About

About Me

I’m Liang Cheng, a PhD candidate in the Department of Informatics at the University of Oslo. My research focuses on combining large language models and knowledge graphs to build intelligent systems for e-waste recycling and circular economy.

My PhD project — Efficient Recycling of E-Waste through Automated and Intelligent Resource Dataflow — explores how structured knowledge can be automatically extracted from scientific literature, product specifications, and unstructured data, then used to support decision-making in waste management and recycling.

🔬 Research Interests +

• Knowledge Graph Construction — ontology-guided LLM extraction, scientific document processing, table-to-KG
• Large Language Models — synthetic data generation, quality assessment, multi-agent extraction pipelines
• Graph Machine Learning — zero-shot link prediction, cross-domain transfer, structural role learning
• E-Waste Recycling — WEEE domain knowledge systems, compositional data, recovery optimization
• RAG / Expert Agents — domain-specific retrieval-augmented systems for scientific decision support
• Applied ML — mixture density networks, transfer learning, dimensionality reduction

🎓 Education +

PhD in Digital Knowledge Systems Apr 2024 – Apr 2027 (expected)

University of Oslo, Department of Informatics | Oslo, Norway

• Project: Efficient Recycling of E-Waste through Automated and Intelligent Resource Dataflow
• Research: LLM + Knowledge Graph methods for automated resource dataflow in e-waste recycling

M.Sc. in Data Science (Machine Learning and Statistics Track) Sep 2021 – Jun 2023

Uppsala University | Uppsala, Sweden

• GPA: 4.0/5.0
• Thesis work on autoencoder-based parameter inference of gene regulatory networks

B.Eng. in Software Engineering Sep 2017 – Jun 2021

Harbin University of Science and Technology | Harbin, China

• Graduated with honors
• Final project: movie recommendation system with collaborative filtering

🛠️ Skills +

Languages +

• Python (primary), JavaScript, SQL, basic C++ / R

ML / DL +

• PyTorch, PyTorch Geometric, scikit-learn, SciPy, pandas, NumPy, SentenceTransformers

Knowledge Graphs +

• RDF, SPARQL, GraphDB, OWL ontology engineering

LLM Workflows +

• Multi-agent extraction pipelines, synthetic data generation, RAG, LLM API integration

Data Viz +

• Matplotlib, Seaborn

Tools +

• Git, COMSOL Multiphysics, Spark (basic), MongoDB (basic), Docker

Languages (spoken) +

• Mandarin (native), Cantonese (native), English (fluent)