• Corpus ID: 14015788

Improving Latent User Models in Online Social Media

@article{Krishnan2017ImprovingLU,
  title={Improving Latent User Models in Online Social Media},
  author={Adit Krishnan and Ashish Sharma and H. Sundaram},
  journal={ArXiv},
  year={2017},
  volume={abs/1711.11124}
}
Modern social platforms are characterized by the presence of rich user-behavior data associated with the publication, sharing and consumption of textual content. Users interact with content and with each other in a complex and dynamic social environment while simultaneously evolving over time. In order to effectively characterize users and predict their future behavior in such a setting, it is necessary to overcome several challenges. Content heterogeneity and temporal inconsistency of behavior… 

Insights from the Long-Tail: Learning Latent Representations of Online User Behavior in the Presence of Skew and Sparsity

TLDR
This framework is the first to jointly address skew and sparsity across graphical behavior models and proposes a generalizable bayesian approach to partition users in the presence of skew while simultaneously learning latent behavior profiles over these partitions to address user-level sparsity.

SELF-ATTENTION NETWORKS

TLDR
DySAT is a novel neural architecture that operates on dynamic graphs and learns node representations that capture both structural properties and temporal evolutionary patterns, and has a significant performance gain over several different state of theart graph embedding baselines.

Dynamic Graph Representation Learning via Self-Attention Networks

TLDR
DySAT is a novel neural architecture that operates on dynamic graphs and learns node representations that capture both structural properties and temporal evolutionary patterns and has a significant performance gain over several different state-of-the-art graph embedding baselines.

An Adversarial Approach to Improve Long-Tail Performance in Neural Collaborative Filtering

TLDR
Experimental results show that even without auxiliary data, adversarial training can boost long-tail recall of state-of-the-art NCF models by up to 25%, without trading-off overall performance.

Meta-GNN: Metagraph Neural Network for Semi-supervised learning in Attributed Heterogeneous Information Networks

TLDR
This paper proposes a novel metagraph convolution operation to extract features from local metagraph-structured neighborhoods, thus capturing semantic higher-order relationships in AHINs and proposes a proposed neural architecture Meta-GNN, which extracts features of diverse semantics by utilizing multiple metagraphs.

Motif-based Convolutional Neural Network on Graphs

TLDR
A novel deep architecture Motif-CNN is developed that employs an attention model to combine the features extracted from multiple patterns, thus effectively capturing high-order structural and feature information.

References

SHOWING 1-10 OF 58 REFERENCES

Insights from the Long-Tail: Learning Latent Representations of Online User Behavior in the Presence of Skew and Sparsity

TLDR
This framework is the first to jointly address skew and sparsity across graphical behavior models and proposes a generalizable bayesian approach to partition users in the presence of skew while simultaneously learning latent behavior profiles over these partitions to address user-level sparsity.

Who should share what?: item-level social influence prediction for users and posts ranking

TLDR
This paper proposes a Hybrid Factor Non-Negative Matrix Factorization (HF-NMF) approach for item-level social influence modeling, and devise an efficient projected gradient method to solve the HF- NMF problem.

Improving User Topic Interest Profiles by Behavior Factorization

TLDR
This work implemented and built a topic recommender predicting user's topical interests using their actions within Google+, and experimentally showed that it obtained better and cleaner signals than baseline methods, and is able to more accurately predict topic interests as well as achieve better coverage.

Mining topic-level influence in heterogeneous networks

TLDR
A generative graphical model is proposed which utilizes the heterogeneous link information and the textual content associated with each node in the network to mine topic-level direct influence and a topic- level influence propagation and aggregation algorithm is proposed to derive the indirect influence between nodes.

Personalized recommendation via cross-domain triadic factorization

TLDR
This paper proposes a generalized Cross Domain Triadic Factorization (CDTF) model over the triadic relation user-item-domain, which can better capture the interactions between domain-specific user factors and item factors.

Short and Sparse Text Topic Modeling via Self-Aggregation

TLDR
A novel model integrating topic modeling with short text aggregation during topic inference is presented, founded on general topical affinity of texts rather than particular heuristics, making the model readily applicable to various short texts.

LCARS: a location-content-aware recommender system

TLDR
LCARS is proposed, a location-content-aware recommender system that offers a particular user a set of venues or events by giving consideration to both personal interest and local preference, and a scalable query processing technique is developed by extending the classic Threshold Algorithm.

A dirichlet multinomial mixture model-based approach for short text clustering

TLDR
This paper proposed a collapsed Gibbs Sampling algorithm for the Dirichlet Multinomial Mixture model for short text clustering and found that GSDMM can infer the number of clusters automatically with a good balance between the completeness and homogeneity of the clustering results, and is fast to converge.

An Adversarial Approach to Improve Long-Tail Performance in Neural Collaborative Filtering

TLDR
Experimental results show that even without auxiliary data, adversarial training can boost long-tail recall of state-of-the-art NCF models by up to 25%, without trading-off overall performance.

Finding Bursty Topics from Microblogs

TLDR
A topic model that simultaneously captures two observations is proposed that helps find event-driven posts on microblogs and helps identify and filter out "personal" posts.
...