CMU SCS 15-826: Multimedia Databases and Data Mining Lecture #16: Text - part III: Vector space model and clustering C. Faloutsos.

Slides:



Advertisements
Similar presentations
CMU SCS : Multimedia Databases and Data Mining Lecture #17: Text - part IV (LSI) C. Faloutsos.
Advertisements

CMU SCS : Multimedia Databases and Data Mining Lecture #7: Spatial Access Methods - IV Grid files, dim. curse C. Faloutsos.
CMU SCS : Multimedia Databases and Data Mining Lecture #14: Text – Part I C. Faloutsos.
CMU SCS : Multimedia Databases and Data Mining Lecture #19: SVD - part II (case studies) C. Faloutsos.
Chapter 5: Introduction to Information Retrieval
CMU SCS : Multimedia Databases and Data Mining Lecture #7: Spatial Access Methods - Metric trees C. Faloutsos.
CMU SCS : Multimedia Databases and Data Mining Lecture#5: Multi-key and Spatial Access Methods - II C. Faloutsos.
CMU SCS : Multimedia Databases and Data Mining Lecture#2: Primary key indexing – B-trees Christos Faloutsos - CMU
CMU SCS : Multimedia Databases and Data Mining Lecture#3: Primary key indexing – hashing C. Faloutsos.
Temple University – CIS Dept. CIS616– Principles of Data Management V. Megalooikonomou Text Databases (some slides are based on notes by C. Faloutsos)
15-826: Multimedia Databases and Data Mining
Search and Retrieval: More on Term Weighting and Document Ranking Prof. Marti Hearst SIMS 202, Lecture 22.
Query Operations: Automatic Local Analysis. Introduction Difficulty of formulating user queries –Insufficient knowledge of the collection –Insufficient.
Text Databases. Outline Spatial Databases Temporal Databases Spatio-temporal Databases Data Mining Multimedia Databases Text databases Image and video.
Database Management Systems, R. Ramakrishnan1 Computing Relevance, Similarity: The Vector Space Model Chapter 27, Part B Based on Larson and Hearst’s slides.
Clustering… in General In vector space, clusters are vectors found within  of a cluster vector, with different techniques for determining the cluster.
Multimedia and Text Indexing. Multimedia Data Management The need to query and analyze vast amounts of multimedia data (i.e., images, sound tracks, video.
Chapter 5: Query Operations Baeza-Yates, 1999 Modern Information Retrieval.
Modern Information Retrieval Chapter 2 Modeling. Can keywords be used to represent a document or a query? keywords as query and matching as query processing.
Introduction to Information Retrieval Introduction to Information Retrieval Hinrich Schütze and Christina Lioma Lecture 16: Flat Clustering 1.
Modeling Modern Information Retrieval
Text and Web Search. Text Databases and IR Text databases (document databases) Large collections of documents from various sources: news articles, research.
1 CS 430 / INFO 430 Information Retrieval Lecture 3 Vector Methods 1.
Information retrieval Finding relevant data using irrelevant keys Example: database of photographic images sorted by number, date. DBMS: Well structured.
10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.
TextMOLE: Text Mining Operations Library and Environment Daniel B. Waegel and April Kontostathis, Ph.D. Ursinus College Collegeville PA.
Ranking by Odds Ratio A Probability Model Approach let be a Boolean random variable: document d is relevant to query q otherwise Consider document d as.
Modern Information Retrieval Chapter 2 Modeling. Can keywords be used to represent a document or a query? keywords as query and matching as query processing.
Multimedia Databases Text II. Outline Spatial Databases Temporal Databases Spatio-temporal Databases Multimedia Databases Text databases Image and video.
