Using random projection, a method to speed up both kernel k-means and centroid initialization with k-means++ is proposed. We approximate the kernel matrix and distances in a lower-dimensional space ${\mathbb{R}}^d$ before the kernel k-means clustering motivated by upper error bounds. With random projections, previous work on bounds for dot products and an improved bound for kernel methods are considered for kernel k-means. The complexities for both kernel k-means with Lloyd’s algorithm and centroid initialization with k-means++ are known to be $O\left(nkD\right)$ and $\Theta \left(nkD\right)$, respectively, with n being the number of data points, the dimensionality of input feature vectors D and the number of clusters k. The proposed method reduces the computational complexity for the kernel computation of kernel k-means from $O\left(n^{2}D\right)$ to $O\left(n^{2}d\right)$ and the subsequent computation for k-means with Lloyd’s algorithm and centroid initialization from $O\left(nkD\right)$ to $O\left(nkd\right)$. Our experiments demonstrate that the speed-up of the clustering method with reduced dimensionality $d=200$ is 2 to 26 times with very little performance degradation (less than one percent) in general.

We propose three algorithms as approximations to k-means++ with complexity only O(nk) which is independent of D with n being the number of data points, k being the number of clusters and D the dimensionality. By combining random projection and k-means++, the proposed algorithm still enjoying the lower-bounded performance by results given for k-means++, on the other hand, the errors introduced by the approximation are also bounded by JL lemma, which is sufficently low in the experiment we presented. By approximating D(x) ² in the k-means++ algorithm, we obtain an approximate algorithm to k-means++ of complexity only O(nk). The approximation to D(x) ² can be obtained through random projections by only perturbing the distances to a provably small extent. Our experiments on real-world datasets show that, when k is small, the performance of minimization of the potential deteriorated only by 3%-5% while the execution time can be shortened to 50% of the original. With a large k, compared to the results of our proposed algorithms with small values of k, the error introduced by random projection is much reduced and the speed up of the execution time is improved. In our experiments, it is shown that the speed up of the proposed algorithms is up to 20 times compared to k-means++.

Behavioral theory assumes that leaders can be identified by their daily behaviors. Social network analysis helps to understand behavioral patterns within their social networks. This work considers leaders as the managerial personnel of the organization and differentiates managements from non-managerial staff by their behavior with five different types of interactions with PageRank and their attributes in modern organizations. PageRank and word embedding using word2vec with phrases from features are adopted to extract new features for the identification of managerial staff. Both traditional machine learning methods and graph neural networks are utilized with real-world data from an Austrian IT company called Knapp System Integration. Our experimental results show that the proposed new features extracted using PageRank with different types of interactions and word2vec with phrases significantly improve the identification accuracy. We also propose to use graph neural networks as an effective learning algorithm to identify managers from organizations. Our approach can identify managerial staff with an accuracy of around 80%, which demonstrates that managers could be identified through social network analysis. By analyzing the behaviors of members, the proposed method is effective as a performance appraisal tool for organizations. The study facilitates sustainable management by helping organizations to retain managerial talents or to invite potential talents to join the management team.

To monitor road safety, billions of records can be generated by Controller Area Network bus each day on public transportation. Automation to determine whether certain driving behaviour of drivers on public transportation can be considered safe on the road using artificial intelligence or machine learning techniques for big data analytics has become a possibility recently. Due to the high false classification rates of the current methods, our goal is to build a practical and accurate method for road safety predictions that automatically determine if the driving behaviour is safe on public transportation. In this paper, our main contributions include (1) a novel feature extraction method because of the lack of informative features in raw CAN bus data, (2) a novel boosting method for driving behaviour classification (safe or unsafe) to combine advantages of deep learning and shallow learning methods with much improved performance, and (3) an evaluation of our method using a real-world data to provide accurate labels from domain experts in the public transportation industry for the first time. The experiments show that the proposed boosting method with our proposed features outperforms seven other popular methods on the real-world dataset by 5.9% and 5.5%.