Location-Aware Query Recommendation for Search Engines at Scale

Slides:



Advertisements
Similar presentations
Green Network Project Contract
Advertisements

Web Information Retrieval
Computer Science and Engineering Inverted Linear Quadtree: Efficient Top K Spatial Keyword Search Chengyuan Zhang 1,Ying Zhang 1,Wenjie Zhang 1, Xuemin.
Diversity Maximization Under Matroid Constraints Date : 2013/11/06 Source : KDD’13 Authors : Zeinab Abbassi, Vahab S. Mirrokni, Mayur Thakur Advisor :
Supporting Top-k join Queries in Relational Databases By:Ihab F. Ilyas, Walid G. Aref, Ahmed K. Elmagarmid Presented by: Calvin R Noronha ( )
Quality Aware Privacy Protection for Location-based Services Zhen Xiao, Xiaofeng Meng Renmin University of China Jianliang Xu Hong Kong Baptist University.
Crawling, Ranking and Indexing. Organizing the Web The Web is big. Really big. –Over 3 billion pages, just in the indexable Web The Web is dynamic Problems:
Pete Bohman Adam Kunk.  Introduction  Related Work  System Overview  Indexing Scheme  Ranking  Evaluation  Conclusion.
1 Routing Techniques in Wireless Sensor networks: A Survey.
Jean-Eudes Ranvier 17/05/2015Planet Data - Madrid Trustworthiness assessment (on web pages) Task 3.3.
DATA MINING LECTURE 12 Link Analysis Ranking Random walks.
Graph Drawing Zsuzsanna Hollander. Reviewed Papers Effective Graph Visualization via Node Grouping Janet M. Six and Ioannis G. Tollis. Proc InfoVis 2001.
Color Aware Switch algorithm implementation The Computer Communication Lab (236340) Spring 2008.
1 Collaborative Filtering and Pagerank in a Network Qiang Yang HKUST Thanks: Sonny Chee.
Dissemination protocols for large sensor networks Fan Ye, Haiyun Luo, Songwu Lu and Lixia Zhang Department of Computer Science UCLA Chien Kang Wu.
Zdravko Markov and Daniel T. Larose, Data Mining the Web: Uncovering Patterns in Web Content, Structure, and Usage, Wiley, Slides for Chapter 1:
Adaptive Sampling in Distributed Streaming Environment Ankur Jain 2/4/03.
(hyperlink-induced topic search)
Personalized Ontologies for Web Search and Caching Susan Gauch Information and Telecommunications Technology Center Electrical Engineering and Computer.
Optimal Crawling Strategies for Web Search Engines Wolf, Sethuraman, Ozsen Presented By Rajat Teotia.
Web Spam Detection: link-based and content-based techniques Reporter : 鄭志欣 Advisor : Hsing-Kuo Pao 2010/11/8 1.
Mobile Application Abstract Future Work The potential applications and integration of this project are vast – many large department and grocery stores.
School of Electronics Engineering and Computer Science Peking University Beijing, P.R. China Ziqi Wang, Yuwei Tan, Ming Zhang.
Suggesting Friends using the Implicit Social Graph Maayan Roth et al. (Google, Inc., Israel R&D Center) KDD’10 Hyewon Lim 1 Oct 2014.
VLDB '2006 Haibo Hu (Hong Kong Baptist University, Hong Kong) Dik Lun Lee (Hong Kong University of Science and Technology, Hong Kong) Victor.
Mehdi Kargar Aijun An York University, Toronto, Canada Keyword Search in Graphs: Finding r-cliques.
1 Web Search Personalization via Social Bookmarking and Tagging Michael G. Noll & Christoph Meinel Hasso-Plattner-Institut an der Universit¨at Potsdam,
1 University of Qom Information Retrieval Course Web Search (Link Analysis) Based on:
Exploring Online Social Activities for Adaptive Search Personalization CIKM’10 Advisor : Jia Ling, Koh Speaker : SHENG HONG, CHUNG.
GraphLab: how I understood it with sample code Aapo Kyrola, Carnegie Mellon Univ. Oct 1, 2009.
Query Suggestions Debapriyo Majumdar Information Retrieval – Spring 2015 Indian Statistical Institute Kolkata.
Department of Computer Science City University of Hong Kong Department of Computer Science City University of Hong Kong 1 Probabilistic Continuous Update.
WEB SEARCH PERSONALIZATION WITH ONTOLOGICAL USER PROFILES Data Mining Lab XUAN MAN.
Presenter: Lung-Hao Lee ( 李龍豪 ) January 7, 309.
استاد : مهندس حسین پور ارائه دهنده : احسان جوانمرد Google Architecture.
Web Search Algorithms By Matt Richard and Kyle Krueger.
Mehdi Kargar Aijun An York University, Toronto, Canada Keyword Search in Graphs: Finding r-cliques.
Improving Web Search Results Using Affinity Graph Benyu Zhang, Hua Li, Yi Liu, Lei Ji, Wensi Xi, Weiguo Fan, Zheng Chen, Wei-Ying Ma Microsoft Research.
Query Suggestion Naama Kraus Slides are based on the papers: Baeza-Yates, Hurtado, Mendoza, Improving search engines by query clustering Boldi, Bonchi,
Answering Similar Region Search Queries Chang Sheng, Yu Zheng.
WebQuery: Searching and Visualizing the Web through Connectivity Rick Kazman Software Engineering Institute Pittsburgh, PA
On Computing Top-t Influential Spatial Sites Authors: T. Xia, D. Zhang, E. Kanoulas, Y.Du Northeastern University, USA Appeared in: VLDB 2005 Presenter:
Jiafeng Guo(ICT) Xueqi Cheng(ICT) Hua-Wei Shen(ICT) Gu Xu (MSRA) Speaker: Rui-Rui Li Supervisor: Prof. Ben Kao.
Intelligent Database Systems Lab N.Y.U.S.T. I. M. Externally growing self-organizing maps and its application to database visualization and exploration.
Post-Ranking query suggestion by diversifying search Chao Wang.
CS483/683 Multi-Agent Systems Lecture 2: Distributed variants of 2 important AI problems: Search and Constraint Satisfaction 21 January 2010 Instructor:
An MPEG-7 Based Semantic Album for Home Entertainment Presented by Chen-hsiu Huang 2003/08/12 Presented by Chen-hsiu Huang 2003/08/12.
Ad Hoc On-Demand Distance Vector Routing (AODV) ietf
The Development of a search engine & Comparison according to algorithms Sung-soo Kim The final report.
Network Partition –Finding modules of the network. Graph Clustering –Partition graphs according to the connectivity. –Nodes within a cluster is highly.
Midterm 1 Summary Grading Scale A 93%
The Anatomy of a Large-Scale Hypertextual Web Search Engine S. Brin and L. Page, Computer Networks and ISDN Systems, Vol. 30, No. 1-7, pages , April.
Ariel Fuxman, Panayiotis Tsaparas, Kannan Achan, Rakesh Agrawal (2008) - Akanksha Saxena 1.
Density-based Place Clustering in Geo-Social Networks Jieming Shi, Nikos Mamoulis, Dingming Wu, David W. Cheung Department of Computer Science, The University.
User Modeling for Personal Assistant
Ananya Das Christman CS311 Fall 2016
Information Retrieval in Practice
Link-Based Ranking Seminar Social Media Mining University UC3M
Graph Search Applications
Efficient Evaluation of k-NN Queries Using Spatial Mashups
Jinhong Jung, Woojung Jin, Lee Sael, U Kang, ICDM ‘16
Learning Literature Search Models from Citation Behavior
Variations of Weighted Fair Queueing
Continuous Density Queries for Moving Objects
Learning to Rank Typed Graph Walks: Local and Global Approaches
Markov Networks.
Information Retrieval and Web Design
On Constructing k-Connected k-Dominating Set in Wireless Networks
Network-Based Coverage of Mutational Profiles Reveals Cancer Genes
WSExpress: A QoS-Aware Search Engine for Web Services
Presentation transcript:

