Research

Ongoing

  • Collaboration with eBay on fraud detection projects, three papers published, one under review, one ongoing.
    • Graph Decomposition:
      (1) DeGNN, accepted at KDD 2021.
    • Heterogeneous Graph and Fraud Detection:
      (2) xFraud, accepted at VLDB 2022.
    • Dynamic Graph and Fraud Detection:
      (3) DHGReg, accepted in the AAAI Workshop on Deep Learning on Graphs 2021.
      (4) DyHGN.
    • Real-time Fraud Detection with Lambda Architecture:
      (5) LNN on DSS,
      (6) BRIGHT, accepted at CIKM 2022.
  • Understanding the connectedness amongst disciplines, project poster.
    1. Peter Egger, Susie Rao, and Ce Zhang. Hierarchical classification of subfields in CS/Econ (Arxiv).
    2. Peter Egger, Susie Rao, and Ce Zhang. Understanding the Web of Science Using Deep Learning: a Novel Hierarchical Multi-class Multi-label Classification System (Scientometrics working paper).
    3. Peter Egger, Susie Rao, and Ce Zhang. A Meta-Science Documentation of Interdisciplinarity: How we Evolve as One Scientific Community (working paper).
  • EconMultiplex – Multiplex Networks in International Trade in collaboration with Swiss Data Science Center, Principal Investigator: Prof. Dr. Egger Peter.
    1. Egger, Peter, Susie Xi Rao, and Sebastiano Papini. “A new algorithm for matching Chinese NBS firm-level with customs data.” China Economic Journal 14.3 (2021): 311-335. (paper)
    2. Peter Egger, Vincent Lohmann, and Susie Rao. How do Firms Choose Their Location: Evidence from China (working paper).

Publication @ Google scholar

Presentation and Talks

Future

  • Our work “BRIGHT - Graph Neural Networks in Real-Time Fraud Detection” (in collab with eBay) got accepted at CIKM2022. The talk will be on Oct 20, 2022 (session: S10-F: ADVERTISING & E-COMMERCE).

  • Last year’s presentation of my dissertation “Pattern Mining in Graph-structured Data” (November 4, 2022).

  • Guest lecture in “Fraud Detection” Seminar at HSLU - Hochschule Luzern (tba).

  • Data science night on “Fraud Detection” at FHNW - Fachhochschule Nordwestschweiz (March 30, 2023).

Recent

Teaching

  • Teaching assistant for Big Data, Department of Computer Science, ETHZ, Autumn Semester, 2022.

  • Teaching assistant for Big Data for Engineers, Department of Computer Science, ETHZ, Spring Semester, 2022.

  • Teaching assistant (online) for Big Data, Department of Computer Science, ETHZ, Autumn Semester, 2020.

  • Economic Network Analysis with Prof. Dr. Egger Peter, World Trade Institute at Uni Berne, in Nov, 2019.

  • Teaching assistant for Big Data for Engineers, Department of Computer Science, ETHZ, Spring Semester, 2019.

  • Teaching certificate by successfully finishing Learning to Teach, a pedagogical training workshop from LET, ETHZ.

Master thesis/Semester project

Research assistant supervision at the Chair of Applied Economics

  • Ongoing:
  • Past:
    • Zuoyue Li: a law database crawler and XML-based information extraction.
    • Jiajing Feng, Xi Chen: information extraction from Chinese yearbooks.
    • Daoye Wang: code management and extension of Chinese database matching.
    • Ada Langenfeld (jointly with Johannes Rausch): Annotations for TableParser (an table annotator built on DocParser).
    • Xi Chen (jointly with Vincent Lohmann): Chinese dataset geocoding.
    • Rahul Rarade (jointly with Vincent Lohmann): data support in Science projects and ORBIS geocoding.

Master student supervision at DS3Lab (BT: bachelor thesis, MT: master thesis, SP: semester project)

  • Ongoing:
  • Past:
    • Livio Kaiser (MT jointly with Johannes Rausch, Ce Zhang): Table element detection with weak supervision support from Excel data.
    • Anton Maksimov (MT jointly with Shuai Zhang, Ce Zhang): Fraud detection in large-scale transaction graph using Hyperbolic Heterogeneous Graph Neural Networks.
    • Alfonso Amayuelas (MT jointly with Shuai Zhang, Ce Zhang): Neural logic reasoning in knowledge graphs, published in ICLR 2022 (review, paper, code).
    • Clémence Lanfranchi (SP jointly with Shuai Zhang, Ce Zhang): Improving suspcious massive registration detection via temporal embedding methods in GNN.
    • Georgios Tsolakis (BT jointly with Ce Zhang, Peter Egger): The influence of international trade agreement on firm-firm network.
    • Piriyatamwong Piriyakorn: Improving the hierarchical classification system for scientific disciplines.

Grant (as co-applicant)

My master thesis

Rao, Xi. Automatic Labeling of Articles in International Investment Agreements : Using Semi-supervised Learning and Word Embeddings (2017). (Link)

Participation (besides conference presentations)