Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks442 from public data
- DeepMove: Predicting Human Mobility with Attentional Recurrent Networks841
In DeepMove, an attentional recurrent network for mobility prediction from lengthy and sparse trajectories, a multi-modal embedding recurrent neural network is designed to capture the complicated sequential transitions by jointly embedding the multiple factors that govern the human mobility.
- 357
- Bridging Collaborative Filtering and Semi-Supervised Learning316
This work proposes to devise a general and principled SSL (semi-supervised learning) framework, to alleviate data scarcity via smoothing among neighboring users and POIs, and treat various context by regularizing user preference based on context graphs.
- Exploiting internal and external semantics for the clustering of short texts using world knowledge269
The proposed method employs a hierarchical three-level structure to tackle the data sparsity problem of original short texts and reconstruct the corresponding feature space with the integration of multiple semantic knowledge bases -- Wikipedia and WordNet.
- Emergence of Indigenous Artemisinin-Resistant Plasmodium falciparum in Africa265
Resistance to artemisinin resistance in Plasmodium falciparum is observed in a patient who was working in Equatorial Guinea.
- 220
- 210
- Beyond Human Parts: Dual Part-Aligned Representations for Person Re-Identification206
This paper addresses the missed contextual cues by exploiting both the accurate human parts and the coarse non-human parts by applying a human parsing model and a self-attention mechanism to capture the soft latent (non-human) part masks.
- Weakly-Supervised Neural Text Classification203
This paper proposes a weakly-supervised method that addresses the lack of training data in neural text classification and achieves inspiring performance without requiring excessive training data and outperforms baseline methods significantly.
- Text Classification Using Label Names Only: A Language Model Self-Training Approach196
This paper uses pre-trained neural language models both as general linguistic knowledge sources for category understanding and as representation learning models for document classification, and achieves around 90% accuracy on four benchmark datasets.
- 183
- 165
- 159
- GMove158
This work proposes GMove, a group-level mobility modeling method using GeoSM data that alternates between user grouping and mobility modeling, and generates an ensemble of Hidden Markov Models (HMMs) to characterize group- level movement regularity.
- Fine-Tuning Pre-trained Language Model with Weak Supervision: A Contrastive-Regularized Self-Training Approach153
This work develops a contrastive self-training framework, COSINE, to enable fine-tuning LMs with weak supervision, underpinned by contrastive regularization and confidence-based reweighting, which gradually improves model fitting while effectively suppressing error propagation.
- Regions, Periods, Activities150
CrossMap is presented, a novel cross-modal representation learning method that uncovers urban dynamics with massive GTSM data and significantly outperforms state-of-the-art methods for activity recovery and classification, but also achieves much better efficiency.
- 148
- GeoBurst148
GeoBurst is a method that enables effective and real-time local event detection from geo-tagged tweet streams and significantly outperforms state-of-the-art methods in precision, and is orders of magnitude faster.
- Robust solution of monotone stochastic linear complementarity problems146
A new error bound for the monotone LCP is given and used to show that solutions of the ERM formulation are robust in the sense that they may have a minimum sensitivity with respect to random parameter variations in SLCP.
- Splitter144
This work proposes Splitter to effectively mine fine-grained sequential patterns, which must satisfy spatial compactness, semantic consistency and temporal continuity simultaneously, and uses an effective algorithm called weighted snippet shift to detect clusters, and leverages a divide-and-conquer strategy to speed up the top-down pattern splitting process.
- Weakly-Supervised Hierarchical Text Classification142
This paper proposes a weakly-supervised neural method for hierarchical text classification that features a hierarchical neural structure, which mimics the given hierarchy and is capable of determining the proper levels for documents with a blocking mechanism.
- TaxoGen140
The method, TaxoGen, uses term embeddings and hierarchical clustering to construct a topic taxonomy in a recursive fashion and consists of an adaptive spherical clustering module for allocating terms to proper levels when splitting a coarse topic into fine-grained ones.
- 132
- 132
- Easing Embedding Learning by Comprehensive Transcription of Heterogeneous Information Networks126
The HEER algorithm is proposed, which embeds HINs via edge representations that are further coupled with properly-learned heterogeneous metrics, and demonstrates the effectiveness of the proposed HEER model and the utility of edge representations and heterogeneity metrics.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-13. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.