SOFIANE ABBAR, HABIBUR RAHMAN, SARAVANA N THIRUMURUGANATHAN, CARLOS CASTILLO, G AUTAM DAS QATAR COMPUTING RESEARCH INSTITUTE UNIVERSITY OF TEXAS AT ARLINGTON.

Slides:



Advertisements
Similar presentations
Statistics for Improving the Efficiency of Public Administration Daniel Peña Universidad Carlos III Madrid, Spain NTTS 2009 Brussels.
Advertisements

Nonnegative Matrix Factorization with Sparseness Constraints S. Race MA591R.
Principal Component Analysis Based on L1-Norm Maximization Nojun Kwak IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008.
Active Learning for Streaming Networked Data Zhilin Yang, Jie Tang, Yutao Zhang Computer Science Department, Tsinghua University.
Fast Algorithms For Hierarchical Range Histogram Constructions
Graph Laplacian Regularization for Large-Scale Semidefinite Programming Kilian Weinberger et al. NIPS 2006 presented by Aggeliki Tsoli.
Second order cone programming approaches for handing missing and uncertain data P. K. Shivaswamy, C. Bhattacharyya and A. J. Smola Discussion led by Qi.
Active Learning and Collaborative Filtering
CUSTOMER NEEDS ELICITATION FOR PRODUCT CUSTOMIZATION Yue Wang Advisor: Prof. Tseng Advanced Manufacturing Institute Hong Kong University of Science and.
10/11/2001Random walks and spectral segmentation1 CSE 291 Fall 2001 Marina Meila and Jianbo Shi: Learning Segmentation by Random Walks/A Random Walks View.
Bayesian Robust Principal Component Analysis Presenter: Raghu Ranganathan ECE / CMR Tennessee Technological University January 21, 2011 Reading Group (Xinghao.
A Generalized Model for Financial Time Series Representation and Prediction Author: Depei Bao Presenter: Liao Shu Acknowledgement: Some figures in this.
Turning Privacy Leaks into Floods: Surreptitious Discovery of Social Network Friendships Michael T. Goodrich Univ. of California, Irvine joint w/ Arthur.
1 Learning Entity Specific Models Stefan Niculescu Carnegie Mellon University November, 2003.
Clustering In Large Graphs And Matrices Petros Drineas, Alan Frieze, Ravi Kannan, Santosh Vempala, V. Vinay Presented by Eric Anderson.
Computing Sketches of Matrices Efficiently & (Privacy Preserving) Data Mining Petros Drineas Rensselaer Polytechnic Institute (joint.
Preference Analysis Joachim Giesen and Eva Schuberth May 24, 2006.
Topic-Sensitive PageRank Taher H. Haveliwala. PageRank Importance is propagated A global ranking vector is pre-computed.
Collaborative Filtering Matrix Factorization Approach
Item-based Collaborative Filtering Recommendation Algorithms
Manifold learning: Locally Linear Embedding Jieping Ye Department of Computer Science and Engineering Arizona State University
Approximating the Algebraic Solution of Systems of Interval Linear Equations with Use of Neural Networks Nguyen Hoang Viet Michal Kleiber Institute of.
Cao et al. ICML 2010 Presented by Danushka Bollegala.
Efficient Model Selection for Support Vector Machines
Performance of Recommender Algorithms on Top-N Recommendation Tasks RecSys 2010 Intelligent Database Systems Lab. School of Computer Science & Engineering.
Distributed Networks & Systems Lab. Introduction Collaborative filtering Characteristics and challenges Memory-based CF Model-based CF Hybrid CF Recent.
The Multiplicative Weights Update Method Based on Arora, Hazan & Kale (2005) Mashor Housh Oded Cats Advanced simulation methods Prof. Rubinstein.
Focused Matrix Factorization for Audience Selection in Display Advertising BHARGAV KANAGAL, AMR AHMED, SANDEEP PANDEY, VANJA JOSIFOVSKI, LLUIS GARCIA-PUEYO,
Non Negative Matrix Factorization
Bayesian Sets Zoubin Ghahramani and Kathertine A. Heller NIPS 2005 Presented by Qi An Mar. 17 th, 2006.
Trust-Aware Optimal Crowdsourcing With Budget Constraint Xiangyang Liu 1, He He 2, and John S. Baras 1 1 Institute for Systems Research and Department.
