Splitting Approaches for Context-Aware Recommendation: An Empirical Study

Size: px
Start display at page:

Download "Splitting Approaches for Context-Aware Recommendation: An Empirical Study"

Transcription

1 Splitting Approaches for Context-Aware Recommendation: An Empirical Study Yong Zheng, Robin Burke, Bamshad Mobasher ACM SIGAPP the 29th Symposium On Applied Computing Gyeongju, South Korea, March 26, 2014

2 Contents ACM SIGAPP the 29th Symposium On Applied Computing Gyeongju, South Korea, March 26, 2014 Context-aware Recommender Systems Context-aware Splitting Approaches Empirical Study & Evaluation Results Discussions, Conclusions & Future work 2

3 Contents ACM SIGAPP the 29th Symposium On Applied Computing Gyeongju, South Korea, March 26, 2014 Context-aware Recommender Systems Context-aware Splitting Approaches Empirical Study & Evaluation Results Discussions, Conclusions & Future work 3

4 Recommender Systems Recommender Systems (RS) Two-dimension rating space: Users Items Ratings M1 M2 M3 U1 U2 U3 U4? 4

5 Context-aware Recommender Systems Context-aware Recommender Systems (CARS) Multi-dimensional space: Users Items Contexts Ratings 5

6 Context-aware Recommender Systems Context-aware Recommender Systems (CARS) Assumptions and Viewpoints in CARS: Users preferences or decisions usually differ from contexts to contexts, even towards the same item. E.g. buy a gift for someone. It s better to infer user s preferences by rating profiles within the same or similar contexts. E.g. look at music others choose within same contexts Context is defined as any information that can be used to characterize the situation of an entity by Dey, Anind K. (2001).However, the actual contexts in CARS and the contextual effects are domain specific. Movie domain: time, location, companion, mood, etc Music domain: time, activity, mood, etc Travel domain: season, weather, companion or trip type, etc 6

7 Context-aware Recommender Systems Context-aware Recommender Systems (CARS) Example of CARS applications: Tour Plan Recommender Yu, Chien-Chih and Chang, Hsiao-ping, "Towards Context-Aware Recommendation for Personalized Mobile Travel Planning". International Conference on Context-Aware Systems and Applications,

8 Context-aware Recommender Systems How to incorporate contexts into RS? There are two methods to categorize those incorporations. 1).In terms of how contexts interacted with the RS algorithms 8

9 Context-aware Recommender Systems How to incorporate contexts into RS? There are two methods to categorize those incorporations. 2).In terms of whether new CARS algorithms required to be developed It can be simply categorized into: a).transformation Algorithms A transformation is required, then all traditional RS algorithms can be applied to. Do NOT need to develop new CARS algorithms. such as Dimensions as Virtual Items (DaVI) and context-aware splitting approaches (CASA). b).adaptation Algorithms CARS algorithms are required, traditional algs can be modified. Such as context-aware matrix factorization (CAMF). 9

10 Contents ACM SIGAPP the 29th Symposium On Applied Computing Gyeongju, South Korea, March 26, 2014 Context-aware Recommender Systems Context-aware Splitting Approaches Empirical Study & Evaluation Results Discussions, Conclusions & Future work 10

11 Context-aware Splitting Approaches (CASA) Context-aware Splitting Approaches (CASA) In terms of the two categorizations (i.e. how to incorporate contexts into recommender systems) above, CASA belongs to pre-filtering and transformation algorithms. There are three context-aware splitting approaches: 1). Item splitting by L. Baltrunas, F. Ricci, ACM RecSys, ). User splitting by A. Said et al, RecSys, ). UI splitting by Y. Zheng et al, RecSys, 2013 User splitting and UI splitting are two approaches derived from Item splitting, examined and evaluated by different authors. 11

12 Context-aware Splitting Approaches (CASA) Item Splitting The underlying idea is that the nature of an item, from the user's point of view, may change in different contextual conditions, hence it may be useful to consider it as two different items. (L. Baltrunas, F. Ricci, RecSys'09) In short, contexts are dependent with items. At Cinema At Home At Swimming Pool Any dependent patterns involved in those ratings? 12

13 Context-aware Splitting Approaches (CASA) Item Splitting -- Example User Item Location Rating U1 M1 Pool 5 U2 M1 Pool 5 U3 M1 Pool 5 U1 M1 Home 2 U4 M1 Home 3 U2 M1 Home 2 High Rating Significant difference? Let s split it!!! Low Rating Same movie, different IDs. M11: being seen at Pool M12: being seen at Home M1 13

14 Context-aware Splitting Approaches (CASA) Item Splitting Step 1. Choose a contextual condition to split each item; The selection process is done by measuring significance of rating differences (such as the two-sample t test); Step 2. Contexts are fused to items and removed from original multidimensional matrix. We get a 2D rating matrix, then traditional algorithms like CF, MF can be applied to; How to select an appropriate contextual conditions for splitting? a). Binary contextual condition b). Impurity criteria and significance test See example in the next. 14

15 Context-aware Splitting Approaches (CASA) Item Splitting Binary Contextual Condition User Item Loc Rating User Item Rating U1 M1 Pool 5 U2 M1 Pool 5 U3 M1 Pool 5 U1 M1 Home 2 U4 M1 Home 3 U2 M1 Cinema 2 Transformation U1 M11 5 U2 M11 5 U3 M11 5 U1 M12 2 U4 M12 3 U2 M12 2 A binary contextual condition for splitting: Pool vs. Non-Pool If there is qualified split, one item will be split to two new ones. Why use a binary condition? To alleviate or avoid cold-start problems! 15

16 Context-aware Splitting Approaches (CASA) Item Splitting Impurity Criteria There could be several binary context conditions, for example, Pool vs Non-Pool, Home vs Non-Home, Weekend vs Non-Weekend. Impurity criteria and significance test are used to make the selection. There are 4 impurity criteria for splitting by L. Baltrunas, et al, RecSys'09; tmean (t-test), tprop (z-test), tchi (chi-square test), tig (Information gain) Take tmean for example, tmean, is defined using the two-sample t test and computes how significantly different are the means of the rating in the two rating subsets, when the split c (c is a context condition, e.g. location = Pool) is used. The bigger the t value of the test is, the more likely the difference of the means in the two partitions is significant (at 95% confidence value). Choose the largest one! 16

17 Context-aware Splitting Approaches (CASA) User Splitting and UI Splitting Similarly, the splitting approach can be applied to user too! User Splitting: is a similar one. Instead of splitting items, it may be useful to consider one user as two different users, if user demonstrates significantly different preferences across contexts. (A. Said et al., CARS@RecSys 2011) In short, contexts are dependent with items. UI Splitting: simply a combination of item splitting and user splitting both approaches are applied to create a new rating matrix new users and new items are created in the rating matrix. (Y. Zheng, et al, Decisions@ACM RecSys 2013). In short, it fuses dependent contexts to users and items simultaneously at the same time. 17

