Search

The paper proposes a robot zero calibration device and method based on spatial parameter clustering identification to overcome the shortcomings of traditional shaft pin and laser tracker calibration. The dial indicator and spatial analytic geometry are used to construct the zero calibration equation， which analyzes the robot’s theoretical post and posture values to obtain the zero-position training parameters. The clustering module identifies multiple sets of zero-position training parameters， and the optimal zero-position parameters of each joint can be solved. The experimental results show that spatial parameter clustering can improve the absolute accuracy by 72.9%， compared to the most commonly used shaft pin calibration method. Compared with the laser tracker calibration method with the highest accuracy， the time efficiency is improved by 87.5%. This method considers the calibration accuracy and efficiency and significantly improves the adaptability to the calibration environment. [ABSTRACT FROM AUTHOR]

Collecting wearable motion sensor signals and using transfer learning to overcome the inconsistency of data distribution to identify the daily behavior of the human body are very popular technologies. Using wearable sensors to collect signals will result in generating noise samples that affect the transfer effect. Traditional algorithms lack the processing of these samples. To solve this problem， the traditional algorithm was improved by introducing a sample screening algorithm based on Mahalanobis distance， and a transfer learning algorithm T-WMD was proposed that can be used for human activity recognition. And compared with other five algorithms on two public human activity recognition data sets， the results show that the algorithm proposed in this paper can effectively improve the effect of transfer learning. [ABSTRACT FROM AUTHOR]

Since it is difficult to online measure the underflow concentration of the thickener in the thickening-dehydration process, a broad learning system(BLS) based soft sensor modeling method is proposed in this paper. The method has high precision and strong generalization capability. First, several pressure sensors are installed inside the thickener, and the historical dataset under normal operating conditions is established. Then, the soft sensor model is trained by employing the BLS method to online predict the underflow concentration of the thickener. Finally, the efficiency of the proposed method is verified by simulation experiments. Compared with other traditional machine learning methods, the BLS method has higher prediction accuracy. [ABSTRACT FROM AUTHOR]

The neural network with piecewise linear activation function ( PLNN) is extensively applied in machine learning. This paper gives a measure of the expressive power of PLNN model, i. e., the number of linear regions, with the mathematical representation of linear regions presented. The relationship between linear regions is analyzed and the number of combined linear regions is calculated. A method for calculating the number of linear regions based on Z curve is developed. A case is given to calculate the number of linear regions of different network structures and the number of linear regions after merging, and the correlation between the number of linear regions and the accuracy of different network structures is analyzed. The results show that the number of linear regions can reflect the expressive power of PLNN model, which has great research significance for selecting network hyperparameters and explaining model boundaries. [ABSTRACT FROM AUTHOR]

The existing spam website detection methods are mainly aimed at self-built spam websites， and not suitable for injected spam websites because of the low efficiency of link detection. This paper proposes a new detection method， in which a detection framework is based on multi-dimensional features of webpage structure and text. The framework divides the webpage into blocks. Then content features are extracted by calculating odd ratio and structural features based on tags， attribute keys and attribute values are extracted by using the one-hot rate. The detection model is generated by proper machine learning and used to detect spam links. The detection accuracy of this framework is increased by up to 13%， compared with the algorithms based on content detection and on blacklist matching. [ABSTRACT FROM AUTHOR]

Interactive data exploration(IDE) is a key technique in a diverse set of discovery-based applications, which focuses on interaction, exploration and discovery and has a wide range of applications in many scenes and areas. The feature adaptive technology of interactive data exploration was studied in this paper with the background of massive academic literature data exploration. Firstly, a framework of interactive data exploration was presented, namely FA-IDE(feature-adaptive interactive data exploration) framework, which can dynamically adjust the subset of features during each iteration to meet the needs of the user’ s interest diversity. Secondly, according to this framework, the evaluation criteria of the balance of feature subsets(BFS) were proposed in the stage of exploration and a sequence forward feature selection algorithm based on BFS was also given. Besides, for the phases of related sample discovery, a division level establishment method was proposed. According to the decision tree model which can divide the user interest area, a strategy of result set sorting based on similarity was proposed. The results of experiments showthat the accuracy and efficiency of the proposed method have been effectively improved. [ABSTRACT FROM AUTHOR]