Advances in Knowledge Discovery and Data Mining
Lecture notes in computer sciencePublished 1 January 2015Open access
Tru H. Cao, Ee‐Peng Lim, Zhi‐Hua Zhou, Tu-Bao Ho, David W. Cheung, Hiroshi Motoda
Citations58
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
Abstract
<p>19th Pacific-Asia Conference, PAKDD 2015, Ho Chi Minh City, Vietnam, May 19-22, 2015, Proceedings, Part II</p>
Keywords
Computer Science
Lecture notes in computer scienceReliable Early Classification on Multivariate Time Series with Numerical and Categorical Attributes
41 Citations2015Yu‐Feng Lin, Hsuan-Hsu Chen +2 more
A novel methodology named REACT (Reliable EArly ClassificaTion), which is the first work addressing the issue of constructing an effective classifier on multivariate time series with numerical and categorical attributes in serial manner so as to guarantee stability of accuracy compared to the classifiers using full-length time series.
Lecture notes in computer scienceDistributed Newton Methods for Regularized Logistic Regression
35 Citations2015Yong Zhuang, Wei-Sheng Chin +2 more
Experiments show that the proposed distributed Newton method for training logistic regression is competitive with or even faster than state-of-the-art approaches such as Alternating Direction Method of Multipliers and Vowpal Wabbit.
Lecture notes in computer scienceClustering on Multi-source Incomplete Data via Tensor Modeling and Factorization
34 Citations2015Weixiang Shao, Lifang He +1 more
A more general framework T-MIC (Tensor based Multi-source Incomplete data Clustering) to integrate multiple incomplete data sources is proposed, which first use the kernel matrices to form an initial tensor across all the multiple sources and formulate a joint tensor factorization process with the sparsity constraint.
Lecture notes in computer scienceEfficient Interactive Training Selection for Large-Scale Entity Resolution
28 Citations2015Qing Wang, Dinusha Vatsalan +1 more
Experiments show that manual labeling efforts can be significantly reduced for training an ER classifier without compromising matching quality, and a noisy oracle where manual labeling might be incorrect is considered.
Lecture notes in computer scienceWhat Is New in Our City? A Framework for Event Extraction Using Social Media Posts
24 Citations2015Chaolun Xia, Jun Hu +2 more
It is shown that integrating Instagram and Twitter post streams can improve event detection accuracy, and properly combining text, image and geolocation information is able to retrieve more relevant photos for events.
Lecture notes in computer scienceScalable Outlying-Inlying Aspects Discovery via Feature Ranking
22 Citations2015Nguyễn Xuân Vinh, Jeffrey Chan +4 more
OARank is presented – a hybrid framework that leverages the efficiency of feature selection based approaches and the effectiveness and versatility of score-and-search based methods and is suitable for mining large data sets.
Lecture notes in computer scienceUnderstanding Community Effects on Information Diffusion
22 Citations2015Shuyang Lin, Qingbo Hu +2 more
This work has found that links between communities function differently from those within communities, and no information diffusion model has yet considered how the community structure affects the diffusion process.
Lecture notes in computer scienceTrend-Based Citation Count Prediction for Research Articles
20 Citations2015Cheng–Te Li, Yu-Jen Lin +2 more
Experiments conducted on the ArnetMiner citation dataset exhibit promising results that T-CCP outperforms CCP and the proposed features are more effective than conventional ones.
Lecture notes in computer scienceConvolutional Nonlinear Neighbourhood Components Analysis for Time Series Classification
20 Citations2015Yi Zheng, Qi Liu +4 more
This paper investigates the distance metric learning and proposes a novel Convolutional Nonlinear Neighbourhood Components Analysis model for time series classification that can improve the classification accuracy to some extent and indicates that it can learn a good distance metric.
Lecture notes in computer scienceA Coupled k-Nearest Neighbor Algorithm for Multi-label Classification
16 Citations2015Chunming Liu, Longbing Cao
Experiments show that the proposed Coupled Multi-Label \(k\) Nearest Neighbor algorithm (CML-\(k\)NN) achieves superior performance than some existing multi-label classification algorithms.
Lecture notes in computer scienceLearning Topic-Oriented Word Embedding for Query Classification
11 Citations2015Hebin Yang, Qinmin Hu +1 more
This paper proposes a topic-oriented word embedding approach to address the query classification problem, and presents four neural network strategies based on the proposed model that achieve the high performance.
Lecture notes in computer scienceDiscovering the Impact of Urban Traffic Interventions Using Contrast Mining on Vehicle Trajectory Data
11 Citations2015Xiaoting Wang, Christopher Leckie +2 more
A framework to assess the impact of traffic intervention measures, such as road closures, on the traffic network and frequent sub-networks of the overall traffic network are discovered to reveal the region that is most affected.