18 Context-aware Splitting Approaches (CASA) An Example of Three CASA After transformation: Item Splitting: User + NewItem; User Splitting: NewUser + Item; UI Splitting: NewUser + NewItem; UI Splitting fuses contexts to both users and items, where it may enlarge the contextual effects, but it also increases sparsity. It is hard to say whether UI splitting will outperform the other two algs or not. It varies from data to data. 18

19 Contents ACM SIGAPP the 29th Symposium On Applied Computing Gyeongju, South Korea, March 26, 2014 Context-aware Recommender Systems Context-aware Splitting Approaches Empirical Study & Evaluation Results Discussions, Conclusions & Future work 19

20 Empirical Study and Evaluations Experimental Goals a> Comparison Among Three Splitting Approaches Which one performs the best? Which splitting criteria is the best appropriate one? Any underlying patterns to indicate which one should be used? b> Comparison Between CASA and other Contextual Algorithms Which one performs the best? How about CASA competing with other CARS algorithms? 20

21 Empirical Study and Evaluations Data Sets Contextual variables in the three survey data sets: Food data: degree of hungriness in real and supposed situations; Movie data: Location (home/cinema), Time (weekend/weekday), Companion (family, etc); LDOS-CoMoDa: Location, Time, Companion, Weather, Emotions, Seasons, etc We use a 5-fold cross validation for all data sets and examined algorithms. 21

22 Empirical Study and Evaluations Baseline Algorithms We choose two other context-aware algorithms as the baselines: 1). Differential Context Modeling (DCM) by Y. Zheng, et al, 2012 There are two approaches falling into this category: Differential Context Relaxation (DCR) Differential Context Weighting (DCW) Basic idea: Using rating profiles with same or similar contexts for rating predictions; Take user-based collaborative filtering for example: Segment alg to various components; Apply context filter to each component; Filters could be different, and not necessary to be the same. Filter could be realized by context relaxation to find same contexts, or context weighting to find similar contexts. Generally, DCW works better than DCR. 22

23 Empirical Study and Evaluations Baseline Algorithms We choose two other context-aware algorithms as the baselines: 2). Context-aware Matrix Factorization (CAMF) by L. Baltrunas, et al, 2011 There are three approaches falling into this category: CAMF_C, CAMF_CI, CAMF_CU CAMF_C: Assume contextual effect is associated with each contextual condition only. CAMF_CI: Assume contextual effect is associated with item-context interactions. CAMF_CU: Assume contextual effect is associated with user-context interactions. CAMF is a kind of contextual modeling approach, where context-aware splitting approaches are contextual pre-filtering approaches. Both of them take advantage of the dependency between contexts and users or items. 23

24 Empirical Study and Evaluations Context-aware Splitting Approaches (CASA) They are pre-filtering approaches. Any traditional recommendation algorithms can be applied to, after the original multi-dimensional rating matrix was transformed to a 2D rating matrix. We evaluate three CASA based on the configuration as follows: 1). Evaluated by different traditional RS algorithms User-based Collaborative Filtering (UBCF), Item-based Collaborative Filtering (IBCF) Traditional Matrix Factorization techniques (MF) without taking contexts into consideration Implemented and evaluated by open-source Toolkit MyMediaLite v3.07 CF algorithms were tuned up by varying # of neighbors; MF algorithms were examined by varying # of factors and training iterations; 2). Evaluated by different impurity criteria in splitting processes tmean (t-test), tprop (z-test), tchi (chi-square test), tig (Information gain) 24

25 Empirical Study and Evaluations Evaluation Metrics We choose three metrics: RMSE, Precision, ROC. RMSE is used to evaluate the accuracy of predicted ratings. Prediction error, is the most popular and common used metric in CARS area, since context-aware data are usually sparse and few users rated a same item for several times. ROC = a visualization between recall and FPR by varying # of N in Top-N recommendations. = x axis is FPT, y axis is Recall In measuring Precision and ROC, we use a rating threshold to judge relevance. For Movie data, the threshold is set as 7, and it is set as 3 for the other two data sets. 25

26 Empirical Study and Evaluations Evaluation Metrics Traditional way to measure Precision and ROC: 1).We have training and testing set; 2).Train a model based on the training set and evaluate it on the testing set; 3).In the testing set, aggregate rating profiles by <user, a list of items he/she rated>; 4).For evaluation purpose, provide a list of ranked Top-N items to each user; 5).And examine the hit ratio between the Top-N list and the list of items rated in the testing; However, in CARS, contexts should be taken into account; CPrecision and CROC curve 1).We have training and testing set; 2).Train a model based on the training set and evaluate it on the testing set; 3).In the testing set, aggregate rating profiles by <user, a list of items he/she rated, contexts>; 4).For evaluation purpose, provide a list of ranked Top-N items to each <user, contexts>; 5).And examine the hit ratio between the Top-N list and the list of items rated in the testing; NOTICE: <user, a list of items he/she rated, contexts>; the list is pretty short and even just one item, because users seldom rated items for several times within different contexts. Thus the value of CPrecision and CROC will be much smaller than traditional ones. 26

27 Empirical Study and Evaluations Evaluation Challenge in CASA (Optional Part) User Item Loc Rating User Item Rating U1 M1 Pool 5 U2 M1 Pool 5 U3 M1 Pool 5 U1 M1 Home 2 U4 M1 Home 3 U2 M1 Cinema 2 Transformation U1 M11 5 U2 M11 5 U3 M11 5 U1 M12 2 U4 M12 3 U2 M12 2 RMSE can be directly evaluated based on the transformed rating matrix in CASA It is because the number of rating profiles in data is NOT changed. CPrecision and CROC cannot be directly evaluated on the transformed rating matrix 1). # of users and # of items could be DIFFERENT 2). It is not comparable to other CARS algorithms Solution: We only use transformed matrix to predict ratings, but evaluate IR metrics on the original multi-dimensional rating matrix. 27

28 Experimental Results Experimental Results (in RMSE) Goal-1: Comparisons among the three context-aware splitting approaches (in RMSE) Q: Which one performs the best? The best impurity criteria? A: UI Splitting using MF as the recommendation algorithm. MF works better than CFs. The best choice varies from data to data. No consistent patterns. Q: Any other patterns? A: For Movie data, item splitting is better than user splitting; But user splitting is better than item splitting for the other two ones, where they have emotional or feeling contextual variables, we assume those contexts are more dependent with users. 28

29 Experimental Results Experimental Results (in RMSE) Goal-2: Comparisons with other CARS algorithms (in terms of RMSE) Q: Which one performs the best? The best impurity criteria? A: UI Splitting using MF as the recommendation algorithm in terms of RMSE. Q: Any other patterns? A: If item splitting is better than user splitting, CAMF_CI is better than CAMF_CU; If user splitting is better than item splitting, then CAMF_CU is better than CAMF_CI; It is because both of them take advantage of context-dependency patterns!! 29

