WSDM 2014

54 papers
The 2014 WSDM conference presented 86 papers centered on large-scale data mining and machine learning for web applications. A primary focus was the challenge of scalability in the "big data" era, with research exploring web-scale classification and efficient clustering methods like ranked retrieval-based K-Means. Other significant themes included social network analysis, specifically regarding community detection and node labeling, as well as the development of nonparametric Bayesian models for multi-modal data. Practical applications in digital advertising and search intent were also prominent. Studies addressed conversion optimization and mobile advertising response prediction using hierarchical factorization machines and multitask learning algorithms. Additionally, research into intent learning utilized heterogeneous graphs to link queries with web pages and external knowledge bases like Wikipedia. Prolific authors contributing to these themes included Jiawei Han, Susan T. Dumais, Philip S. Yu, and Patrick Gallinari. The year’s proceedings reflected a shift toward integrating complex network structures with predictive modeling to improve information retrieval and user behavior analysis.

Top Research Directions

The 19 research directions with WSDM 2014 papers, ranked by paper count, next to 2013. Share is the fraction of the edition’s papers in the direction.

#Direction20142013Share 2014Share 20132013 rank
1Balancing Relevance and Diversity with Maximal Marginal Relevance101.9%0.0%—
2Document Length Normalization in Information Retrieval and Learning101.9%0.0%—
3Embedding Temporal Dynamics in Knowledge Graphs101.9%0.0%—
4Hierarchical Feature Sharing for Long Tailed Classification101.9%0.0%—
5Hybrid Pointwise Pairwise Optimization for Robust Ranking101.9%0.0%—
6Localized Subgraph Extraction for Relational Reasoning101.9%0.0%—
7Mitigating the Cold Start Problem in Recommender Systems101.9%0.0%—
8Modeling Asymmetric Social Influence in Multi-Agent Systems101.9%0.0%—
9Modeling Statistical Dependencies in Networked Data101.9%0.0%—
10Modularity Maximization for Network Community Detection111.9%1.8%10
11Multi-Constraint Graph Partitioning101.9%0.0%—
12Overcoming Semantic Sparsity in Short Texts101.9%0.0%—
13Overcoming the Document Independence Assumption101.9%0.0%—
14Reducing Annotation Costs through Active Sample Selection101.9%0.0%—
15Relational Specificity Weighting in Knowledge Graphs101.9%0.0%—
16Relative Preference Ranking for Quality Assessment101.9%0.0%—
17Scaling Graph Learning to Massive Networks101.9%0.0%—
18Search Result Diversification for Information Redundancy Removal101.9%0.0%—
19The Topological Information Loss of Graph Linearization101.9%0.0%—

Download all directions as CSV

Institutions

Institutions credited by WSDM 2014 papers: a paper credits every institution named in its affiliations, once. 4 institutions share 3 credited papers.

#InstitutionPapersShare20132013 rank
1009yjbg58 133.3%0—
2Stanford University 133.3%13
3Tsinghua University 133.3%0—
4University of Notre Dame 133.3%0—

Download all institutions as CSV

About this data

Papers. WSDM 2014 lists 54 papers on Lacuna, duplicate versions counted once; rejected and withdrawn submissions are left out.

Directions. Directions are Lacuna’s clustering of the concept elements extracted from each paper, so a paper can belong to several and direction counts add up to more than the paper count. A direction’s share is its papers divided by the edition’s papers. Comparisons with 2013 use shares, since the edition went from 55 to 54 papers.

Institutions. A paper credits each institution named in its affiliations, resolved to a ROR record, once. 3 of 54 papers (5.6%) credit at least one institution and are the base for institution shares; the rest have no resolved affiliation yet.

Data. Counts come from Lacuna’s data build of 2026-11-17 and change as papers are added. Tables as CSV: directions · institutions. Cite as: Lacuna, “WSDM 2014”, https://lacuna.tiptreesystems.com/venue/wsdm/2014.

Top Authors

AuthorPapers
Jiawei Han3
Patrick Gallinari3
Ryen W White3
Abhimanyu Das2
Ahmed Hassan Awadallah2
Alex Smola2
Amr Ahmed2
Brian D. Davison2
Dawei Yin2
Hongning Wang2
Jure Leskovec2
Ludovic Denoyer2
Philip S. Yu2
Xiangnan Kong2
Xiang Ren2
Xiao Yu2
Aristides Gionis1
Bin Bi1
Bin Xu1
Bo Long1

Impactful Papers

Papers from WSDM 2014 that shaped research clusters.

Influential

Papers

Title
1st workshop on diffusion networks and cascade analytics
Active learning for networked data based on non-progressive diffusion model
Adapting deep RankNet for personalized search
A few good predictions: selective node labeling in a social network
Big graph mining for the web and social media: algorithms, anomaly detection, and applications
Chinese-English mixed text normalization
Customized tour recommendations in urban areas
Democracy is good for ranking: towards multi-view rank learning and adaptation in web search
Detecting cohesive and 2-mode communities indirected and undirected networks
Detecting non-gaussian geographical topics in tagged photo collections
Discovering common motifs in cursor movement data for improving web search
Entity linking at the tail: sparse signals, unknown entities, and phrase models
Estimating ad group performance in sponsored search
Exploiting contextual factors for click modeling in sponsored search
Exploiting user disagreement for web search evaluation: an experimental approach
FENNEL: streaming graph partitioning for massive scale graphs
Going beyond Corr-LDA for detecting specific comments on news & blogs
Heterogeneous graph-based intent learning with queries, web pages and Wikipedia concepts
Improving search relevance for short queries in community question answering
Inferring the impacts of social media on crowdfunding
Is a picture really worth a thousand words?: - on the role of images in e-commerce
Knowledge-based graph document modeling
LASER: a scalable response prediction platform for online advertising
Latent dirichlet allocation based diversified retrieval for e-commerce search
Learning latent representations of nodes for classifying in heterogeneous social networks
Learning social network embeddings for predicting information diffusion
Lessons from the journey: a query log analysis of within-session learning
Modeling dwell time to predict click-level satisfaction
Modeling opinion dynamics in social networks
Multilingual probabilistic topic modeling and its applications in web mining and search
Nonparametric bayesian upstream supervised multi-modal topic models
On building entity recommender systems using user click log and freebase knowledge
Partner tiering in display advertising
Personalized entity recommendation: a heterogeneous information network approach
Predicting response in mobile advertising with hierarchical importance-aware factorization machine
Ranking in heterogeneous social media
Relative confidence sampling for efficient on-line ranker evaluation
Sampling dilemma: towards effective data sampling for click prediction in sponsored search
Scalable hierarchical multitask learning algorithms for conversion optimization in display advertising
Scalable K-Means by ranked retrieval
Scalable topic-specific influence analysis on microblogs
Search engine click spam detection based on bipartite graph propagation
Sentiment analysis on evolving social streams: how self-report imbalances can help
Spatial compactness meets topical consistency: jointly modeling links and content for community detection
Struggling or exploring?: disambiguating long search sessions
Taxonomy discovery for personalized recommendation
The last click: why users give up information network navigation
Transferring heterogeneous links across location-based social networks
Understanding and promoting micro-finance activities in Kiva.org
User modeling in search logs via a nonparametric bayesian approach

Page 1/2 next »