题目：Finding Orders Within Heterogeneous Information Networks
主讲：阳极，伊利诺伊大学厄巴纳-香槟分校(University of Illinois at Urbana-Champaign，UIUC)
时间：下午14:00pm - 16:00pm
Data are being generated everywhere in extraordinary speed and amount nowadays. Due to advanced data warehousing techniques and the much effort put into knowledge acquisition, lots of data are stored with structures, ready to be modeled under the setting of heterogeneous information networks. However, data in the networks are still noisy and incomplete, and entities interact in complex ways. Traditional network analyses are mostly focused on links and node attributes. They are ineffective in leveraging complex features and inefficient in exploring various network structures, which results in a huge gap between structured data and real-world applications.
In this talk, I will cover three pieces of my recent work, under the theme of finding orders within complex structured data. Firstly, I will introduce a joint learning framework for predicting user links and attributes in large social networks. The main idea is to gradually approach a precise and complete graph, via addressing smoothness on social graphs iteratively through two directions. Next, I will talk about a neural framework that is designed to capture social patterns underlying network communities, in terms of both user attributes and neighborhood local structures. The framework also explores supervision from example communities with end-to-end node embedding. Then I will discuss a joint learning neural framework for POI recommendation, through mining the user and POI networks. The framework explores users’ preference on POIs by fitting the feedback data, while preserving various user/POI context through smoothing on the heterogeneous context graphs. Finally, I will share some ideas on network mining through neural architectures, with a focus on pattern-aware network embedding.
Carl Yang is a fourth-year Ph.D. student with Prof. Jiawei Han in Computer Science at University of Illinois, Urbana Champaign. He received his B.Eng. in Computer Science at Zhejiang University under Xiaofei He in 2014. In his research, he develops data-driven techniques and neural architectures for learning with context-rich heterogeneous networks. His interests span data mining, machine learning and statistics, with a focus on leveraging graph analysis and deep learning techniques, to a wide range of questions including information network construction, entity/relation profiling, pattern-based network embedding and etc. Carl’s research results have been published in top conferences like KDD, WWW and ICML. Carl is experienced in multi-dimensional social media modeling, with multiple projects done in social graph construction and completion, user growth modeling and interest based user profiling and recommendation, by integrating users’ social activities with their mobility data. He has also taught or assisted the teaching of multiple graduate courses in University of Illinois at Urbana Champaign, including introduction to data mining, database systems and advanced data management.