30 Experimental Results Experimental Results (CPrecision & CROC) Goal-1: Comparisons among the three context-aware splitting approaches In CPrecision, UI splitting > Item splitting > User Splitting; In ROC Curve, UI splitting > User splitting > Item Splitting; Goal-2: Comparisons with other CARS algorithms In CPrecision, UI splitting > CAMF_CI > CAMF_CU > DCW > DCR; In ROC Curve, UI splitting > CAMF_CU > CAMF_CI > DCW > DCR; Patterns: UI Splitting is the best in RMSE and IR metrics for LDOS-CoMoDa; Consistent findings in context-dependency pattern in EACH METRIC; In RMSE, context is more dependent with user; 30

31 Experimental Results Experimental Results (CPrecision & CROC) Goal-1: Comparisons among the three context-aware splitting approaches In CPrecision, Item splitting > UI splitting > User splitting; In ROC Curve, same patter as above; Goal-2: Comparisons with other CARS algorithms In CPrecision, Item splitting > UI splitting > CAMF_CI > CAMF_CU > DCW; In ROC Curve, Item splitting > UI splitting > CAMF_CI > DCW > CAMF_CU; Patterns: Item Splitting is the best in RMSE and IR metrics for Movie data; Consistent findings in context-dependency pattern in EACH METRIC; 31

32 Experimental Results Experimental Results (CPrecision & CROC) Goal-1: Comparisons among the three context-aware splitting approaches In CPrecision, UI splitting > User splitting > Item Splitting; In ROC Curve, same pattern as above; Goal-2: Comparisons with other CARS algorithms In CPrecision, UI splitting > CAMF_CU > CAMF_CI > DCR > DCW; In ROC Curve, DCR > UI splitting > DCW > CAMF_CU > CAMF_CI Patterns: Overall, UI Splitting is the best in RMSE and IR metrics for Food Data; Consistent findings in context-dependency pattern; 32

33 Contents ACM SIGAPP the 29th Symposium On Applied Computing Gyeongju, South Korea, March 26, 2014 Context-aware Recommender Systems Context-aware Splitting Approaches Empirical Study & Evaluation Results Discussions, Conclusions & Future work 33

34 Conclusions Conclusions & Future Work Which one performs the best? Generally speaking, UI splitting is the best; In Movie data, UI splitting is the best on RMSE, but item splitting is the best on IR metrics; If context is not that dependent with users, merging effects by UI splitting may decrease the joint effect on recommendations. Any patterns or guidelines to choose which context-aware algorithms? In terms of choices between item splitting & user splitting, and CAMF_CI & CAMF_CU, it totally depends on which one contexts are more dependent to, user or item? Whether UI splitting performs the best depends on three factors: 1). The dependency between contexts and users and items; 2). The sparsity after rating matrix transformation cold-start problems in CASA; 3). The performance difference between user splitting and item splitting. If one of them performs bad, it is not guaranteed that the joint effect UI splitting will perform better; Future work: 1).how to judge contexts are more dependent with users or items? Any numeric metrics to validate it? PS: Impurity values? no consistent patterns. 2). How to alleviate the cold-start problems in UI splitting. 34

35 Conclusions References Item Splitting 1) L. Baltrunas, and F. Ricci. "Context-based splitting of item ratings in collaborative filtering." ACM RecSys, ) L. Baltrunas, and F. Ricci. "Experimental evaluation of context-dependent collaborative filtering using item splitting." User Modeling and User-Adapted Interaction (2013): User Splitting 1) L. Batrunas and X. Amatriain."Towards Time-Dependent Recommendation Based on Implicit Feedback." ) A. Said, E. Luca, S. Albayrak. "Inferring contextual user profiles improving recommender performance UI Splitting 1) Y. Zheng, R. Burke, B. Mobasher. "The Role of Emotions in Context-aware Recommendation" ) Y. Zheng, R. Burke, B. Mobasher, Splitting Approaches for Context-Aware Recommendation: An Empirical Study, ACM SAC, 2014 Context-aware Matrix Factorization 1) L. Baltrunas, B. Ludwig, F. Ricci. "Matrix factorization techniques for context aware recommendation." ACM RecSys Differential Context Modeling 1) Y. Zheng, R. Burke, B. Mobasher. "Differential Context Relaxation for Context-aware Travel Recommendation". EC-WEB, ) Y. Zheng, R. Burke, B. Mobasher. "Optimal Feature Selection for Context-Aware Recommendation using Differential Relaxation" ) Y. Zheng, R. Burke, B. Mobasher. "Recommendation with Differential Context Weighting". In UMAP,

36 ACM SIGAPP the 29th Symposium On Applied Computing Gyeongju, South Korea, March 26, 2014

Context-aware recommendation

Context-aware recommendation Context-aware recommendation Eirini Kolomvrezou, Hendrik Heuer Special Course in Computer and Information Science User Modelling & Recommender Systems Aalto University Context-aware recommendation 2 Recommendation

More information

Experimental evaluation of context-dependent collaborative filtering using item splitting

Experimental evaluation of context-dependent collaborative filtering using item splitting User Model User-Adap Inter (2014) 24:7 34 DOI 10.1007/s11257-012-9137-9 ORIGINAL PAPER Experimental evaluation of context-dependent collaborative filtering using item splitting Linas Baltrunas Francesco

More information

Context-Aware Movie Recommendations: An Empirical Comparison of Pre-Filtering, Post-Filtering and Contextual Modeling Approaches

Context-Aware Movie Recommendations: An Empirical Comparison of Pre-Filtering, Post-Filtering and Contextual Modeling Approaches Context-Aware Movie Recommendations: An Empirical Comparison of Pre-Filtering, Post-Filtering and Contextual Modeling Approaches Pedro G. Campos 1,2, Ignacio Fernández-Tobías 2, Iván Cantador 2, Fernando

More information

Recommending hotels based on multi-dimensional customer ratings

Recommending hotels based on multi-dimensional customer ratings Recommending hotels based on multi-dimensional customer ratings Dietmar Jannach a, Fatih Gedikli a, Zeynep Karakaya a, and Oliver Juwig b a Fakultät für Informatik TU Dortmund, Germany {firstname.lastname}@tu-dortmund.de

More information

Prediction of Personalized Rating by Combining Bandwagon Effect and Social Group Opinion: using Hadoop-Spark Framework

Prediction of Personalized Rating by Combining Bandwagon Effect and Social Group Opinion: using Hadoop-Spark Framework Prediction of Personalized Rating by Combining Bandwagon Effect and Social Group Opinion: using Hadoop-Spark Framework Lu Sun 1, Kiejin Park 2 and Limei Peng 1 1 Department of Industrial Engineering, Ajou

More information

A Simple but Effective Method to Incorporate Trusted Neighbors in Recommender Systems

A Simple but Effective Method to Incorporate Trusted Neighbors in Recommender Systems A Simple but Effective Method to Incorporate Trusted Neighbors in Recommender Systems Guibing Guo, Jie Zhang, and Daniel Thalmann School of Computer Engineering Nanyang Technological University, Singapore

More information

Neighbor selection for cold users in collaborative filtering with positive-only feedback

