Hi! I am a first-year PhD student in Department of Computer Science, Stanford University, currently rotating with Prof. Dan McFarland and Prof. Dan Jurafsky. In previous quarters, I rotated with Prof. Michael Bernstein and Prof. Jure Leskovec.
I received my bachelor degree (with honors) from Department of Electronic Engineering, Tsinghua University in 2018. I studied as an exchange student in A. James Clark School of Engineering, University of Maryland, College Park in Fall 2016.
I have been a research assistant in Future Communications and Internet Lab, Tsinghua University, advised by Prof. Yong Li and Prof. Vassilis Kostakos (University of Melbourne) since 2015. In Fall 2016, I was fortunate enough to work under the supervision of Distinguished University Professor Prof. Hanan Samet at University of Maryland. In Summer 2017, I was a visiting student and research assistant at Human Dynamics Group, MIT Media Lab, advised by Prof. Alex 'Sandy' Pentland and Prof. Xiaowen Dong.
My research interests lie in data mining, ubiquitous computing, computational social science and complex networks.News: [July 2019] Paper on social commerce accepted to ICWSM'20! [preprint] News: [May 2019] Received SIGCHI Student Travel Grant for UbiComp'19. See you in London!
Sep. 2018 - Jun. 2023 (Expected), Department of Computer Science, Stanford University,
Doctor of Philosophy.
Aug. 2014 - Jul. 2018, Department of Electronic Engineering, Tsinghua University,
Bachelor of Engineering (with honors), GPA: 92.2/100, Rank: 12/262.
Aug. 2016 - Dec. 2016, A. James Clark School of Engineering, University of Maryland, College Park,
Exchange Student, GPA: 3.83/4.0.
Jun. 2017 - Sep. 2017, Human Dynamics Group, Media Laboratory, Massachusetts Institute of Technology,
Visiting Student and Research Assistant.
Jul. 2018 - Sep. 2018, Map Service, Mobile Internet Group, Tencent Inc,
H. Cao*, Z. Chen*, F. Xu, T. Wang, Y. Xu, L. Zhang, Y. Li, When Your Friends Become Sellers: An Empirical Study of Social Commerce Site Beidian, to appear in AAAI ICWSM 2020. [preprint]
H. Cao, Z. Chen, F. Xu, Y. Li, V. Kostakos, Revisitation in Urban Space vs. Online: A Comparison across POIs, Websites, and Smartphone Apps, in PACM IMWUT (UbiComp 2019). [pdf]
H. Cao, J. Feng, Y. Li, V. Kostakos, Uniqueness in the City: Urban Morphology and Location Privacy, in PACM IMWUT (UbiComp 2018). [pdf]
H. Cao, F. Xu, J. Sankaranarayanan, Y. Li, H. Samet, Habit2vec: Trajectory Semantic Embedding for Living Pattern Recognition in Population, in IEEE TMC (IF: 4.098). [pdf]
H. Cao, J. Sankaranarayanan, J. Feng, Y. Li, H. Samet, Understanding Metropolitan Crowd Mobility via Mobile Cellular Accessing Data, in ACM TSAS . [pdf]
M. Zeng, H. Cao, M. Chen, Y. Li, User Behavior Modeling, Recommendations, and Purchase Prediction During Online Shopping Festivals, in Springer Electronic Markets (EM, IF: 3.818). [pdf]
H. Shi, H. Cao, X. Zhou, Y. Li, C. Zhang, V. Kostakos, F. Sun, F. Meng, Semantics-Aware HMM for Human Mobility Modelling, in SDM 2019. [pdf]
F. Xu, T. Xia, H. Cao, Y. Li, F. Sun, F. Meng, Detecting Popular Temporal Modes in Population-scale Unlabelled Trajectory Data, in PACM IMWUT (UbiComp 2018). [pdf]
Purchasing Pattern Recognition in Metropolis,
Hancheng Cao, Xiaowen Dong, Alex 'Sandy' Pentland,
Jun 2017 - Present, MIT Media Lab,
We recognize the leading purchasing pattern of a major European city through a representation learning and Monte Carlo simulation based method. Leveraging a real-world transaction dataset from the country’s most popular bank, we analyze the city’s purchasing pattern at weekly and monthly level and investigate the relationship between purchasing behavior and user social-demographic features (e.g., sex, age, income, educational level, home location), churn pattern as well as social learning, which contribute to our understanding of the city's economic structure.
Uniqueness in the City: Urban Morphology and Location Privacy,
Hancheng Cao, Jie Feng, Yong Li, Vassilis Kostakos,
Sep 2017 - Nov 2017, University of Melbourne,
We provide the first investigation on the potential for privacy leaks when users reveal their nearby Points-of-Interest (POIs), e.g., using location-based recommendation service. Specifically, we investigate whether and how a person's location can be reverse-engineered when that person simply reveals their nearby POI types (e.g. 2 schools and 3 restaurants). We approach our analysis by introducing a "Location Re-identification" algorithm that is computationally efficient. Using data from Open Street Map, we show that urban morphology has a clear link to location privacy, and we highlight a number of urban factors that contribute to location privacy.