Multimedia Databases LSI and SVD. Text - Detailed outline text problem full text scanning inversion signature files clustering information filtering and.
IR Models: Review Vector Model and Probabilistic.
1 CS 430 / INFO 430 Information Retrieval Lecture 3 Searching Full Text 3.
10-603/15-826A: Multimedia Databases and Data Mining Text - part II C. Faloutsos.
CMU SCS : Multimedia Databases and Data Mining Lecture #30: Conclusions C. Faloutsos.
Chapter 5: Information Retrieval and Web Search
Clustering Unsupervised learning Generating “classes”
1 Vector Space Model Rong Jin. 2 Basic Issues in A Retrieval Model How to represent text objects What similarity function should be used? How to refine.
Modern Information Retrieval: A Brief Overview By Amit Singhal Ranjan Dash.
Query Operations J. H. Wang Mar. 26, The Retrieval Process User Interface Text Operations Query Operations Indexing Searching Ranking Index Text.
Xiaoying Gao Computer Science Victoria University of Wellington Intelligent Agents COMP 423.
Clustering Supervised vs. Unsupervised Learning Examples of clustering in Web IR Characteristics of clustering Clustering algorithms Cluster Labeling 1.
Basic Machine Learning: Clustering CS 315 – Web Search and Data Mining 1.
1 Motivation Web query is usually two or three words long. –Prone to ambiguity –Example “keyboard” –Input device of computer –Musical instruments How can.
Term Frequency. Term frequency Two factors: – A term that appears just once in a document is probably not as significant as a term that appears a number.
Chapter 6: Information Retrieval and Web Search
CMU SCS : Multimedia Databases and Data Mining Lecture #12: Fractals - case studies Part III (quadtrees, knn queries) C. Faloutsos.
1 Computing Relevance, Similarity: The Vector Space Model.
Introduction to Digital Libraries hussein suleman uct cs honours 2003.
University of Malta CSA3080: Lecture 6 © Chris Staff 1 of 20 CSA3080: Adaptive Hypertext Systems I Dr. Christopher Staff Department.
Collocations and Information Management Applications Gregor Erbach Saarland University Saarbrücken.
Information Retrieval Lecture 6 Introduction to Information Retrieval (Manning et al. 2007) Chapter 16 For the MSc Computer Science Programme Dell Zhang.
Vector Space Models.
CS 8751 ML & KDDData Clustering1 Clustering Unsupervised learning Generating “classes” Distance/similarity measures Agglomerative methods Divisive methods.
E.G.M. PetrakisText Clustering1 Clustering  “Clustering is the unsupervised classification of patterns (observations, data items or feature vectors) into.
CIS750 – Seminar in Advanced Topics in Computer Science Advanced topics in databases – Multimedia Databases V. Megalooikonomou Text Databases (some slides.
Basic Machine Learning: Clustering CS 315 – Web Search and Data Mining 1.
CMU SCS : Multimedia Databases and Data Mining Lecture #18: SVD - part I (definitions) C. Faloutsos.
1 CS 430 / INFO 430 Information Retrieval Lecture 3 Searching Full Text 3.
Natural Language Processing Topics in Information Retrieval August, 2002.
Xiaoying Gao Computer Science Victoria University of Wellington COMP307 NLP 4 Information Retrieval.
CMU SCS : Multimedia Databases and Data Mining Lecture #7: Spatial Access Methods - Metric trees C. Faloutsos.
Information Retrieval and Web Search
15-826: Multimedia Databases and Data Mining
15-826: Multimedia Databases and Data Mining
15-826: Multimedia Databases and Data Mining
15-826: Multimedia Databases and Data Mining
15-826: Multimedia Databases and Data Mining
15-826: Multimedia Databases and Data Mining
Chapter 5: Information Retrieval and Web Search
15-826: Multimedia Databases and Data Mining
Presentation transcript:

