Microservice is a popular architecture to construct applications from a set of small independent services in cloud environment, leading to high cohesion, high availability, low coupling, and decent scalability. Due to large number of independent services in a microservice system, system faults generated from a single service would propagate to multiple services, eventually degraded the overall system performance and Quality of Service (QoS). Thus, it is crucial to efficiently and autonomously diagnose the runtime system fault. However, the complexity and dynamism of microservice systems and cloud environment pose unique challenges to precisely and robustly identify the faults and localize the root causes. In this paper, we propose an Autonomous Model Selection-Ensemble-Stacking (AMSES) framework for microservice system fault identification. The proposed framework can automatically select, ensemble, and stack optimal models from candidate unsupervised detection models for identifying different fault types robustly. In addition, AMSES can adaptively localize the fault services using autoselected root cause localization model. Moreover, by exploiting the fault degree and causal inferring score, we can diagnose the detected system fault precisely and interpretably. To evaluate the effectiveness, we empirically compare AMSES with state-of-the-art models on three kinds of faults on two microservice benchmarks: Sock-Shop and Train-Ticket. The experimental results show that AMSES can achieve$\mathbf{8 7. 1 \%}$and$\mathbf{9 1. 4 \%}$macroF1 average for fault type identification on Sock-Shop and TrainTicket, respectively. Meanwhile, AMSES could outperform its competitors for root cause localization with an average Avg@5 of 0.856 on Sock-Shop and 0.633 on Train-Ticket.

Spoken language proficiency is critically important for children’s growth and personal development. Due to the limited and imbalanced educational resources in China, elementary students barely have chances to improve their oral language skills in classes. Verbal fluency tasks (VFTs) were invented to let the students practice their spoken language proficiency after school. VFTs are simple but concrete math related questions that ask students to not only report answers but speak out the entire thinking process. In spite of the great success of VFTs, they bring a heavy grading burden to elementary teachers. To alleviate this problem, we develop Dolphin, a spoken language proficiency assessment system for Chinese elementary education. Dolphin is able to automatically evaluate both phonological fluency and semantic relevance of students’ VFT answers. We conduct a wide range of offline and online experiments to demonstrate the effectiveness of Dolphin. In our offline experiments, we show that Dolphin improves both phonological fluency and semantic relevance evaluation performance when compared to state-of-the-art baselines on real-world educational data sets. In our online A/B experiments, we test Dolphin with 183 teachers from 2 major cities (Hangzhou and Xi’an) in China for 10 weeks and the results show that VFT assignments grading coverage is improved by 22%.