EMIS 8381 – Spring Netflix and Your Next Movie Night Nonlinear Programming Ron Andrews EMIS 8381.
Chengjie Sun,Lei Lin, Yuan Chen, Bingquan Liu Harbin Institute of Technology School of Computer Science and Technology 1 19/11/ :09 PM.
Monte Carlo Methods Versatile methods for analyzing the behavior of some activity, plan or process that involves uncertainty.
Online Learning for Collaborative Filtering
Designing Example Critiquing Interaction Boi Faltings Pearl Pu Marc Torrens Paolo Viappiani IUI 2004, Madeira, Portugal – Wed Jan 14, 2004 LIAHCI.
CS 782 – Machine Learning Lecture 4 Linear Models for Classification  Probabilistic generative models  Probabilistic discriminative models.
Exploiting Context Analysis for Combining Multiple Entity Resolution Systems -Ramu Bandaru Zhaoqi Chen Dmitri V.kalashnikov Sharad Mehrotra.
Tanja Magoč, François Modave, Xiaojing Wang, and Martine Ceberio Computer Science Department The University of Texas at El Paso.
Distributed Nonnegative Matrix Factorization for Web- Scale Dyadic Data Analysis on MapReduce Challenge : the scalability of available tools Definition.
Graph-based Text Classification: Learn from Your Neighbors Ralitsa Angelova , Gerhard Weikum : Max Planck Institute for Informatics Stuhlsatzenhausweg.
Recommender Systems Debapriyo Majumdar Information Retrieval – Spring 2015 Indian Statistical Institute Kolkata Credits to Bing Liu (UIC) and Angshul Majumdar.
Exploiting Group Recommendation Functions for Flexible Preferences.
Pairwise Preference Regression for Cold-start Recommendation Speaker: Yuanshuai Sun
8.4.2 Quantum process tomography 8.5 Limitations of the quantum operations formalism 量子輪講 2003 年 10 月 16 日 担当:徳本 晋
ICDCS 2014 Madrid, Spain 30 June-3 July 2014
CoNMF: Exploiting User Comments for Clustering Web2.0 Items Presenter: He Xiangnan 28 June School of Computing National.
De novo discovery of mutated driver pathways in cancer Discussion leader: Matthew Bernstein Scribe: Kun-Chieh Wang Computational Network Biology BMI 826/Computer.
NONNEGATIVE MATRIX FACTORIZATION WITH MATRIX EXPONENTIATION Siwei Lyu ICASSP 2010 Presenter : 張庭豪.
Large-Scale Matrix Factorization with Missing Data under Additional Constraints Kaushik Mitra University of Maryland, College Park, MD Sameer Sheoreyy.
FISM: Factored Item Similarity Models for Top-N Recommender Systems
GeoMF: Joint Geographical Modeling and Matrix Factorization for Point-of-Interest Recommendation Defu Lian, Cong Zhao, Xing Xie, Guangzhong Sun, EnhongChen,
Matrix Factorization and its applications By Zachary 16 th Nov, 2010.
Adaptive Multi-view Clustering via Cross Trace Lasso
Collaborative Filtering via Euclidean Embedding M. Khoshneshin and W. Street Proc. of ACM RecSys, pp , 2010.
STATIC ANALYSIS OF UNCERTAIN STRUCTURES USING INTERVAL EIGENVALUE DECOMPOSITION Mehdi Modares Tufts University Robert L. Mullen Case Western Reserve University.
Identifying “Best Bet” Web Search Results by Mining Past User Behavior Author: Eugene Agichtein, Zijian Zheng (Microsoft Research) Source: KDD2006 Reporter:
A Kernel Approach for Learning From Almost Orthogonal Pattern * CIS 525 Class Presentation Professor: Slobodan Vucetic Presenter: Yilian Qin * B. Scholkopf.
Collaborative Deep Learning for Recommender Systems
1 Dongheng Sun 04/26/2011 Learning with Matrix Factorizations By Nathan Srebro.
Ranking: Compare, Don’t Score Ammar Ammar, Devavrat Shah (LIDS – MIT) Poster ( No preprint), WIDS 2011.
Learning to Align: a Statistical Approach
Chapter 7. Classification and Prediction
Singular Value Decomposition, and Application to Recommender Systems
Computing and Compressive Sensing in Wireless Sensor Networks
Multimodal Learning with Deep Boltzmann Machines
Machine Learning Basics
Clustering (3) Center-based algorithms Fuzzy k-means
Collaborative Filtering Matrix Factorization Approach
The Communication Complexity of Distributed Set-Joins
Presentation transcript:

