Artificial intelligence-based collaborative filtering method with ensemble learning for personalized lung cancer medicine without genetic sequencing.

In personalized medicine, many factors influence the choice of compounds. Hence, the selection of suitable medicine for patients with non-small-cell lung cancer (NSCLC) is expensive. To shorten the decision-making process for compounds, we propose a computationally efficient and cost-effective collaborative filtering method with ensemble learning. The ensemble learning is used to handle small-sample sizes in drug response datasets as the typical number of patients in a cancer dataset is very small. Moreover, the proposed method can be used to identify the most suitable compounds for patients without genetic data. To the best of our knowledge, this is the first method to provide effective recommendations without genetic data. We also constructed a reliable dataset that includes eight NSCLC cell lines and ten compounds that have been approved by the Food and Drug Administration. With the new dataset, the experimental results demonstrated that the dataset shift phenomenon that commonly occurs in practical biomedical data does not occur in this problem. The experimental results demonstrated that our proposed method can outperform two state-of-the-art recommender system techniques on both the NCI60 dataset and our new dataset. Our model can be applied to the prediction of drug sensitivity with less labor-intensive experiments in the future.

Complementary Deep and Shallow Learning with Boosting for Public Transportation Safety.

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%.