Neighbor selection for cold users in collaborative filtering with positive-only feedback Neighbor selection for cold users in collaborative filtering with positive-only feedback Alejandro Bellogín 1, Ignacio Fernández-Tobías 2, Iván Cantador 1, and Paolo Tomeo 3 1 Universidad Autónoma de Madrid,

More information

Model Selection, Evaluation, Diagnosis

Model Selection, Evaluation, Diagnosis Model Selection, Evaluation, Diagnosis INFO-4604, Applied Machine Learning University of Colorado Boulder October 31 November 2, 2017 Prof. Michael Paul Today How do you estimate how well your classifier

More information

Predicting user rating for Yelp businesses leveraging user similarity

Predicting user rating for Yelp businesses leveraging user similarity Predicting user rating for Yelp businesses leveraging user similarity Kritika Singh kritika@eng.ucsd.edu Abstract Users visit a Yelp business, such as a restaurant, based on its overall rating and often

More information

arxiv: v1 [cs.ir] 22 Jan 2019

arxiv: v1 [cs.ir] 22 Jan 2019 Managing Popularity Bias in Recommender Systems with Personalized Re-ranking Himan Abdollahpouri, Robin Burke and 2 Bamshad Mobasher That Recommender Systems Lab,Department of Information Science, University

More information

SOCIAL MEDIA MINING. Behavior Analytics

SOCIAL MEDIA MINING. Behavior Analytics SOCIAL MEDIA MINING Behavior Analytics Dear instructors/users of these slides: Please feel free to include these slides in your own material, or modify them as you see fit. If you decide to incorporate

More information

RECOMMENDER SYSTEM IN RETAIL

RECOMMENDER SYSTEM IN RETAIL RECOMMENDER SYSTEM IN RETAIL 2015 Shoppers Drug Mart. All rights reserved. Unauthorized duplication or distribution in whole or in part via any channel without written permission strictly prohibited. TRADITIONAL

More information

MATRIX FACTORIZATION TECHNIQUES FOR RECOMMENDER SYSTEMS. Yehuda Koren, Yahoo Research Robert Bell and Chris Volinsky, AT&T Labs - Research

MATRIX FACTORIZATION TECHNIQUES FOR RECOMMENDER SYSTEMS. Yehuda Koren, Yahoo Research Robert Bell and Chris Volinsky, AT&T Labs - Research MATRIX FACTORIZATION TECHNIQUES FOR RECOMMENDER SYSTEMS Yehuda Koren, Yahoo Research Robert Bell and Chris Volinsky, AT&T Labs - Research Recommender System Strategies Recommender System Strategies Content

More information

A Novel Approach to Trust Based Recommender Systems Leveraged by Context Attributes

A Novel Approach to Trust Based Recommender Systems Leveraged by Context Attributes A Novel Approach to Trust Based Recommender Systems Leveraged by Context Attributes PallabDutta #1, Dr. A. Kumaravel *2 #1 Research Scholar, Department of CSE, Bharath University, Chennai, India pallabd14@yahoo.in

More information

Model of Personal Discount Sensitivity in Recommender Systems

Model of Personal Discount Sensitivity in Recommender Systems Model of Personal Discount Sensitivity in Recommender Systems Masahiro Sato, Hidetaka Izumo, Takashi Sonoda Fuji Xerox Co., Ltd. 6-1 Minatomirai, Nishi-ku, Yokohama, Kanagawa, Japan {sato.masahiro, izumo.hidetaka,

More information

Web Usage Mining. Recommender Systems. Web Log Mining. Introduction Memory-Based Recommender Systems Model-Based Recommender Systems. 1 J.

Web Usage Mining. Recommender Systems. Web Log Mining. Introduction Memory-Based Recommender Systems Model-Based Recommender Systems. 1 J. Web Usage Mining Recommender Systems Introduction Memory-Based Recommender Systems Model-Based Recommender Systems Web Log Mining 1 J. Fürnkranz Recommender Systems Scenario: Users have a potential interest

More information

Using Decision Tree to predict repeat customers

Using Decision Tree to predict repeat customers Using Decision Tree to predict repeat customers Jia En Nicholette Li Jing Rong Lim Abstract We focus on using feature engineering and decision trees to perform classification and feature selection on the

More information

Yelp Recommendation System Using Advanced Collaborative Filtering

Yelp Recommendation System Using Advanced Collaborative Filtering Yelp Recommendation System Using Advanced Collaborative Filtering 1. INTRODUCTION Chee Hoon Ha Stanford Univerisy cheehoon@stanford.edu Thanks to the advancement in technology, we live in a world where

More information

Collaborative Filtering in Social Networks

Collaborative Filtering in Social Networks THE FACULTIES OF ENGINEERING, SCIENCE AND MEDICINE - AALBORG UNIVERSITY DEPARTMENT OF COMPUTER SCIENCE AALBORG UNIVERSITY DE4 GROUP D624A Collaborative Filtering in Social Networks Similarity analysis

More information

A Fine-Grained Analysis on the Evolutionary Coupling of Cloned Code

A Fine-Grained Analysis on the Evolutionary Coupling of Cloned Code A Fine-Grained Analysis on the Evolutionary Coupling of Cloned Code Manishankar Mondal Chanchal K. Roy Kevin A. Schneider Software Research Laboratory, Department of Computer Science, University of Saskatchewan,

More information

Research Article Research on E-Commerce Platform-Based Personalized Recommendation Algorithm

Research Article Research on E-Commerce Platform-Based Personalized Recommendation Algorithm Applied Computational Intelligence and So Computing Volume 2016, Article ID 5160460, 7 pages http://dx.doi.org/.1155/2016/5160460 Research Article Research on E-Commerce Platform-Based Personalized Recommendation

More information

An Analysis Framework for Content-based Job Recommendation. Author(s) Guo, Xingsheng; Jerbi, Houssem; O'Mahony, Michael P.

An Analysis Framework for Content-based Job Recommendation. Author(s) Guo, Xingsheng; Jerbi, Houssem; O'Mahony, Michael P. Provided by the author(s) and University College Dublin Library in accordance with publisher policies. Please cite the published version when available. Title An Analysis Framework for Content-based Job

More information

HybridRank: Ranking in the Twitter Hybrid Networks

HybridRank: Ranking in the Twitter Hybrid Networks HybridRank: Ranking in the Twitter Hybrid Networks Jianyu Li Department of Computer Science University of Maryland, College Park jli@cs.umd.edu ABSTRACT User influence in social media may depend on multiple

More information

Context-Aware Recommender Systems: A Comparison Of Three Approaches

Context-Aware Recommender Systems: A Comparison Of Three Approaches Context-Aware Recommender Systems: A Comparison Of Three Approaches Umberto Panniello and Michele Gorgoglione Politecnico di Bari (Italy), Viale Japigia 182 Tel. +39 080 5962765 Fax: +39 080 5962766 u.panniello@poliba.it,

