kinetic parameters of the system. Analyzing the behavior of a biological network is a complicated task. In our approach, the machine learning method is integrated with the framework of system dynamics so that its findings are expressed in a form of system dynamics model. An application of the method to the cell division cycle model has shown that the method can discover approximate parametric values of the system and reproduce the input behavior.     

Distributed patching for mobile wireless sensor networks

Mobile wireless sensor networks (MWSNs), the continuous movement of sensor nodes, may cause complete disconnection of the network or at best a part of it. The design of such networks should guarantee that all sensor nodes at all times have a path to the sink node(s). Prolonging the lifetimes of the MWSNs is a crucial design issue but should not be at the expense of other essential functions such as connectivity. In this paper, we propose an energy efficient distributed framework for connectivity maintenance of MWSNs. In the proposed framework, sensor nodes of the MWSNs schedule and control their radio frequency (RF) modules based on dynamic coordinated reconstruction mechanism, some senor nodes may lose their connectivity with the network due to mobility, depletion of energy and/or incurred faults. Two protocols are developed to patch up the disconnection of the MWSNs. The first protocol reestablishes the network connection consuming far less energy than state-of-the-art alternatives. However, it is capable of reestablishing the connection when no more than 20% of the MWSN backbone nodes lose connection concurrently. The second protocol extends this limit to operate when up to 35% of the MWSN backbone nodes lose connection simultaneously.     

一种谱分解降维的模糊有监督局部保持投影策略

提出一种谱分解降维的模糊有监督局部保持投影策略。首先针对监督局部保持投影SLPP存在过学习和不能较好地保持图像空间的差异信息等问题,通过最小化局部离散度和最大化差异离散度准则提取投影方向,找到一种线性鉴别分析的等价形式。其次,通过采用模糊k近邻(FKNN)方法得到相应的样本分布隶属度信息,同时考虑到离群样本对整个分类结果的不利影响,提出一种模糊化方法,根据样本的隶属度对样本分布矩阵重定义所做的贡献,将每个样本的隶属度融入到SLPP特征抽取的过程中,从而得到完整有效的模糊样本特征向量集,有效解决了小样本问题的特征抽取问题。第三,提出一种谱分解的矩阵分析方法,在SLPP投影准则下,对散布矩阵实现降维。在ORL和NUST603人脸库上的实验结果验证了该方法的有效性。     

流媒体直播系统的多路并发流自适应传输控制

为了解决流媒体直播在Internet中传输带宽不足、用户接入多样性等问题,本文探讨了流媒体在IP网络中一般的传输控制技术,提出了一种基于联播的流媒体实时传输控制策略以适应网络带宽的动态变化。首先通过组播方式以及RTP/RTCP协议实现流媒体的实时传输;之后对自适应传输的算法进行了详细的分析和设计,并采用多线程技术实现了网络状况监测和实时码流切换控制;最后通过研究及实验数据分析对比,证明该方案可提高多路并发流传输效率的有效性,并且可以给客户端用户带来较好的用户体验。     

Fuzzy arithmetic based reliability allocation approach during early design and development

During early design and development stages, every engineering system has to meet its specific reliability goals. The target reliability of the system is achieved through application of an effective reliability apportionment technique to its subsystems. There are various traditional methods exist to perform the reliability allocation based on engineering factors that are assessed in a subjective manner. The conventional reliability allocation approach requires the assessment of factors like complexity, cost, and maintenance. This may not be realistic in real applications if they are assessed in a crisp manner by the domain experts of their varied expertise and background.In this paper, we treat allocation factors as fuzzy numbers, which are evaluated in fuzzy linguistic terms. As a result, fuzzy proportionality factor scales are proposed for the subsystems. In order to accomplish fuzzy division to evaluate the fuzzy proportionality factor, an approximation method based on linear programming for trapezoidal fuzzy numbers is also proposed in this paper. For the evaluation of weighting factors from fuzzy proportionality factors, centroid method of defuzzification is being employed. The allocated reliability of each subsystem is computed with the help of weighting factor thereafter. An example is provided to illustrate the potential application of the proposed fuzzy based reliability allocation approach.     

Leveraging clustering approaches to solve the gray-sheep users problem in recommender systems

Recommender systems apply data mining and machine learning techniques for filtering unseen information and can predict whether a user would like a given item. This paper focuses on gray-sheep users problem responsible for the increased error rate in collaborative filtering based recommender systems. This paper makes the following contributions: we show that (1) the presence of gray-sheep users can affect the performance – accuracy and coverage – of the collaborative filtering based algorithms, depending on the data sparsity and distribution; (2) gray-sheep users can be identified using clustering algorithms in offline fashion, where the similarity threshold to isolate these users from the rest of community can be found empirically. We propose various improved centroid selection approaches and distance measures for the K-means clustering algorithm; (3) content-based profile of gray-sheep users can be used for making accurate recommendations. We offer a hybrid recommendation algorithm to make reliable recommendations for gray-sheep users. To the best of our knowledge, this is the first attempt to propose a formal solution for gray-sheep users problem. By extensive experimental results on two different datasets (MovieLens and community of movie fans in the FilmTrust website), we showed that the proposed approach reduces the recommendation error rate for the gray-sheep users while maintaining reasonable computational performance.     

A weighted multivariate Fuzzy C-Means method in interval-valued scientific production data

Clustering is the process of organizing objects into groups whose members are similar in some way. Most of the clustering methods involve numeric data only. However, this representation may not be adequate to model complex information which may be: histogram, distributions, intervals. To deal with these types of data, Symbolic Data Analysis (SDA) was developed. In multivariate data analysis, it is common some variables be more or less relevant than others and less relevant variables can mask the cluster structure. This work proposes a clustering method based on fuzzy approach that produces weighted multivariate memberships for interval-valued data. These memberships can change at each iteration of the algorithm and they are different from one variable to another and from one cluster to another. Furthermore, there is a different relevance weight associated to each variable that may also be different from one cluster to another. The advantage of this method is that it is robust to ambiguous cluster membership assignment since weights represent how important the different variables are to the clusters. Experiments are performed with synthetic data sets to compare the performance of the proposed method against other methods already established by the clustering literature. Also, an application with interval-valued scientific production data is presented in this work. Clustering quality results have shown that the proposed method offers higher accuracy when variables have different variabilities.     

Estimation and forecasting with logarithmic autoregressive conditional duration models: A comparative study with an application

This paper presents a semi-parametric method of parameter estimation for the class of logarithmic ACD (Log-ACD) models using the theory of estimating functions (EF). A number of theoretical results related to the corresponding EF estimators are derived. A simulation study is conducted to compare the performance of the proposed EF estimates with corresponding ML (maximum likelihood) and QML (quasi maximum likelihood) estimates. It is argued that the EF estimates are relatively easier to evaluate and have sampling properties comparable with those of ML and QML methods. Furthermore, the suggested EF estimates can be obtained without any knowledge of the distribution of errors is known. We apply all these suggested methodology for a real financial duration dataset. Our results show that Log-ACD (1, 1) fits the data well giving relatively smaller variation in forecast errors than in Linear ACD (1, 1) regardless of the method of estimation. In addition, the Diebold–Mariano (DM) and superior predictive ability (SPA) tests have been applied to confirm the performance of the suggested methodology. It is shown that the new method is slightly better than traditional methods in practice in terms of computation; however, there is no significant difference in forecasting ability for all models and methods.