SOFIANE ABBAR, HABIBUR RAHMAN, SARAVANA N THIRUMURUGANATHAN, CARLOS CASTILLO, G AUTAM DAS QATAR COMPUTING RESEARCH INSTITUTE UNIVERSITY OF TEXAS AT ARLINGTON Ranking Item Features by Mining Online User-Item Interactions

Outline Introduction Motivation and Challenge Model and Extensions Experimental Evaluation Related Works Conclusion

Business owners relies on user's feedback for the success of their businesses. It is important for them to understand what are the features which makes an item popular. User's put feedback on items in the form of reviews, tags, likes or +1's etc. Can we leverage this information to and the ranking of features in an item ? Can we and the global ranking or popularity of the features?

Introduction The main focus in this paper is the investigation of a novel-problem: how to rank the features of each item from user-item interactions. The principal problem investigated in this paper is stated as FEATURE RANKING(FR) PROBLEM: Where a set of features, and rudimentary user-item interactions (either at aggregate or individual level) is given, and how to identify the most important features per item (alternatively, a ranked list of features per item).

In this paper, the approach propose a probabilistic model that describes user-item interactions in terms of user preference distribution over feature and a feature-item transition matrix that determine the probability that an item will be chosen, given a feature.

This paper, used a database of items, where each item is des cribed by a set of attributes, some of which are multi valued. We refer to each of the distinct attribute values of an item as features(or equivalently, an item can be described as a set of features) Sparsity assumption. This paper assumes that among all the ℓ features available, each user expresses preference over a relatively small fraction of them:

Motivation For example Netflix, a simple user-item interaction would Involve whether the user watched the movie. While some users could have watched the movie because it starred Tom Hanks, others could have watched it because, in addition it was also directed by Steven Spielberg. Similarly, while some users might buy a car due to its manufacturer, others might buy it for the model and transmission type.

Example

Challenges

Models A ranking is a relationship between a set of items such that, for any two items, the first is either “ranked higher than”, “ranked lower than” or “ranked equal” ranking is the popularity of items features and suggesting popular item features.

Feature Ranking with Aggregate interaction information This model assumed that user u first picked a single feature j based on their individual preference vector h u and then selected an item i containing j with probability proportional to W ij

FR-AGG-W Algorithm: Input: Database D and aggregate visit vector v 1: W = Estimate feature item transition matrix 2: constraints = { ∀ i ∈ [1, n] hi ≥ 0, ||h||1= 1 } 3: h = argmin Error(v,Wh) subject to constraints h 4: Compute Xi = Wi· ◦ h ∀ i ∈ [1, n] 5: return X = {X1,X2,...,Xn}

FR-AGG-h Algorithm : Input: Database D and aggregate visit vector v 1: W = Estimate feature-item presence matrix 2: h = Estimate aggregate preference vector 3: constraints = { W ≤ W and ∀ j||W·j ||1= 1 and ∀ i, jWij ≥ 0 } 4: W = argmin Error(v,Wh) subject to constraints W 5: Compute Xi = Wi· ◦ h ∀ i ∈ [1, n] 6: return X = {X1,X2,...,Xn}

Variant Problem 1: (FR-AGG): Given a database D and Aggregate interaction Vector v, estimate the item-featurevis it vector X (where Xi=Wi·◦h) For each item I such that Error (v, W h) is minimized. Variant Problem 2 (FR-INDIV): Given a database D and ind ividual interaction matrix V, estimate the item-feature vi sit vector Xi for each item i (where Xi = Wi· ◦ h, is the average of columns of H) such that Error (V, W H) is minimized.