More information

Using Trust in Collaborative Filtering Recommendation

Using Trust in Collaborative Filtering Recommendation Using Trust in Collaborative Filtering Recommendation Chein-Shung Hwang and Yu-Pin Chen Department of Information Management, Chinese Culture University, 55, Hwa-Kang Road, Yang-Ming-Shan, Taipei, Taiwan,

More information

Evaluation of Recommendations: Rating-Prediction and Ranking

Evaluation of Recommendations: Rating-Prediction and Ranking Evaluation of Recommendations: Rating-Prediction and Ranking Harald Steck Netflix Inc. Los Gatos, California hsteck@netflix.com ABSTRACT The literature on recommender systems distinguishes typically between

More information

From Ratings to Trust: an Empirical Study of Implicit Trust in Recommender Systems

From Ratings to Trust: an Empirical Study of Implicit Trust in Recommender Systems From Ratings to Trust: an Empirical Study of Implicit Trust in Recommender Systems Guibing Guo 1, Jie Zhang 1, Daniel Thalmann 1, Anirban Basu 2, Neil Yorke-Smith 3 1 Nanyang Technological University,

More information

2 Maria Carolina Monard and Gustavo E. A. P. A. Batista

2 Maria Carolina Monard and Gustavo E. A. P. A. Batista Graphical Methods for Classifier Performance Evaluation Maria Carolina Monard and Gustavo E. A. P. A. Batista University of São Paulo USP Institute of Mathematics and Computer Science ICMC Department of

More information

Business Analytics & Data Mining Modeling Using R Dr. Gaurav Dixit Department of Management Studies Indian Institute of Technology, Roorkee

Business Analytics & Data Mining Modeling Using R Dr. Gaurav Dixit Department of Management Studies Indian Institute of Technology, Roorkee Business Analytics & Data Mining Modeling Using R Dr. Gaurav Dixit Department of Management Studies Indian Institute of Technology, Roorkee Lecture - 02 Data Mining Process Welcome to the lecture 2 of

More information

Part 15: Context Dependent Recommendations. Francesco Ricci Free University of Bozen-Bolzano Italy

Part 15: Context Dependent Recommendations. Francesco Ricci Free University of Bozen-Bolzano Italy Part 15: Context Dependent Recommendations Francesco Ricci Free University of Bozen-Bolzano Italy fricci@unibz.it Content p What is context? p Types of context p Context impact on recommendations and ratings

More information

Applying an Interactive Machine Learning Approach to Statutory Analysis

Applying an Interactive Machine Learning Approach to Statutory Analysis Applying an Interactive Machine Learning Approach to Statutory Analysis Jaromir Savelka a,b, Gaurav Trivedi a, Kevin D. Ashley a,b,c a Intelligent Systems Program, University of Pittsburgh, USA b Learning

More information

COMP 465: Data Mining Recommender Systems

COMP 465: Data Mining Recommender Systems //0 COMP : Data Mining Recommender Systems Slides Adapted From: www.mmds.org (Mining Massive Datasets) Customer X Buys Metallica CD Buys Megadeth CD Customer Y Does search on Metallica Recommender system

More information

Predicting Purchase Behavior of E-commerce Customer, One-stage or Two-stage?

Predicting Purchase Behavior of E-commerce Customer, One-stage or Two-stage? 2016 International Conference on Artificial Intelligence and Computer Science (AICS 2016) ISBN: 978-1-60595-411-0 Predicting Purchase Behavior of E-commerce Customer, One-stage or Two-stage? Chen CHEN

More information

THE internet has changed different aspects of our lives, job

THE internet has changed different aspects of our lives, job 1 Recommender Systems for IT Recruitment João Almeida and Luís Custódio Abstract Recruitment processes have increasingly become dependent on the internet. Companies post job opportunities on their websites

More information

Social Media Recommender Systems: Review and Open Research Issues

Social Media Recommender Systems: Review and Open Research Issues Received December 17, 2017, accepted February 13, 2018, date of publication February 27, 2018, date of current version April 4, 2018. Digital Object Identifier 10.1109/ACCESS.2018.2810062 Social Media

More information

TOWARD MORE DIVERSE RECOMMENDATIONS: ITEM RE-RANKING METHODS FOR RECOMMENDER SYSTEMS

TOWARD MORE DIVERSE RECOMMENDATIONS: ITEM RE-RANKING METHODS FOR RECOMMENDER SYSTEMS TOWARD MORE DIVERSE RECOMMENDATIONS: ITEM RE-RANKING METHODS FOR RECOMMENDER SYSTEMS Gediminas Adomavicius YoungOk Kwon Department of Information and Decision Sciences Carlson School of Management, University

More information

Netflix Optimization: A Confluence of Metrics, Algorithms, and Experimentation. CIKM 2013, UEO Workshop Caitlin Smallwood

Netflix Optimization: A Confluence of Metrics, Algorithms, and Experimentation. CIKM 2013, UEO Workshop Caitlin Smallwood Netflix Optimization: A Confluence of Metrics, Algorithms, and Experimentation CIKM 2013, UEO Workshop Caitlin Smallwood 1 Allegheny Monongahela Ohio River 2 TV & Movie Enjoyment Made Easy Stream any video

More information

Information Sciences

Information Sciences Information Sciences 179 (2009) 3505 3519 Contents lists available at ScienceDirect Information Sciences journal homepage: www.elsevier.com/locate/ins A hybrid of sequential rules and collaborative filtering

More information

Semantic ratings and heuristic similarity for collaborative filtering

Semantic ratings and heuristic similarity for collaborative filtering Semantic ratings and heuristic similarity for collaborative filtering Robin Burke Department of Information and Computer Science University of California, Irvine burke@ics.uci.edu Abstract Collaborative

More information

Insights on Social Recommender System

Insights on Social Recommender System Insights on Social Recommender System Wolney L. de Mello Neto Vrije Universiteit Brussel CoMo Lab Brussels, Belgium wdemello@vub.ac.be Ann Nowé Vrije Universiteit Brussel CoMo Lab Brussels, Belgium ann.nowe@vub.ac.be

More information

Credibility: Evaluating What s Been Learned

Credibility: Evaluating What s Been Learned Evaluation: the Key to Success Credibility: Evaluating What s Been Learned Chapter 5 of Data Mining How predictive is the model we learned? Accuracy on the training data is not a good indicator of performance

More information

Analytics for Banks. September 19, 2017

Analytics for Banks. September 19, 2017 Analytics for Banks September 19, 2017 Outline About AlgoAnalytics Problems we can solve for banks Our experience Technology Page 2 About AlgoAnalytics Analytics Consultancy Work at the intersection of

More information

Semantic ratings and heuristic similarity for collaborative filtering

Semantic ratings and heuristic similarity for collaborative filtering From: AAAI Technical Report WS-00-04. Compilation copyright 2000, AAAI (www.aaai.org). All rights reserved. Semantic ratings and heuristic similarity for collaborative filtering Robin Burke Department

More information

Cold-start Solution to Location-based Entity Shop. Recommender Systems Using Online Sales Records

