Tensor Data Imputation by PARAFAC with Updated Chaotic Biases by Adam Optimizer

The big data pattern analysis suffers from incorrect responses due to missing data entries in the real world. Data collected for digital movie platforms like Netflix and intelligent transportation systems is Spatio-temporal data. Extracting the latent and explicit features from this data is a challenge. We present the high dimensional data imputation problem as a higher-order tensor decomposition. The regularized and biased PARAFAC decomposition is proposed to generate the missing data entries. The biases are created and updated by a chaotic exponential factor in Adam's optimization, which reduces the imputation error. This chaotic perturbed exponentially update in the learning rate replaces the fixed learning rate in the bias update by Adam optimization. The idea has experimented with Netflix and traffic datasets from Guangzhou, China.

[1]  Vipin Kumar,et al.  Parallel Multilevel k-way Partitioning Scheme for Irregular Graphs , 1996, Proceedings of the 1996 ACM/IEEE Conference on Supercomputing.

[2]  Christopher J. Hillar,et al.  Most Tensor Problems Are NP-Hard , 2009, JACM.

[3]  Yehuda Koren,et al.  Matrix Factorization Techniques for Recommender Systems , 2009, Computer.

[4]  Ken-ichi Kawarabayashi,et al.  Expected Tensor Decomposition with Stochastic Gradient Descent , 2016, AAAI.

[5]  Peter D. Jarvis,et al.  Tensor Rank, Invariants, Inequalities, and Applications , 2012, SIAM J. Matrix Anal. Appl..

[6]  Ambuj K. Singh,et al.  GPOP: Scalable Group-level Popularity Prediction for Online Content in Social Networks , 2017, WWW.

[7]  H. Kiers,et al.  Selecting among three-mode principal component models of different types and complexities: a numerical convex hull based method. , 2006, The British journal of mathematical and statistical psychology.

[8]  Yifan Hu,et al.  Collaborative Filtering for Implicit Feedback Datasets , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[9]  Xinyu Chen,et al.  Spatial-temporal traffic speed patterns discovery and incomplete data recovery via SVD-combined tensor decomposition , 2018 .

[10]  Hong-Yuan Mark Liao,et al.  Simultaneous Tensor Decomposition and Completion Using Factor Priors , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Stefan Stieglitz,et al.  Social media analytics - Challenges in topic discovery, data collection, and data preparation , 2018, Int. J. Inf. Manag..

[12]  Yongdong Zhang,et al.  Sequential Prediction of Social Media Popularity with Deep Temporal Context Networks , 2017, IJCAI.

[13]  Thomas Demeester,et al.  Modeling and predicting the popularity of online news based on temporal and content-related features , 2017, Multimedia Tools and Applications.

[14]  Yehuda Koren,et al.  Collaborative filtering with temporal dynamics , 2009, KDD.

[15]  Wannes Meert,et al.  Predicting the popularity of online articles with random forests , 2014 .

[16]  Ryutaro Ichise,et al.  Predicting the Popularity of Social Curation , 2014, KSE.

[17]  Mohammed Shamsul Alam,et al.  Predicting the popularity of online news from content metadata , 2016, 2016 International Conference on Innovations in Science, Engineering and Technology (ICISET).

[18]  Mohammed Erritali,et al.  Analyzing Social Media through Big Data using InfoSphere BigInsights and Apache Flume , 2017, EUSPN/ICTH.

[19]  KarypisGeorge,et al.  Multilevelk-way Partitioning Scheme for Irregular Graphs , 1998 .

[20]  Paulo Rita,et al.  Predicting social media performance metrics and evaluation of the impact on brand building: A data mining approach , 2016 .

[21]  P. Paatero.  Construction and analysis of degenerate PARAFAC models , 2000 .

[22]  Masoud Makrehchi,et al.  Mining Social Media Content for Crime Prediction , 2016, 2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI).

[23]  Lieven De Lathauwer,et al.  Optimization-Based Algorithms for Tensor Decompositions: Canonical Polyadic Decomposition, Decomposition in Rank-(Lr, Lr, 1) Terms, and a New Generalization , 2013, SIAM J. Optim..

[24]  Lijun Sun,et al.  A Bayesian tensor decomposition approach for spatiotemporal traffic data imputation , 2019, Transportation Research Part C: Emerging Technologies.

[25]  S.T. Barnard,et al.  PMRSB: Parallel Multilevel Recursive Spectral Bisection , 1995, Proceedings of the IEEE/ACM SC95 Conference.

[26]  Avik Bhattacharya,et al.  Predicting the popularity of instagram posts for a lifestyle magazine using deep learning , 2017, 2017 2nd International Conference on Communication Systems, Computing and IT Applications (CSCITA).

[27]  Yong Jae Lee,et al.  Who Will Share My Image?: Predicting the Content Diffusion Path in Online Social Networks , 2018, WSDM.

[28]  Jiawei Wang,et al.  Missing traffic data imputation and pattern discovery with a Bayesian augmented tensor factorization model , 2019, Transportation Research Part C: Emerging Technologies.

[29]  Luis E. Ortiz,et al.  Chic or Social: Visual Popularity Analysis in Online Fashion Networks , 2014, ACM Multimedia.

[30]  Donald Goldfarb,et al.  Robust Low-Rank Tensor Recovery: Models and Algorithms , 2013, SIAM J. Matrix Anal. Appl..

[31]  Liqing Zhang,et al.  Bayesian CP Factorization of Incomplete Tensors with Automatic Rank Determination , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[32]  Radu State,et al.  VecHGrad for solving accurately complex tensor decomposition , 2019, ArXiv.

[33]  Changjun Hu,et al.  Predicting the popularity of viral topics based on time series forecasting , 2016, Neurocomputing.

[34]  Ting Lie,et al.  Advances in Intelligent Systems and Computing , 2014 .

[35]  Dan Cosley,et al.  Predictability of Popularity: Gaps between Prediction and Understanding , 2016, ICWSM.

[36]  Brandon Victor Syiem,et al.  Popularity Analysis on Social Network: A Big Data Analysis , 2015 .