• 강사 : 황지영 교수 (KAIST 전산학부) 
• 일시 : 2020. 10. 8 (목) 17:00~18:30

Graphs are useful tools to model real-world data that is best represented by a set of objects and the relationships between the objects, e.g., WWW, social networks, and biological networks among others. This talk mainly focuses on mining and learning methods for graphs with three specific topics: clustering, hypergraphs, and representation learning. Traditional clustering algorithms, such as K-Means, output a clustering that is disjoint and exhaustive, i.e., every single data point is assigned to exactly one cluster. However, in many real-world datasets, clusters can overlap and there are often outliers that do not belong to any cluster. We propose NEO-K-Means (Non-Exhaustive, Overlapping K-Means) that captures the issues of overlap and non-exhaustiveness in a unified manner. Complex relationships among entities can be modeled very effectively using hypergraphs. Hypergraphs model real-world data by allowing a hyperedge to include two or more entities. We propose a semi-supervised clustering framework for hypergraphs that is able to easily incorporate not only multiple attributes of the entities but also auxiliary relationships among the entities from diverse sources. Furthermore, by showing the close relationship between the hypergraph normalized cut and the weighted kernel K-Means, we also develop an efficient multilevel hypergraph clustering method which provides a good initialization with our semi-supervised multi-view clustering algorithm. On a web graph, a node indicates a web page and a directed edge indicates a hyperlink between the web pages. The hyperlinks are created for different reasons, and thus, may play different roles in the graph. We formally define a hyperlink classification problem in web search by classifying hyperlinks into three classes based on their roles: navigation, suggestion, and action. We approach the hyperlink classification problem from a structured graph embedding (also known as representation learning) perspective, and show that we can solve the problem by modifying the recently proposed knowledge graph embedding techniques.