Location-Aware Query Recommendation for Search Engines at Scale Zhipeng Huang, The University of Hong Kong Nikos Mamoulis, University of Ioannina

Query Recommendation Suggest important and useful query alternatives to search engine users. “SSTD”

Location-Related Query Suppose you want to find a KFC in the nearby, We need better location-aware query recommendation!

Location from query issuer Dropping price of Smartphones Increasing coverage and bandwidth of mobile networks Leading to more and more search engine queries from mobile users, which have location information as an important feature. We study using the location information to improve the quality of query recommendation.

Location of a Web page Distribution got from frequency of mentioned locations.

Location of a query Query-Document bipartite graph Location Aggregation: Q D

Spatial similarity Given a search engine user u, the spatial similarity between him and a Web query q is: Where r is a range threshold (by default 100km).

Existing Recommendation models QFG (Query-Flow Graph) Recommendation: do a PPR from the input query, and recommend those with large PPR scores (or probabilities). [P. Boldi, F. Bonchi, C. Castillo, D. Donato, A. Gionis, and S. Vigna. The query-flow graph: model and applications. In CIKM , pages 609–618. ACM, 2008.]

Existing Recommendation models TQGraph (Term-Query Flow Graph) Recommendation: do a PPR from each term, and aggregate the PPR scores. [F. Bonchi, R. Perego, F. Silvestri, H. Vahabi, and R. Venturini. Efficient query recommendations in the long tail via center-piece subgraphs. In SIGIR, pages 345–354. ACM, 2012.]

Location-aware PPR Adjust the weights according to the spatial similarity between a query and the user: Then we do PPRs over the adjusted graph. Location-aware QFG Location-aware TQGraph

Computing PPR Efficiently Bookmark Coloring Algorithm (BCA) model the RWR process as a bookmark coloring process, in which some portion of the ink in a processed node is sent to its neighbors, while the remaining ink is retained at the node. (suppose ) 0.255 0.3 0.3 1.0 0.15 0.7 0.7 0.595

Lazy Update Mechanism In BCA, we need to maintain a priority query of nodes sorted by the amount of ink to be propagated. the pushing a node into the priority queue is delayed until the ink it receives is greater than a threshold. (suppose e = 0.1) Red: In queue Green: Not in queue 0.1 0.015 0.1 0.4 0.2 0.4 0.4 0.03 0.5 0.03 0.5 0.5 0.119 0.085

Spatial indeces Speed up the spatial similarity computing by calculating an approximation based on grids. We need to check at most cells, where a is the size of the cells.

Experiments Metrics: Competitors: Coverage: This is the percentage of input queries that can be served with at least one recommendation precision@k: spatialSim@k: the spatial similarity between the recommended queries to the location of user Competitors: LKS: a most recently location-aware keyword suggestion approach. SQFG: our location aware Query STQGraph: our location-aware Term-Query-Flow Graph approach. STQGraph*: approximation version. [S. Qi, D. Wu, and N. Mamoulis. Location aware keyword query suggestion based on document proximity. TKDE, 28(1):82–97, 2016]

Experiments Results: coverage results:

Experiments Running time: our default setting takes around 0.3s.

Q&A