Cold-start Solution to Location-based Entity Shop. Recommender Systems Using Online Sales Records Cold-start Solution to Location-based Entity Shop Recommender Systems Using Online Sales Records Yichen Yao 1, Zhongjie Li 2 1 Department of Engineering Mechanics, Tsinghua University, Beijing, China yaoyichen@aliyun.com

More information

Exploitation and Exploration in a Performance based Contextual Advertising System

Exploitation and Exploration in a Performance based Contextual Advertising System Exploitation and Exploration in a Performance based Contextual Advertising System Wei Li, Xuerui Wang, Ruofei Zhang, Ying Cui Jianchang Mao and Rong Jin Performance-based online advertising Performance

More information

Context-Aware Rank-Oriented Recommender Systems. Brian Ackerman

Context-Aware Rank-Oriented Recommender Systems. Brian Ackerman Context-Aware Rank-Oriented Recommender Systems by Brian Ackerman A Thesis Presented in Partial Fulfillment of the Requirements for the Degree Master of Science Approved June 2012 by the Graduate Supervisory

More information

Predicting Yelp Ratings From Business and User Characteristics

Predicting Yelp Ratings From Business and User Characteristics Predicting Yelp Ratings From Business and User Characteristics Jeff Han Justin Kuang Derek Lim Stanford University jeffhan@stanford.edu kuangj@stanford.edu limderek@stanford.edu I. Abstract With online

More information

Supply Chain MICROSOFT BUSINESS SOLUTIONS DEMAND PLANNER

Supply Chain MICROSOFT BUSINESS SOLUTIONS DEMAND PLANNER Supply Chain MICROSOFT BUSINESS SOLUTIONS DEMAND PLANNER DEMAND PLANNING FOR BUSINESSES Demand planning is the first step towards business planning. As businesses are moving towards a demand-centric environment

More information

Segmenting Customer Bases in Personalization Applications Using Direct Grouping and Micro-Targeting Approaches

Segmenting Customer Bases in Personalization Applications Using Direct Grouping and Micro-Targeting Approaches Segmenting Customer Bases in Personalization Applications Using Direct Grouping and Micro-Targeting Approaches Alexander Tuzhilin Stern School of Business New York University (joint work with Tianyi Jiang)

More information

Similarity and trust metrics used in Recommender Systems: A survey

Similarity and trust metrics used in Recommender Systems: A survey Similarity and trust metrics used in Recommender Systems: A survey Maryam Jallouli 1, Sonia Lajmi 1,2, Ikram Amous 1 1 MIRACL Laboratory, Technopole of Sfax, University of Sfax, P.O.Box 242, 3031 Sfax,

More information

Introduction to Recommendation Engines

Introduction to Recommendation Engines Introduction to Recommendation Engines A guide to algorithmically predicting what your customers want and when. By Tuck Ngun, PhD Introduction Recommendation engines have become a popular solution for

More information

A Social Influence based Trust Model for Recommender Systems

A Social Influence based Trust Model for Recommender Systems A Social Influence based Trust Model for Recommender Systems Jian-Ping Mei a,, Han Yu c, Zhiqi Shen b,c, Chunyan Miao b,c a College of Computer Science and Technology, Zhejiang University of Technology,

More information

Weka Evaluation: Assessing the performance

Weka Evaluation: Assessing the performance Weka Evaluation: Assessing the performance Lab3 (in- class): 21 NOV 2016, 13:00-15:00, CHOMSKY ACKNOWLEDGEMENTS: INFORMATION, EXAMPLES AND TASKS IN THIS LAB COME FROM SEVERAL WEB SOURCES. Learning objectives

More information

Multi-Dimensional Representations -- How Many Dimensions? Cluster Stack Visualization for Market Segmentation Analysis

Multi-Dimensional Representations -- How Many Dimensions? Cluster Stack Visualization for Market Segmentation Analysis Multi-Dimensional Representations -- How Many Dimensions? Cluster Stack Visualization for Market Segmentation Analysis William Wright Oculus Info Inc. An earlier version of this paper appeared in the Symposium

More information

Collaborative recommendation

Collaborative recommendation 2 Collaborative recommendation The main idea of collaborative recommendation approaches is to exploit information about the past behavior or the opinions of an existing user community for predicting which

More information

A Short Survey of Recommendation Technologies in Travel and Tourism

A Short Survey of Recommendation Technologies in Travel and Tourism ÖGAI Journal 25/2 1 A Short Survey of Recommendation Technologies in Travel and Tourism Abstract A. Felfernig, S. Gordea, D. Jannach, E. Teppan, M. Zanker University Klagenfurt Institut für Wirtschaftsinformatik

More information

Statistical Methods for Network Analysis of Biological Data

Statistical Methods for Network Analysis of Biological Data The Protein Interaction Workshop, 8 12 June 2015, IMS Statistical Methods for Network Analysis of Biological Data Minghua Deng, dengmh@pku.edu.cn School of Mathematical Sciences Center for Quantitative

More information

Text Analysis of American Airlines Customer Reviews

Text Analysis of American Airlines Customer Reviews SESUG 2016 Paper EPO-281 Text Analysis of American Airlines Customer Reviews Rajesh Tolety, Oklahoma State University Saurabh Kumar Choudhary, Oklahoma State University ABSTRACT Which airline should I

More information

An Effective Recommender System by Unifying User and Item Trust Information for B2B Applications

An Effective Recommender System by Unifying User and Item Trust Information for B2B Applications An Effective Recommender System by Unifying User and Item Trust Information for B2B Applications Qusai Shambour a,b, Jie Lu a, a Lab of Decision Systems and e-service Intelligence, Centre for Quantum Computation

More information

Combinational Collaborative Filtering: An Approach For Personalised, Contextually Relevant Product Recommendation Baskets

Combinational Collaborative Filtering: An Approach For Personalised, Contextually Relevant Product Recommendation Baskets Combinational Collaborative Filtering: An Approach For Personalised, Contextually Relevant Product Recommendation Baskets Research Project - Jai Chopra (338852) Dr Wei Wang (Supervisor) Dr Yifang Sun (Assessor)

More information

Trust-Aware Recommender Systems

Trust-Aware Recommender Systems Mohammad Ali Abbasi, Jiliang Tang, and Huan Liu Computer Science and Engineering, Arizona State University {Ali.Abbasi, Jiliang.Tang, Huan.Liu}@asu.edu Trust-Aware Recommender Systems Chapter 1 Trust-Aware

More information

A Comparative Study of Filter-based Feature Ranking Techniques

A Comparative Study of Filter-based Feature Ranking Techniques Western Kentucky University From the SelectedWorks of Dr. Huanjing Wang August, 2010 A Comparative Study of Filter-based Feature Ranking Techniques Huanjing Wang, Western Kentucky University Taghi M. Khoshgoftaar,

More information

XPLODIV: An Exploitation-Exploration Aware Diversification Approach for Recommender Systems

