Our research on recommendation and social media mining

1. Recommendation and User Behaviour Modeling
2. POI knowledge graphs & recommendation
3. Mining Social Media, Reviews, and Forums

1. Recommendation and User Behaviour Modeling

Figure 1: Overview of our research on recommendation

Selected publications:

• Data issues in recommendation (WSDM’22) (Best paper runner up award)
• HyperML: A Boosting Metric Learning Approach in Hyperbolic Space for Recommender Systems. WSDM 2020 (Best paper award runner-up)
• Global Context Enhanced Graph Nerual Networks for Session-based Recommendation, SIGIR 2020
• Interact and Decide: Medley of Sub-Attention Networks for Effective Group Recommendation (SIGIR 19)
• Group Recommendation based on topic models(KDD14)    

Main contributors: Quan Yuan, Shanshan Feng, Pham Nguyen Tuan-Anh, Xutao Li, Xiucheng Li, Lucas VINH Tran, Jin Yao Chin

Collaborator: Sun Aixin, Wei Wei.

2. POI knowledge graph & recommendation

Figure 2: Overview about our research on POI knowledge graph & recommendation

Selected publications:

• HME: A Hyperbolic Metric Embedding Approach for Next-POI Recommendation, SIGIR 2020
• A new POI recommentdation approach, which performs better than previous approaches in experiments (SIGIR 2015)
• SAR: A sentiment-aspect-region model for user preference analysis and POI/user recommendation. The model provides explanations for recommendation results.  (ICDE 2015)
• A general graph model for recommendation in heterogeneous networks and its applications in event-based social networks (ICDE 2015)
• Diversity-aware POI recommendation (AAAI 2015)
• Time-aware POI recommendation (SIGIR13, CIKM14).  Datasets available here
• Mining significant semantic locations from user generated GPS data for recommendation (PVLDB10)

• W4: Discovering spatio-temporal topics for individual users and its various applications, e.g., requirement-aware POI recommendation  (KDD13, TOIS15).   Datasets available here

Main contributors: Quan Yuan, Shanshan Feng, Pham Nguyen Tuan-Anh, Tao Guo, Kaiqi Zhao, Zongcheng Ji, Dezhong Yao, Yile Chen, Junghoon Kim, Pasquale BALSEBRE

Collaborator: Sun Aixin, Jialong Han.

3. Mining Reviews, Social Media, and Forums

• We consider the impact of users' attributes, time factor, and novelty decay (Repeated exposures of an individual to an idea may have diminishing influence on the individual) for finding influential users.
• We develop techniques for review mining and sentiment analysis. 
• We also develop techniques for mining social media, including Micro-blogs (e.g., Twitter), and Community Based Question Answering Sites (e.g., Yahoo! Q&A).

Selected Publications:

• Inf2vec: Latent Representation Model for Social Influence Embedding (ICDE 18)
• DynaDiffuse: A dynamic diffusion model for continuous time constrained influence maximization (AAAI 15)
• Finding influential event organizers in event based social networks (SIGMOD14)
• Influence maximization with novelty decay (AAAI14)
• Time constrained influence maximization in social networks ( ICDM12 , TKDE . Source code)
• Computing top-k influential nodes (KDD10, AAAI 11)  
• Detecting user intents from tweets (AAAI 15)
• Coarse-to-fine review selection via supervised joint aspect and sentiment model (SIGIR14)
• One seed to find them all: Mining opinion features via association (CIKM12)
• Geolocation prediction for social images by exploring user profiles (JASIST14) 
• On predicting popularity of newly emerging hashtags in Twitter (JASIST13)
• Short text classification ( WWW12 poster, evaluation paper JASIST ) and hierarchy maintenance ( SIGIR12).  Annotated dataset  for our SIGIR12 paper is available here.
• Using categorization information to improve question search in community based question answering services ( CIKM09, WWW2010, TOIS12). Annotated dataset is available here
• Extracting Question-Answer pairs from forums to build the QA database (SIGIR08, ACL08)
• Routing questions to expert users ( ICDE09)

Acknowledgement: Some of our projects are supported by grants awarded by Ministry of Education, NRF, IAF, Singtel/NCS, Roll-Royce, Alibaba, and Microsoft.