Lecture notes in computer sciencePredicting Future Links Between Disjoint Research Areas Using Heterogeneous Bibliographic Information Network
11 Citations2015Yakub Sebastian, Eu-Gene Siew +1 more
The experimental results showed that HBIN meta path features could predict future co-citation links between these papers with high accuracy, outperforming the existing document similarity algorithms such as LDA, TF-IDF, and Bibliographic Coupling.
Lecture notes in computer scienceMinimal Jumping Emerging Patterns: Computation and Practical Assessment
10 Citations2015Bamba Kane, Bertrand Cuissart +1 more
An efficient method to mine the whole set of the minimal JEPs is detailed and fully proven and it is shown that this leads to the discovery of new valuable pieces of information.
Lecture notes in computer scienceAn Empirical Study of Personal Factors and Social Effects on Rating Prediction
10 Citations2015Zhijin Wang, Yan Yang +2 more
This paper proposes a novel recommendation approach, namely Personal factors with Weighted Social effects Matrix Factorization (PWS), which utilities both user’s taste and social effects to provide recommendations.
Lecture notes in computer scienceCentroid-Means-Embedding: An Approach to Infusing Word Embeddings into Features for Text Classification
10 Citations2015Mohammad Golam Sohrab, Makoto Miwa +1 more
Experimental results show that the proposed CME-based SAS-VSM approaches are promising over the different weighting approaches on the 20 Newsgroups and RCV1-v2/LYRL2004 datasets using Support Vector Machine (SVM) classifiers to enhance the classification tasks.
Lecture notes in computer scienceIterative Use of Weighted Voronoi Diagrams to Improve Scalability in Recommender Systems
9 Citations2015Joydeep Das, Subhashis Majumder +2 more
A decomposition based Recommendation Algorithm using Multiplicatively Weighted Voronoi Diagrams is proposed to reduce the running time without compromising the recommendation quality much, allowing the Recommender Systems to tackle bigger datasets using the same resources.
Lecture notes in computer scienceA New Feature Sampling Method in Random Forests for Predicting High-Dimensional Data
9 Citations2015Thanh-Tung Nguyen, He Zhao +3 more
A new approach for feature sampling for RF to deal with high-dimensional data maintains the diversity and the randomness of the forest and enables one to generate trees with a lower prediction error.
Lecture notes in computer scienceA Comparative Study on Parallel LDA Algorithms in MapReduce Framework
9 Citations2015Yang Gao, Zhenlong Sun +4 more
A novel multi-channel MapReduce framework is built to compare fairly three representative parallel LDA algorithms such as parallel variational Bayes (PVB), parallel Gibbs sampling (PGS) and parallel belief propagation (PBP), confirming that PGS yields the best application performance in search engine and online advertising system of Tencent.
Lecture notes in computer scienceIdentifying Hesitant and Interested Customers for Targeted Social Marketing
8 Citations2015Guowei Ma, Qi Liu +2 more
This paper proposes a general approach to figure out the targeted users for social marketing, taking both user interests and multiple social influences into consideration, and argues that these valuable users should have the best balanced influence entropy and utility scores.
Lecture notes in computer scienceMining Text Enriched Heterogeneous Citation Networks
7 Citations2015Jan Kralj, Anita Valmarska +2 more
An approach to mining text enriched heterogeneous information networks, applied to a task of categorizing papers from a large citation network of scientific publications in the field of psychology indicates that increasing the network size and combining both types of information is beneficial for improving the accuracy of paper categorization.
Lecture notes in computer scienceLink Prediction in Aligned Heterogeneous Networks
7 Citations2015Fangbing Liu, Shu‐Tao Xia
This paper attempts to solve the user-user link prediction problem for new users by utilizing data in a similar social network (source network) and proposes the Aligned Factor Graph (AFG) model, which works well when users leave little data in target network.
Lecture notes in computer scienceA Stochastic Framework for Solar Irradiance Forecasting Using Condition Random Field
7 Citations2015Jin Xu, Shinjae Yoo +5 more
A novel stochastic solar prediction framework using Conditional Random Fields is proposed, which utilizes features extracted from both cloud images taken by Total Sky Imagers and historical statistics to synergistically reduce the prediction error.
Lecture notes in computer scienceAutomated Detection for Probable Homologous Foodborne Disease Outbreaks
7 Citations2015Xiao Xiao, Yong Ge +5 more
The purpose of this paper is to provide approaches that can be applied to an online system to rapidly find local and sporadic foodborne disease outbreaks out of the collected data and propose a new approach named K-CPS (K-Means Clustering with Pattern Similarity) to detect sporadic outbreaks.
