Xiaoke Yang Email & Phone Number

Recommender systems, discriminative and generative learning algorithms development, Alibaba Cloud, China.• Various projects across the product lines either carried out or supervised by me. • The recommender system selects potential customers for a given product using data-based, content-embedding, and behaviour-embedding based methods. A domain name recommendation function was also developed using similar ideas. • Discriminative models include content risk detection (text classification via LMs), trademark search (image classification and similarity detection using CNN and multi-modality models), and automated examination for design patent and software copyright applications.• Generative models include text generation for trademark/company/website names, etc using language models and logo generation using diffusion models. A ChatGPT-like digital assistant is also under development.Some of the functions are online at wanwang.aliyun.com with their introductions, and some serve at the internal system to assist the examination process.

Beijing City, China

Airline booking systems anomaly detection and mitigation, Alibaba Group, China. • A Java Spring Boot application was developed, consisting of measurement, detection, decision making and actuation modules. The system accepts inputs from message queues, databases, API calls etc., performs anomaly detection, isolates the root cause, and mitigate the risks.• Short term problems were detected by filtering of the signals (e.g. via AR/MA models), long term problems were isolated by sensitivity analysis and structure learning for Bayesian networks. • Skills: Domain-driven design, SQL, concurrent systems, anomaly detectionAirline flight cancellation prediction.• The project was to predict whether a given flight on a certain departure date will be cancelled or not. The accuracy of algorithm was over 95%. • The algorithm followed a classic machine learning paradigm. Extensive feature engineering was done, features were manually crossed and selected by information values, logistic regression and gradient boosted decision tree models were used.

Beijing City, China

My research focuses on machine learning algorithms for aircraft avionics systems and embedded systems hardware and software design for unmanned aerial vehicle and IoT applications.

Cambridge, United Kingdom

Undergraduate supervisor for third year course: Systems and Control.

Cambridge, United Kingdom

Undergraduate teaching lab demonstrators for second and third year experiments: C++, Matlab, Evaporator, Ball and Beam, and Inverted Pendulum.

Doctor Of Philosophy (Phd), Information Engineering

Activities and Societies: Wolfson College Table Tennis Team, Cambridge Chinese Classics SocietyDissertation “Fault-tolerant predictive.

Master Of Engineering (M.Eng.), Automatic Control

Dissertation "Modelling and control of aircraft with variations in mass and centre of gravity", Research involves mathematical modelling.

Bachelor'S Degree, Automatic Control And Information Technology