Arabic sign language intelligent translator

ABSTRACT

Arabic sign language (ArSL) is method of communication between deaf communities in Arab countries; therefore, the development of systemsthat can recognize the gestures provides a means for the Deaf to easily integrate into society. In this research we implemented a computational structurefor an intelligent interpreter that automatically recognizes the isolated dynamic gestures. The proposed system recognizes and translates gesturesperformed with one or both hands. It comprises five subsystems, building dataset, video processing, feature extraction, mapping between ArSL and Arabictext, and text generation. To apply the system, 100-signs of ArSL was used, which was applied on 1500 video files. It's were divided into five classes:alphabet, numbers, "prepositions, pronouns and question words", Arabic life expressions, and "nouns and verbs". The evaluation indicated that thesystem automatically recognizes and translates isolated dynamic ArSL gestures by highly accurate manner. The results showed that the system accuracy is 95.8%.     

Food recognition using neural network classifier and multiple hypotheses image segmentation

ABSTRACT

This paper proposes the multiple-hypotheses image segmentation and feed-forward neural network classifier for food recognition to improve the performance. Initially, the food or meal image is given as input. Then, the segmentation is applied to identify the regions, where a particular food item is located using salient region detection, multi-scale segmentation, and fast rejection. Then, the features of every food item are extracted by the global feature and local feature extraction. After the features are obtained, the classification is performed for each segmented region using a feed-forward neural network model. Finally, the calorie value is computed with the aid of (i) food volume and (ii) calorie and nutrition measure based on mass value. The experimental results and performance evaluation are validated. The outcome of the proposed method attains 0.947 for Macro Average Accuracy (MAA) and 0.959 for Standard Accuracy (SA), which provides better classification performance.     

MOOC Learner’s Final Grade Prediction Based on an Improved Random Forests Method

Massive Open Online Course (MOOC) has become a popular way of online learning used across the world by millions of people. Meanwhile, a vast amount of information has been collected from the MOOC learners and institutions. Based on the educational data, a lot of researches have been investigated for the prediction of the MOOC learner’s final grade. However, there are still two problems in this research field. The first problem is how to select the most proper features to improve the prediction accuracy, and the second problem is how to use or modify the data mining algorithms for a better analysis of the MOOC data. In order to solve these two problems, an improved random forests method is proposed in this paper. First, a hybrid indicator is defined to measure the importance of the features, and a rule is further established for the feature selection; then, a Clustering-Synthetic Minority Over-sampling Technique (SMOTE) is embedded into the traditional random forests algorithm to solve the class imbalance problem. In experiment part, we verify the performance of the proposed method by using the Canvas Network Person-Course (CNPC) dataset. Furthermore, four well-known prediction methods have been applied for comparison, where the superiority of our method has been proved.     

基于特征贡献率的机械故障分类方法

为提高往复压缩机、航空发动机等复杂机械故障分类的准确率，依据特征参数对不同故障的敏感度存在差异的特性，提出一种狄利克雷过程混合模型(Dirichlet process mixture model,简称DPMM)与贝叶斯推断贡献(Bayesian inference contribution,简称BIC)相结合的分析方法。采用DPMM方法自学习机械振动信号高维特征的统计分布模型，并依据BIC理论计算得到各特征参数对模型的贡献率，通过对比观测数据与各类故障数据特征贡献率间的差异实现故障分类。试验结果表明，该方法的平均分类准确率比基于高斯混合模型(Gaussian mixture model,简称GMM)的故障诊断方法的平均分类准确率提高19.29%，比基于Relief算法的故障诊断方法的平均分类准确率提高32.71%，且该方法的时效性高，泛化性能强，能够更有效地进行复杂机械故障分类。     

Automatic recommendation to appropriate log levels

A log statement is one of the key tactics for a developer to record and monitor important run-time behaviors of our system in a development phase and a maintenance phase. It composes of a message for stating log contents, and a log level (eg, debug or warn) to denote the severity of a message and controlling its visibility at run time. In spite of its usefulness, a developer does not tend to deeply consider which log level is appropriate in writing source code, which causes the system to be unmaintainable. To address this issue, this paper proposes an automatic approach to validating the appropriateness of the log level in consideration of the semantic and syntactic features and recommending a proper alternative log level. We first build the semantic feature vector to quantify the semantic similarity among application log messages using the word vector space, and the syntactic feature vector to capture the application context that surrounds the log statement. Based on the feature vectors and machine learning techniques, the log level is automatically validated, and an alternative log level is recommended if the log level is invalid. For the evaluation, we collected 22 open-source projects from three application domains, and obtained the 77% of precision and 75% of recall in validating the log levels. Also, our approach showed 6% higher accuracy than that of the developer group who has 7 to 8 years of work experience, and 72% of the developers accepted our recommendation.     