Lecture notes in computer scienceDistributed Document Representation for Document Classification
6 Citations2015Rumeng Li, Hiroyuki Shindo
This paper proposes a supervised framework (Compound RNN) for document classification based on document-level distributed representations learned from deep learning architecture that outperforms existing document representations such as bag-of-words, LDA in multiple text classification/regression tasks.
Lecture notes in computer sciencePrediciton of Emergency Events: A Multi-Task Multi-Label Learning Approach
6 Citations2015Budhaditya Saha, Sunil Gupta +1 more
The proposed model is formulated as an iterative optimization problems and solved using a scalable and efficient Block co-ordinate descent (BCD) method, and applies on two hospital cohorts - Cancer and Acute Myocardial Infarction patients collected over a two year period from a large hospital emergency department.
Lecture notes in computer scienceOnline Prediction of Chess Match Result
5 Citations2015Mohammad Mehedy Masud, Ameera Al-Shehhi +4 more
This work proposes a novel ensemble based learning technique where a profile-based segmentation is done on the training dataset, and one classifier is trained from each such segment, and the ensemble of classifiers is used to predict the outcome of new chess matches.
Lecture notes in computer scienceEfficient Discovery of Recurrent Routine Behaviours in Smart Meter Time Series by Growing Subsequences
5 Citations2015Jin Wang, Rachel Cardell‐Oliver +1 more
A novel algorithm is proposed to efficiently discover recurrent routine behaviours in smart meter time series by growing subsequences and is evaluated on synthetic data and demonstrated on a real-world dataset from the city of Kalgoorlie-Boulder in Western Australia.
Lecture notes in computer scienceImmune Centroids Over-Sampling Method for Multi-Class Classification
4 Citations2015Xusheng Ai, Jian Wu +4 more
This work utilizes an artificial immune network to generate synthetic examples on clusters with high data densities to address the problem of synthetic minority oversampling techniques, which lacks of the reflection on groups of training examples.
Lecture notes in computer scienceModeling User Interest and Community Interest in Microbloggings: An Integrated Approach
4 Citations2015Tuan-Anh Hoang
The CPI model is built based on the principle that a user’s personal interest is different from that of her community, and a regularization technique to bias the model to learn more socially meaningful topics for each community is developed.
Lecture notes in computer scienceMultiple Factors-Aware Diffusion in Social Networks
3 Citations2015Chung‐Kuang Chou, Ming-Syan Chen⋆
The aim of this work is to design a diffusion model in which factors considered are flexible to extend and change, and to propose a framework of learning parameters of the model, which is independent of factors considered.
Lecture notes in computer scienceGraph Based Relational Features for Collective Classification
3 Citations2015Immanuel Bayer, Uwe Nagel +1 more
This paper introduces new relational features for standard ML methods by extracting information from direct and indirect relations and shows empirically on three standard benchmark datasets that the authors' relational features yield results comparable to collective inference methods.
Lecture notes in computer scienceScale-Adaptive Group Optimization for Social Activity Planning
3 Citations2015Hong-Han Shuai, De-Nian Yang +2 more
A new randomized algorithm, named Budget-Aware Randomized Group Selection (BARGS), is designed to optimally allocate the computation budgets for effective selection of the group size and participants, and it is proved that BARGS can acquire the solution with a guaranteed performance bound.
Lecture notes in computer scienceOptimizing Classifiers for Hypothetical Scenarios
3 Citations2015R. A. Johnson, Troy Raeder +1 more
This work derives a relationship between two evaluation measures, H measure and cost curves, that may be used to address uncertainty in classifier performance and develops a boosting-based algorithm that directly mitigates classifier risk and outperforms AdaBoost on a diverse selection of datasets.
Lecture notes in computer scienceLearning Focused Hierarchical Topic Models with Semi-Supervision in Microblogs
2 Citations2015Anton Slutsky, Xiaohua Hu +1 more
The new Semi-Supervised Microblog-hLDA (SS-Micro-h LDA) model is proposed to discover topic hierarchies in short, noisy microblog documents in a way that allows users to focus topic discovery around interesting areas.
Lecture notes in computer sciencePairwise One Class Recommendation Algorithm
2 Citations2015Huimin Qiu, Chunhong Zhang +1 more
This work addresses the problem of one class recommendation for a special implicit feedback scenario, where training data only contain binary relevance data that indicate user’ selection or non-selection in social network with a pairwise rank-oriented model derived on basis of a rank-biased measure Mean Average Precision raised in Information Retrieval.
Lecture notes in computer scienceAn Incremental Local Distribution Network for Unsupervised Learning
2 Citations2015Youlu Xing, Tongyi Cao +3 more
An Incremental Local Distribution Network for unsupervised learning, which combines the merits of matrix learning and incremental learning, has a statistics based merging mechanism and thus can obtain a precise and concise representation of the learning data called relaxation representa- tion.