CMU SCS : Multimedia Databases and Data Mining Lecture #16: Text - part III: Vector space model and clustering C. Faloutsos

CMU SCS Copyright: C. Faloutsos (2012)2 Must-Read Material MM Textbook, Chapter 6

CMU SCS Copyright: C. Faloutsos (2012)3 Outline Goal: ‘Find similar / interesting things’ Intro to DB Indexing - similarity search Data Mining

CMU SCS Copyright: C. Faloutsos (2012)4 Indexing - Detailed outline primary key indexing secondary key / multi-key indexing spatial access methods fractals text multimedia...

CMU SCS Copyright: C. Faloutsos (2012)5 Text - Detailed outline text –problem –full text scanning –inversion –signature files –clustering –information filtering and LSI

CMU SCS Copyright: C. Faloutsos (2012)6 Vector Space Model and Clustering keyword queries (vs Boolean) each document: -> vector (HOW?) each query: -> vector search for ‘similar’ vectors

CMU SCS Copyright: C. Faloutsos (2012)7 Vector Space Model and Clustering main idea: document...data... aaron zoo data V (= vocabulary size) ‘indexing’

CMU SCS Copyright: C. Faloutsos (2012)8 Vector Space Model and Clustering Then, group nearby vectors together Q1: cluster search? Q2: cluster generation? Two significant contributions ranked output relevance feedback

CMU SCS Copyright: C. Faloutsos (2012)9 Vector Space Model and Clustering cluster search: visit the (k) closest superclusters; continue recursively CS TRs MD TRs

CMU SCS Copyright: C. Faloutsos (2012)10 Vector Space Model and Clustering ranked output: easy! CS TRs MD TRs

CMU SCS Copyright: C. Faloutsos (2012)11 Vector Space Model and Clustering relevance feedback (brilliant idea) [Roccio’73] CS TRs MD TRs

CMU SCS Copyright: C. Faloutsos (2012)12 Vector Space Model and Clustering relevance feedback (brilliant idea) [Roccio’73] How? CS TRs MD TRs

CMU SCS Copyright: C. Faloutsos (2012)13 Vector Space Model and Clustering How? A: by adding the ‘good’ vectors and subtracting the ‘bad’ ones CS TRs MD TRs

CMU SCS Copyright: C. Faloutsos (2012)14 Outline - detailed main idea cluster search cluster generation evaluation

CMU SCS Copyright: C. Faloutsos (2012)15 Cluster generation Problem: – given N points in V dimensions, –group them

CMU SCS Copyright: C. Faloutsos (2012)16 Cluster generation Problem: – given N points in V dimensions, –group them

CMU SCS Copyright: C. Faloutsos (2012)17 Cluster generation We need Q1: document-to-document similarity Q2: document-to-cluster similarity

CMU SCS Copyright: C. Faloutsos (2012)18 Cluster generation Q1: document-to-document similarity (recall: ‘bag of words’ representation) D1: {‘data’, ‘retrieval’, ‘system’} D2: {‘lung’, ‘pulmonary’, ‘system’} distance/similarity functions?

CMU SCS Copyright: C. Faloutsos (2012)19 Cluster generation A1: # of words in common A2: normalized by the vocabulary sizes A3:.... etc About the same performance - prevailing one: cosine similarity

CMU SCS Copyright: C. Faloutsos (2012)20 Cluster generation cosine similarity: similarity(D1, D2) = cos( θ ) = sum(v 1,i * v 2,i ) / len(v 1 )/ len(v 2 ) θ D1 D2

CMU SCS Copyright: C. Faloutsos (2012)21 Cluster generation cosine similarity - observations: related to the Euclidean distance weights v i,j : according to tf/idf θ D1 D2

CMU SCS Copyright: C. Faloutsos (2012)22 Cluster generation tf (‘term frequency’) high, if the term appears very often in this document. idf (‘inverse document frequency’) penalizes ‘common’ words, that appear in almost every document

CMU SCS Copyright: C. Faloutsos (2012)23 Cluster generation We need Q1: document-to-document similarity Q2: document-to-cluster similarity ?

CMU SCS Copyright: C. Faloutsos (2012)24 Cluster generation A1: min distance (‘single-link’) A2: max distance (‘all-link’) A3: avg distance A4: distance to centroid ?

CMU SCS Copyright: C. Faloutsos (2012)25 Cluster generation A1: min distance (‘single-link’) –leads to elongated clusters A2: max distance (‘all-link’) –many, small, tight clusters A3: avg distance –in between the above A4: distance to centroid –fast to compute

CMU SCS Copyright: C. Faloutsos (2012)26 Cluster generation We have document-to-document similarity document-to-cluster similarity Q: How to group documents into ‘natural’ clusters

CMU SCS Copyright: C. Faloutsos (2012)27 Cluster generation A: *many-many* algorithms - in two groups [VanRijsbergen]: theoretically sound (O(N^2)) –independent of the insertion order iterative (O(N), O(N log(N))

CMU SCS Copyright: C. Faloutsos (2012)28 Cluster generation - ‘sound’ methods Approach#1: dendrograms - create a hierarchy (bottom up or top-down) - choose a cut-off (how?) and cut cat tigerhorsecow

CMU SCS Copyright: C. Faloutsos (2012)29 Cluster generation - ‘sound’ methods Approach#2: min. some statistical criterion (eg., sum of squares from cluster centers) –like ‘k-means’ –but how to decide ‘k’?