YOLSO: You Only Look Small Object

Small object detection is challenging and far from satisfactory. Most general object detectors suffer from two critical issues with small objects: (1) Feature extractor based on classification network cannot express the characteristics of small objects reasonably due to insufficient appearance information of targets and a large amount of background interference around them. (2) The detector requires a much higher location accuracy for small objects than for general objects. This paper proposes an effective and efficient small object detector YOLSO to address the above problems. For feature representation, we analyze the drawbacks in previous backbones and present a Half-Space Shortcut(HSSC) module to build a background-aware backbone. Furthermore, a coarse-to-fine Feature Pyramid Enhancement(FPE) module is introduced for layer-wise aggregation at a granular level to enhance the semantic discriminability. For loss function, we propose an exponential L1 loss to promote the convergence of regression, and a focal IOU loss to focus on prime samples with high classification confidence and high IOU. Both of them significantly improves the location accuracy of small objects. The proposed YOLSO sets state-of-the-art results on two typical small object datasets, MOCOD and VeDAI, at a speed of over 200 FPS. In the meantime, it also outperforms the baseline YOLOv3 by a wide margin on the common COCO dataset.     

Feature selection methods for root-cause analysis among top-level product attributes

Manufacturing companies not only strive to deliver flawless products but also monitor product failures in the field to identify potential quality issues. When product failures occur, quality engineers must identify the root cause to improve any affected product and process. This root-cause analysis can be supported by feature selection methods that identify relevant product attributes, such as manufacturing dates with an increased number of product failures. In this paper, we present different methods for feature selection and evaluate their ability to identify relevant product attributes in a root-cause analysis. First, we compile a list of feature selection methods. Then, we summarize the properties of product attributes in warranty case data and discuss these properties regarding the challenges they pose for machine learning algorithms. Next, we simulate datasets of warranty cases, which emulate these product properties. Finally, we compare the feature selection methods based on these simulated datasets. In the end, the univariate filter information gain is determined to be a suitable method for a wide range of applications. The comparison based on simulated data provides a more general result than other publications, which only focus on a single use case. Due to the generic nature of the simulated datasets, the results can be applied to various root-cause analysis processes in different quality management applications and provide a guideline for readers who wish to explore machine learning methods for their analysis of quality data.     

A New Optimized Wrapper Gene Selection Method for Breast Cancer Prediction

Machine-learning algorithms have been widely used in breast cancer diagnosis to help pathologists and physicians in the decision-making process. However, the high dimensionality of genetic data makes the classification process a challenging task. In this paper, we propose a new optimized wrapper gene selection method that is based on a nature-inspired algorithm (simulated annealing (SA)), which will help select the most informative genes for breast cancer prediction. These optimal genes will then be used to train the classifier to improve its accuracy and efficiency. Three supervised machine-learning algorithms, namely, the support vector machine, the decision tree, and the random forest were used to create the classifier models that will help to predict breast cancer. Two different experiments were conducted using three datasets: Gene expression (GE), deoxyribonucleic acid (DNA) methylation, and a combination of the two. Six measures were used to evaluate the performance of the proposed algorithm, which include the following: Accuracy, precision, recall, specificity, area under the curve (AUC), and execution time. The effectiveness of the proposed classifiers was evaluated through comprehensive experiments. The results demonstrated that our approach outperformed the conventional classifiers as expected in terms of accuracy and execution time. High accuracy values of 99.77%, 99.45%, and 99.45% have been achieved by SA-SVM for GE, DNA methylation, and the combined datasets, respectively. The execution time of the proposed approach was significantly reduced, in comparison to that of the traditional classifiers and the best execution time has been reached by SA-SVM, which was 0.02, 0.03, and 0.02 on GE, DNA methylation, and the combined datasets respectively. In regard to precision and specificity, SA-RF obtained the best result of 100 on GE dataset. While SA-SVM attained the best recall result of 100 on GE dataset.     

空间曲线的特征识别与高质量非均匀采样

为避免传统均匀采样方法因忽视曲线重要特征而生成不理想的采样结果,获得给定数量且由特征点和辅助点组成的采样点序列,提出基于特征识别的高质量空间曲线非均匀采样方法.首先使用抛物线插值法得到曲线上所有曲率极大值点和挠率极大值点的近似位置,经筛选后产生特征点,以更好地抓住空间曲线的轮廓特征.然后定义基于弧长、曲率和挠率加权组合的特征函数,并以此自适应地选取曲线上的辅助点.与3种主流采样方法比较的实验结果表明,该方法能够获得更高质量的采样结果且具有更好的实用性,从而进一步改善空间曲线的B样条拟合效果.     

基于SVD和SSA-VMD降噪的轴承故障特征提取

针对强噪声背景下轴承故障特征提取困难的问题，提出一种基于奇异值分解和参数优化变分模态分解联合降噪的轴承故障特征提取方法(SSVMD)：首先，对原始信号进行奇异值分解(Singular Value Decomposition，SVD)处理，运用奇异值差分谱法选取有效奇异值并将原始信号重构得到初步降噪信号；其次，为防止故障信息丢失，将残余信号进行麻雀算法(Sparrow Search Algorithm，SSA)优化的变分模态分解(Variational Mode Decomposition，VMD）算法处理，得到最佳的模态个数K和惩罚参数α，选取峭度值最大、包络熵最小的IMF分量与初步降噪信号叠加得到最终降噪信号，并对信号进行包络分析；最后，通过仿真和试验数据分析得出，该方法能在信噪比很低的情况下降低噪声含量并提取轴承故障特征，为设备的状态监测和故障诊断提供理论依据。