Network Flow In this, they consider a graph-based representation of t he problem that maps to the element. This algorithm finds feature to item transition matrix (W) by minimizing |V-Wh| error

Extensions Feature Ranking with Composite Features. Baselines. Algorithms - FR-AGG-W-LS - FR-AGG-h-LS - FR-AGG-h-NF Evaluation Metrics Ranking quality

Proposed method(FR-INDIV-MNMF) We choose Kullback-Leibler divergence D(V||W H) in order to measure the reconstruction error Between V and W H. This choice (instead of other Measures such as L2 distance) allows us to design an algorithm that preserves the column stochasticity constraints in the solution. In what follows, They propose a four-step algorithm to solve the problem of ranking item features in the presence of individual interaction matrix.

Step 1: Imposing sparsity constraints over H. They impose a (row) sparsity constraint over the factor W by assuming a sparse binary matrix W such that W ≤ W An entry(W)ij = 0 iff item I does not contain feature j A seemingly similar approach can be used to also impose (column) sparsity constraints over the Factor H by defining a sparse binary matrix H such that H ≤ H, where an entry (H) jk= 0 if user k has not visited any item that contains feature j However, this straightforward approach may not generate adequate sparsity constraints, since the union of distinct features of the items that a user has visited may be quite large

Step 2: Iterative algorithm with multiplicative update rules. In the second step, they propose modifications to the algorithm to discover factors W and H such that the Reconstruction error D (V ||W H)is minimized

Step 3: Imposing stochastic constraints on W and H The matrices W and H produced by Step 2 satisfy the sparsity requirements, however, they may not satisfy the col- umn stochastic constraints, which requires that the weights of each column of W and H sum to 1. In this step we describe a procedure for further modifying W and H such that the stochastic constraints are satisfied. We make use of the following theorem by Ho and Dooren

Step 4: Computing item-feature visit vectors Xi. Once the feature-item transition matrix W and individual preference matrix H are obtained, then the feature ranking of any Item can be computed as follows. First, compute the aggregate preference vector h by averaging all column-wise vectors H.j ∈, then perform a component wise multiplication between the item’s feature transition vector Wi. And h,i.e. Xi = Wi. ◦ h.

FR-INDIV-MNMF Algorithm: Input: Database D and individual interaction matrix V 1: W = Estimate feature-item presence matrix 2: H0 = Initialize a column-wise sparse individual preferen ce matrix using setCover (Step 1) 3: Compute W1, H1 = M-NMF(W, H0) (Step 2) 4: W, H = Impose stochastic constraints (Step 3) 5: Compute h = average (H) 6: Compute Xi = Wi oh ∀ I ∈ [1, n](Step 4) 7: return X ={X1, X2,..., Xn}

Experiment They conduct a comprehensive set of experiments to evaluate the effectiveness and efficiency of various Methods for ranking item features. The ranking quality measured within two scenarios: prediction of the most prominent feature and overall ranking of item features Dataset: MovieLens joint with cast data from IMDB

Result

Related Work Nonnegative Matrix Factorization (NMF) Attributes ranking Feature Ranking.

Conclusion In this paper, they consider the feature ranking problem that ranks features of an item by only considering user-item interaction information such as visits., defined two variants problem based on the granularity of the interaction information available and proposed different algorithms (based on constrained convex optimization, network flow approximation and marginal NMF) to solve these variants. In the future, they wish to investigate a variant where users can choose an item through a weighted combination of features.

Thank You