Habit2vec: Trajectory Semantic Embedding for Living Pattern Recognition in Population,
Hancheng Cao, Fengli Xu, Jagan Sankaranarayanan, Yong Li, Hanan Samet,
May 2017 - Nov 2017 University of Maryland,
We introduce representation learning in semantic-rich trajectory data analysis and propose a framework named habit2vec to represent user trajectory semantics in vector space for living pattern recognition. We evaluated our proposed system on a large-scale real-world dataset provided by a popular social network vendor. The results justify the representation ability of our system in preserving user habit pattern, and demonstrate its effectiveness in clustering users with similar living habits.
User Behavior Modeling, Recommendations and Purchase Prediction,
Ming Zeng, Hancheng Cao, Min Chen, Yong Li,
Jun 2017 - Oct 2017, Tsinghua University,
We provide an exhaustive case study on user online browsing and purchasing behaviors during a large shopping festival in China using 31 million logs generated on 11st November, 2016 (Double 11 Shopping Day), which aims at improving online shopping experience for consumers, increasing sales for merchants and achieving effective warehousing and delivery.
Semantic-Aware HMM for Human Mobility Modeling,
Hongzhi Shi, Hancheng Cao, Xiangxin Zhou, Yong Li, Chao Zhang, Vassilis Kostakos,
Jun 2017 - Oct 2017, Tsinghua University,
We propose a novel semantic-aware mobility model that captures human mobility motivation using large-scale semantic-rich spatial temporal data from location-based social networks. We first develop a multimodal embedding method, then we use hidden Markov model to learn latent states and transitions. We further propose a von Mises-Fisher mixture clustering for user grouping so as to tackle data sparsity. We evaluate our proposed method on two large-scale real-world datasets, where we validate the ability of our method to produce high-quality mobility models and show that our model outperforms baseline mobility models in various tasks.
Detecting Popular Temporal Modes in Population-scale Unlabelled Trajectory Data,
Fengli Xu, Tong Xia, Hancheng Cao, Yong Li, Funing Sun, Fanchao Meng,
Jun 2017 - Oct 2017, Tsinghua University,
We investigate how people allocate their time in their daily routine from spatial temporal big data. We propose a pipeline system consisting of a noise handler, a feature extractor and a mode detector to recognize popular temporal modes in population. Through three real-world dataset, our proposed system reveals insightful correlations between popular temporal modes and individual social economic status, such as occupation.
Understanding Metropolitan Crowd Mobility via Mobile Cellular Accessing Data,
Hancheng Cao, Jagan Sankaranarayanan, Jie Feng, Yong Li, Hanan Samet,
Sep 2016 - Apr 2017, University of Maryland,
We propose a novel approach for analyzing crowd mobility on a "city block" level through mobile cellular accessing data for deeper urban dynamics understanding. We propose algorithms to detect homes, working places and stay regions for individual user trajectories, as well as a method for analyzing commute patterns and spatial correlation. Leveraging a large-scale dataset in Shanghai, we discover commute patterns, spatial correlation rules as well as a hidden structure of the city based on crowd mobility analysis.
I was a member of Tsinghua Student Art Troupe Clavier team and served as vice captain in 2015 and 2016. I have been playing classical piano since 2001 and attained the highest level for amateur player in 2008. I also play the trombone and attained the highest level in 2007.
I enjoy works by Ludwig van Beethoven (who provides me with immense power when I'm feeling down), Johannes Brahms, Franz Schubert, J.S. Bach, Claudio Monteverdi, Anton Bruckner, Gabriel Fauré, Claude Debussy and Alexander Scriabin. Glenn Gould, Yehudi Menuhin, Sergiu Celibidache, Maria Callas are some of my favourite interpreters.Right now I am practicing Beethoven's piano sonata No.30 (OP. 109). I would love to try chamber music as well, e.g., Brahms violin sonata No.1 (Op.78) and Schubert's fantasy for piano four hands (D.940).
I enjoy British poems, Chinese modern literature, visual arts (painting, sculpture and architecture), history and social science. I am especially passionate about the field of social history (where I have drawn inspirations for my research).
I am also keen on sharing my knowledge with others. I was captain of volunteer docents at Shanghai Soong Ching-ling Memorial Residence in 2012 and 2013.
I enjoy swimming (I was a member of Tsinghua EE swimming team), jogging, table tennis and working out. Currently I am learning Taekwondo and Karate, and I plan to start on rock climbing and camping.Ping me if you are interested as well!