XPLODIV: An Exploitation-Exploration Aware Diversification Approach for Recommender Systems Proceedings of the Twenty-Eighth International Florida Artificial Intelligence Research Society Conference XPLODIV: An Exploitation-Exploration Aware Diversification Approach for Recommender Systems Andrea

More information

Near-Balanced Incomplete Block Designs with An Application to Poster Competitions

Near-Balanced Incomplete Block Designs with An Application to Poster Competitions Near-Balanced Incomplete Block Designs with An Application to Poster Competitions arxiv:1806.00034v1 [stat.ap] 31 May 2018 Xiaoyue Niu and James L. Rosenberger Department of Statistics, The Pennsylvania

More information

Client Clustering for Hiring Modeling in Work Marketplaces

Client Clustering for Hiring Modeling in Work Marketplaces Client Clustering for Hiring Modeling in Work Marketplaces ABSTRACT Vasilis Verroios Stanford University verroios@stanford.edu Ramesh Johari Stanford University ramesh.johari@stanford.edu An important

More information

Chapter 5 Evaluating Classification & Predictive Performance

Chapter 5 Evaluating Classification & Predictive Performance Chapter 5 Evaluating Classification & Predictive Performance Data Mining for Business Intelligence Shmueli, Patel & Bruce Galit Shmueli and Peter Bruce 2010 Why Evaluate? Multiple methods are available

More information

Estimating With Objects - Part III

Estimating With Objects - Part III Estimating With Objects - Part III Contents The size estimating problem The comparison problem Estimating part size Selecting a proxy Relationship to development effort The proxy parts in a product can

More information

Application of Data Mining Techniques for Crop Productivity Prediction

Application of Data Mining Techniques for Crop Productivity Prediction Application of Data Mining Techniques for Crop Productivity Prediction Zekarias Diriba zekariaskifle@gmail.com Berhanu Borena PhD Candidate, Addis Ababa University, Ethiopia berhanuborena@gmail.com Abstract

More information

Using Metrics Through the Technology Assisted Review (TAR) Process. A White Paper

Using Metrics Through the Technology Assisted Review (TAR) Process. A White Paper Using Metrics Through the Technology Assisted Review (TAR) Process A White Paper 1 2 Table of Contents Introduction....4 Understanding the Data Set....4 Determine the Rate of Relevance....5 Create Training

More information

Chapter 3. Integrating AHP, clustering and association rule mining

Chapter 3. Integrating AHP, clustering and association rule mining Chapter 3. Integrating AHP, clustering and association rule mining This chapter presents a novel product recommendation methodology that combines group decision-making and data mining. This method has

More information

Improving Recommendation using Forecast based approach and Re-ranking Approache

Improving Recommendation using Forecast based approach and Re-ranking Approache IOSR Journal of Computer Engineering (IOSR-JCE) e-issn: 2278-0661,p-ISSN: 2278-8727, Volume 16, Issue 5, Ver. IV (Sep Oct. 2014), PP 145-153 Improving Recommendation using Forecast based approach and Re-ranking

More information

A STUDY ON STATISTICAL BASED FEATURE SELECTION METHODS FOR CLASSIFICATION OF GENE MICROARRAY DATASET

A STUDY ON STATISTICAL BASED FEATURE SELECTION METHODS FOR CLASSIFICATION OF GENE MICROARRAY DATASET A STUDY ON STATISTICAL BASED FEATURE SELECTION METHODS FOR CLASSIFICATION OF GENE MICROARRAY DATASET 1 J.JEYACHIDRA, M.PUNITHAVALLI, 1 Research Scholar, Department of Computer Science and Applications,

More information

You are Who You Know and How You Behave: Attribute Inference Attacks via Users Social Friends and Behaviors

You are Who You Know and How You Behave: Attribute Inference Attacks via Users Social Friends and Behaviors You are Who You Know and How You Behave: Attribute Inference via Users Social Friends and Behaviors Neil Zhenqiang Gong ECE Department, Iowa State University neilgong@iastate.edu Bin Liu MSIS Department,

More information

An Analysis of Emotion and user Behavior in Context-Aware Travel Recommendation Systems using Pre-Filtering and Tensor Factorization Techniques

An Analysis of Emotion and user Behavior in Context-Aware Travel Recommendation Systems using Pre-Filtering and Tensor Factorization Techniques Global Journal of Computer Science and Technology: D Neural & Artificial Intelligence Volume 18 Issue 1 Version 1.0 Type: Double Blind Peer Reviewed International Research Journal Publisher: Global Journals

More information

Optimizing your promotional campaign with uplift modelling

Optimizing your promotional campaign with uplift modelling ALLMARK FACT BASED MARKETING Amsterdam, 2017 Optimizing your promotional campaign with uplift modelling White paper: Optimizing your promotional campaign with uplift modelling 2017 Optimizing your promotional

More information

Networked Life (CSE 112)

Networked Life (CSE 112) Networked Life (CSE 112) Prof. Michael Kearns Final Examination May 3, 2006 The final exam is closed-book; you should have no materials present other than the exam and a pen or pencil. NAME: PENN ID: Exam

More information

Comparative evaluation of top-n recommenders in e-commerce: an industry perspective

Comparative evaluation of top-n recommenders in e-commerce: an industry perspective Comparative evaluation of top-n recommenders in e-commerce: an industry perspective Dimitris Paraschakis Dept. of Computer Science Malmö University SE-205 06 Malmö, Sweden Email: dimitris.paraschakis@mah.se

More information

CHAPTER VI FUZZY MODEL

CHAPTER VI FUZZY MODEL CHAPTER VI This chapter covers development of a model using fuzzy logic to relate degree of recording/evaluation with the effectiveness of financial decision making and validate the model vis-à-vis results

More information

A Personalized Company Recommender System for Job Seekers Yixin Cai, Ruixi Lin, Yue Kang

A Personalized Company Recommender System for Job Seekers Yixin Cai, Ruixi Lin, Yue Kang A Personalized Company Recommender System for Job Seekers Yixin Cai, Ruixi Lin, Yue Kang Abstract Our team intends to develop a recommendation system for job seekers based on the information of current

More information

Modelling buyer behaviour/3 How survey results can mislead

Modelling buyer behaviour/3 How survey results can mislead Publishing Date: June 1993. 1993. All rights reserved. Copyright rests with the author. No part of this article may be reproduced without written permission from the author. Modelling buyer behaviour/3

More information

EVALUATING COMPETITIVENESS FROM LARGE DATASETS BY MINING COMPETITORS

EVALUATING COMPETITIVENESS FROM LARGE DATASETS BY MINING COMPETITORS EVALUATING COMPETITIVENESS FROM LARGE DATASETS BY MINING COMPETITORS Prof.D.Ravi, Ms.K.Sowmiya, Mr.U.Abhilash, Mr.K.Vignesh Abstract Success of any competitive business, is based on the ability to make

More information

INSIGHTS. Demand Planner for Microsoft Dynamics. Product Overview. Date: November,

INSIGHTS. Demand Planner for Microsoft Dynamics. Product Overview. Date: November, INSIGHTS Demand Planner for Microsoft Dynamics Product Overview Date: November, 2007 www.microsoft.com/dynamics Contents Demand Planning for Business... 1 Product Overview... 3 Multi-dimensional Data Visibility...

