A Large-Scale k -Nearest Neighbor Classification Algorithm Based on Neighbor Relationship Preservation

Owing to the absence of hypotheses of the underlying distributions of the data and the strong generation ability, the k -nearest neighbor (kNN) classification algorithm is widely used to face recognition, text classification, emotional analysis, and other fields. However, kNN needs to compute the similarity between the unlabeled instance and all the training instances during the prediction process; it is difficult to deal with large-scale data. To overcome this difficulty, an increasing number of acceleration algorithms based on data partition are proposed. However, they lack theoretical analysis about the effect of data partition on classification performance. This paper has made a theoretical analysis of the effect using empirical risk minimization and proposed a large-scale k -nearest neighbor classification algorithm based on neighbor relationship preservation. The process of searching the nearest neighbors is converted to a constrained optimization problem. Then, it gives the estimation of the difference on the objective function value under the optimal solution with data partition and without data partition. According to the obtained estimation, minimizing the similarity of the instances in the different divided subsets can largely reduce the effect of data partition. The minibatch k -means clustering algorithm is chosen to perform data partition for its effectiveness and efficiency. Finally, the nearest neighbors of the test instance are continuously searched from the set generated by successively merging the candidate subsets until they do not change anymore, where the candidate subsets are selected based on the similarity between the test instance and cluster centers. Experiment results on public datasets show that the proposed algorithm can largely keep the same nearest neighbors and no significant difference in classification accuracy as the original kNN classification algorithm and better results than two state-of-the-art algorithms.

Dynamic Facial Expression Recognition Using Sparse Reserved Projection Algorithm for Low Illumination Images

In this paper, a novel approach for facial expression recognition based on sparse retained projection is proposed. The locality preserving projection (LPP) algorithm is used to reduce the dimension of face image data that ensures the local near-neighbor relationship of face images. The sparse representation method is used to solve the partial occlusion of human face and the problem of light imbalance. Through sparse reconstruction, the sparse reconstruction information of expression is retained as well as the local neighborhood information of expression, which can extract more effective and judgmental internal features from the original expression data, and the obtained projection is relatively stable. The recognition results based on CK + expression database show that this method can effectively improve the facial expression recognition rate.

Spatial co-location pattern mining based on the improved density peak clustering and the fuzzy neighbor relationship

<abstract> <p>Spatial co-location pattern mining discovers the subsets of spatial features frequently observed together in nearby geographic space. To reduce time and space consumption in checking the clique relationship of row instances of the traditional co-location pattern mining methods, the existing work adopted density peak clustering to materialize the neighbor relationship between instances instead of judging the neighbor relationship by a specific distance threshold. This approach had two drawbacks: first, there was no consideration in the fuzziness of the distance between the center and other instances when calculating the local density; second, forcing an instance to be divided into each cluster resulted in a lack of accuracy in fuzzy participation index calculations. To solve the above problems, three improvement strategies are proposed for the density peak clustering in the co-location pattern mining in this paper. Then a new prevalence measurement of co-location pattern is put forward. Next, we design the spatial co-location pattern mining algorithm based on the improved density peak clustering and the fuzzy neighbor relationship. Many experiments are executed on the synthetic and real datasets. The experimental results show that, compared to the existing method, the proposed algorithm is more effective, and can significantly save the time and space complexity in the phase of generating prevalent co-location patterns.</p> </abstract>

Sistem Identifikasi Biometrik Ekpresi Wajah Menggunakan Metode Transformasi Hough

The movement of the eye ball affects the condition of the pupil to dilate to become smaller or vice versa, it indicates a person's mood changes very quickly. The image of the eyeball is not necessarily in accordance with the condition of a person's heart, so it is necessary to analyze the movement of the pupil of the eye. Facial expression using the Hough Transformation Method focuses on the movement of the pupil of the eye. The Hough transform works by looking for the neighbor relationship between pixels using straight line equations to detect lines and circular equations to detect circles. Hough line transform is a technique most commonly used to detect curved objects such as lines, circles, ellipses and parabolas. The detection accuracy of the pupil is influenced by the accuracy of the extraction of the edges of the eye. If the outer circle identification is not detected, Hough Transform will be identified. The segmentation step carried out can identify the pupil circle region with a detection success of 80-85%.

Analysis of Biological Screening Compounds with Single- or Multi-Target Activity via Diagnostic Machine Learning

Predicting compounds with single- and multi-target activity and exploring origins of compound specificity and promiscuity is of high interest for chemical biology and drug discovery. We present a large-scale analysis of compound promiscuity including two major components. First, high-confidence datasets of compounds with multi- and corresponding single-target activity were extracted from biological screening data. Positive and negative assay results were taken into account and data completeness was ensured. Second, these datasets were investigated using diagnostic machine learning to systematically distinguish between compounds with multi- and single-target activity. Models built on the basis of chemical structure consistently produced meaningful predictions. These findings provided evidence for the presence of structural features differentiating promiscuous and non-promiscuous compounds. Machine learning under varying conditions using modified datasets revealed a strong influence of nearest neighbor relationship on the predictions. Many multi-target compounds were found to be more similar to other multi-target compounds than single-target compounds and vice versa, which resulted in consistently accurate predictions. The results of our study confirm the presence of structural relationships that differentiate promiscuous and non-promiscuous compounds.

The geometric formulas of the Lewis’s law and Aboav-Weaire’s law in two dimensions based on ellipse packing

The two-dimensional (2D) Lewis’s law and Aboav-Weaire’s law are two simple formulas derived from empirical observations. Numerous attempts have been made to improve the empirical formulas. In this study, we simulated a series of Voronoi diagrams by randomly disordered the seed locations of a regular hexagonal 2D Voronoi diagram, and analyzed the cell topology based on ellipse packing. Then, we derived and verified the improved formulas for Lewis’s law and Aboav-Weaire’s law. Specifically, we found that the upper limit of the second moment of edge number is 3. In addition, we derived the geometric formula of the von Neumann-Mullins’s law based on the new formula of the Aboav-Weaire’s law. Our results suggested that the cell area, local neighbor relationship, and cell growth rate are closely linked to each other, and mainly shaped by the effect of deformation from circle to ellipse and less influenced by the global edge distribution.

The geometric formulas of the Lewis’s law and Aboav-Weaire’s law in two dimensions based on ellipse packing

The two-dimensional (2D) Lewis’s law and Aboav-Weaire’s law are two simple formulas derived from empirical observations. Numerous attempts have been made to improve the empirical formulas. In this study, we simulated a series of Voronoi diagrams by randomly disordered the seed locations of a regular hexagonal 2D Voronoi diagram, and analyzed the cell topology based on ellipse packing. Then, we derived and verified the improved formulas for Lewis’s law and Aboav-Weaire’s law. Specifically, we found that the upper limit of the second moment of edge number is 3. In addition, we derived the geometric formula of the von Neumann-Mullins’s law based on the new formula of the Aboav-Weaire’s law. Our results suggested that the cell area, local neighbor relationship, and cell growth rate are closely linked to each other, and mainly shaped by the effect of deformation from circle to ellipse and less influenced by the global edge distribution.