CMU SCS Copyright: C. Faloutsos (2012)30 Cluster generation - ‘sound’ methods Approach#3: Graph theoretic [Zahn]: –build MST; –delete edges longer than 3* std of the local average

CMU SCS Copyright: C. Faloutsos (2012)31 Cluster generation - ‘sound’ methods Result: why ‘3’? variations Complexity?

CMU SCS Copyright: C. Faloutsos (2012)32 Cluster generation - ‘iterative’ methods general outline: Choose ‘seeds’ (how?) assign each vector to its closest seed (possibly adjusting cluster centroid) possibly, re-assign some vectors to improve clusters Fast and practical, but ‘unpredictable’

CMU SCS Copyright: C. Faloutsos (2012)33 Cluster generation - ‘iterative’ methods general outline: Choose ‘seeds’ (how?) assign each vector to its closest seed (possibly adjusting cluster centroid) possibly, re-assign some vectors to improve clusters Fast and practical, but ‘unpredictable’

CMU SCS Copyright: C. Faloutsos (2012)34 Cluster generation one way to estimate # of clusters k: the ‘cover coefficient’ [Can+] ~ SVD

CMU SCS Copyright: C. Faloutsos (2012)35 Outline - detailed main idea cluster search cluster generation evaluation

CMU SCS Copyright: C. Faloutsos (2012)36 Evaluation Q: how to measure ‘goodness’ of one distance function vs another? A: ground truth (by humans) and –‘precision’ and ‘recall’

CMU SCS Copyright: C. Faloutsos (2012)37 Evaluation precision = (retrieved & relevant) / retrieved –100% precision -> no false alarms recall = (retrieved & relevant)/ relevant –100% recall -> no false dismissals

CMU SCS Copyright: C. Faloutsos (2012)38 Evaluation recall precision ideal No false alarms No false dismissals

CMU SCS Copyright: C. Faloutsos (2012)39 Evaluation compressing such a curve into a single number: –11-point average precision –etc

CMU SCS Copyright: C. Faloutsos (2012)40 Conclusions – main ideas ‘bag of words’ idea + keyword queries Cosine similarity Ranked output Relevance feedback

CMU SCS Copyright: C. Faloutsos (2012)41 References Modern Information Retrieval R. Baeza-Yates, Acm Press, Berthier Ribeiro-Neto, February 1999R. Baeza-YatesAcm Press Berthier Ribeiro-Neto Can, F. and E. A. Ozkarahan (Dec. 1990). "Concepts and Effectiveness of the Cover-Coefficient-Based Clustering Methodology for Text Databases." ACM TODS 15(4): Noreault, T., M. McGill, et al. (1983). A Performance Evaluation of Similarity Measures, Document Term Weighting Schemes and Representation in a Boolean Environment. Information Retrieval Research, Butterworths.

CMU SCS Copyright: C. Faloutsos (2012)42 References Rocchio, J. J. (1971). Relevance Feedback in Information Retrieval. The SMART Retrieval System - Experiments in Automatic Document Processing. G. Salton. Englewood Cliffs, New Jersey, Prentice-Hall Inc. Salton, G. (1971). The SMART Retrieval System - Experiments in Automatic Document Processing. Englewood Cliffs, New Jersey, Prentice-Hall Inc.

CMU SCS Copyright: C. Faloutsos (2012)43 References Salton, G. and M. J. McGill (1983). Introduction to Modern Information Retrieval, McGraw-Hill. Van-Rijsbergen, C. J. (1979). Information Retrieval. London, England, Butterworths. Zahn, C. T. (Jan. 1971). "Graph-Theoretical Methods for Detecting and Describing Gestalt Clusters." IEEE Trans. on Computers C-20(1):