More information

MODELING THE EXPERT. An Introduction to Logistic Regression The Analytics Edge

MODELING THE EXPERT. An Introduction to Logistic Regression The Analytics Edge MODELING THE EXPERT An Introduction to Logistic Regression 15.071 The Analytics Edge Ask the Experts! Critical decisions are often made by people with expert knowledge Healthcare Quality Assessment Good

More information

Extensive Experimental Validation of a Personalized Approach for Coping with Unfair Ratings in Reputation Systems

Extensive Experimental Validation of a Personalized Approach for Coping with Unfair Ratings in Reputation Systems Extensive Experimental Validation of a Personalized Approach for Coping with Unfair Ratings in Reputation Systems Nanyang Technological University, School of Computer Engineering, zhangj@ntu.edu.sg Received

More information

Targeted Advertising with Inventory Management

Targeted Advertising with Inventory Management Targeted Advertising with Inventory Management David Maxwell Chickering Microsoft Research Redmond, WA 98052 dmax@microsoft.com David Heckerman Microsoft Research Redmond, WA 98052 heckerma@microsoft.com

More information

Evaluation of Machine Learning Algorithms in Recommender Systems

Evaluation of Machine Learning Algorithms in Recommender Systems Evaluation of Machine Learning Algorithms in Recommender Systems Candidate Recommender Systems in the Staffing Industry Master s Thesis in Software Engineering Adam Myrén Piotr Skupniewicz Neto Department

More information

# of Problems Written # of Homeworks Corrected Tom 2 26 Yoshi 3 27 Gary 6 15

# of Problems Written # of Homeworks Corrected Tom 2 26 Yoshi 3 27 Gary 6 15 Economics 101 Spring 2015 Answers to Homework #2 Due Thursday, February 19, 2015 Directions: The homework will be collected in a box before the lecture. Please place your name on top of the homework (legibly).

More information

Expert Based Prediction of User Preferences

Expert Based Prediction of User Preferences Expert Based Prediction of User Preferences M. Dima Department of Informatics National and Kapodistrian University of Athens, Greece D. Vogiatzis Institute of Informatics and Telecommunications, National

More information

Differential Neighborhood Selection in Memory-Based Group Recommender Systems

Differential Neighborhood Selection in Memory-Based Group Recommender Systems Proceedings of the Twenty-Seventh International Florida Artificial Intelligence Research Society Conference Differential Neighborhood Selection in Memory-Based Group Recommender Systems Nadia A. Najjar

More information

Assistant Professor Neha Pandya Department of Information Technology, Parul Institute Of Engineering & Technology Gujarat Technological University

Assistant Professor Neha Pandya Department of Information Technology, Parul Institute Of Engineering & Technology Gujarat Technological University Feature Level Text Categorization For Opinion Mining Gandhi Vaibhav C. Computer Engineering Parul Institute Of Engineering & Technology Gujarat Technological University Assistant Professor Neha Pandya

More information

Preface to the third edition Preface to the first edition Acknowledgments

Preface to the third edition Preface to the first edition Acknowledgments Contents Foreword Preface to the third edition Preface to the first edition Acknowledgments Part I PRELIMINARIES XXI XXIII XXVII XXIX CHAPTER 1 Introduction 3 1.1 What Is Business Analytics?................

More information

Starting Workflow Tasks Before They re Ready

Starting Workflow Tasks Before They re Ready Starting Workflow Tasks Before They re Ready Wladislaw Gusew, Björn Scheuermann Computer Engineering Group, Humboldt University of Berlin Agenda Introduction Execution semantics Methods and tools Simulation

More information

Predict Commercial Promoted Contents Will Be Clicked By User

Predict Commercial Promoted Contents Will Be Clicked By User Predict Commercial Promoted Contents Will Be Clicked By User Gary(Xinran) Guo garyguo@stanford.edu SUNetID: garyguo Stanford University 1. Introduction As e-commerce, social media grows rapidly, advertisements

More information

THE INFLUENCE OF ORGANIZATIONAL STRUCTURE ON SOFTWARE QUALITY: AN EMPIRICAL CASE STUDY

THE INFLUENCE OF ORGANIZATIONAL STRUCTURE ON SOFTWARE QUALITY: AN EMPIRICAL CASE STUDY THE INFLUENCE OF ORGANIZATIONAL STRUCTURE ON SOFTWARE QUALITY: AN EMPIRICAL CASE STUDY Nachiappan Nagappan Microsoft Research Brendan Murphy Microsoft Research Victor R. Basili University of Maryland Presentation:

More information

Data in the Aggregate: Discovering Honest Signals and Predictable Patterns Within Ultra Large Data Sets

Data in the Aggregate: Discovering Honest Signals and Predictable Patterns Within Ultra Large Data Sets Data in the Aggregate: Discovering Honest Signals and Predictable Patterns Within Ultra Large Data Sets Kathleen Lossau and Jonathan Larson April 03 Getting Precision Search from Big Data Google type search

More information

Methods and tools for exploring functional genomics data

Methods and tools for exploring functional genomics data Methods and tools for exploring functional genomics data William Stafford Noble Department of Genome Sciences Department of Computer Science and Engineering University of Washington Outline Searching for

More information

Recommender Systems RELATED WORKS. Md Feroj Ahmod Md Jahidul Mozumder Yassine boutabia

Recommender Systems RELATED WORKS. Md Feroj Ahmod Md Jahidul Mozumder Yassine boutabia Recommender Systems RELATED WORKS Md Feroj Ahmod Md Jahidul Mozumder Yassine boutabia Studies PackageBuilder: From Tuples to Packages[M.Brucato, R. Ramakrishna, A.Abouzied, A.Meliou]2015 A Package Recommendation

More information

NOVEC Project Proposal

NOVEC Project Proposal NOVEC Project Proposal Authors: Abdulrahaman Alsabti Kevin Mitchell Dan Package 10/02/2014 Contents I. Introduction... 1 a. Background... 1 b. Problem Statement... 1 c. Stakeholders... 2 d. Scope... 2

More information

A Comparative Study of Recommendation Methods for Mobile OSN Users

A Comparative Study of Recommendation Methods for Mobile OSN Users A Comparative Study of Recommendation Methods for Mobile OSN Users Shyam Krishna K 1, Dr. Vince Paul 2 M.Tech Student, Department of Computer Science &Engineering, Sahrdaya College of Engineering & Technology,

More information

How to drive profits with customer segmentation. A beginner s guide to breaking down your market into targetable groups

How to drive profits with customer segmentation. A beginner s guide to breaking down your market into targetable groups How to drive profits with customer segmentation A beginner s guide to breaking down your market into targetable groups Table of contents 3 5 Introduction Why segmentation? 8 9 11 13 17 21 5 steps to build

More information