Chapter9
-
Upload
akhila-chilukuri -
Category
Engineering
-
view
70 -
download
1
Transcript of Chapter9
1
January 17, 2001 Data Mining: Concepts and Techniques 1
Data Mining: Concepts and Techniques
— Slides for Textbook —— Chapter 9 —
©Jiawei Han and Micheline Kamber
Intelligent Database Systems Research Lab
School of Computing Science
Simon Fraser University, Canada
http://www.cs.sfu.caJanuary 17, 2001 Data Mining: Concepts and Techniques 2
Chapter 9. Mining Complex Types of Data
n Multidimensional analysis and descriptive mining of
complex data objects
n Mining spatial databases
n Mining multimedia databases
n Mining time-series and sequence data
n Mining text databases
n Mining the World-Wide Web
n Summary
January 17, 2001 Data Mining: Concepts and Techniques 3
Mining Complex Data Objects: Generalization of Structured Data
n Set-valued attributen Generalization of each value in the set into its
corresponding higher-level conceptsn Derivation of the general behavior of the set, such
as the number of elements in the set, the types or value ranges in the set, or the weighted average for numerical data
n E.g., hobby = {tennis, hockey, chess, violin, nintendo_games} generalizes to {sports, music, video_games}
n List-valued or a sequence-valued attributen Same as set-valued attributes except that the order
of the elements in the sequence should be observed in the generalization
January 17, 2001 Data Mining: Concepts and Techniques 4
Generalizing Spatial and Multimedia Data
n Spatial data:n Generalize detailed geographic points into clustered regions,
such as business, residential, industrial, or agricultural areas, according to land usage
n Require the merge of a set of geographic areas by spatial operations
n Image data:n Extracted by aggregation and/or approximationn Size, color, shape, texture, orientation, and relative positions
and structures of the contained objects or regions in the image n Music data:
n Summarize its melody: based on the approximate patterns that repeatedly occur in the segment
n Summarized its style: based on its tone, tempo, or the major musical instruments played
January 17, 2001 Data Mining: Concepts and Techniques 5
Generalizing Object Data
n Object identifier: generalize to the lowest level of class in the class/subclass hierarchies
n Class composition hierarchiesn generalize nested structured datan generalize only objects closely related in semantics to the current
onen Construction and mining of object cubes
n Extend the attribute-oriented induction methodn Apply a sequence of class-based generalization operators on
different attributesn Continue until getting a small number of generalized objects that
can be summarized as a concise in high-level termsn For efficient implementation
n Examine each attribute, generalize it to simple-valued data n Construct a multidimensional data cube (object cube)n Problem: it is not always desirable to generalize a set of values
to single-valued dataJanuary 17, 2001 Data Mining: Concepts and Techniques 6
An Example: Plan Mining by Divide and Conquer
n Plan: a variable sequence of actionsn E.g., Travel (flight): <traveler, departure, arrival, d-time, a-time,
airline, price, seat> n Plan mining: extraction of important or significant generalized
(sequential) patterns from a planbase (a large collection of plans)n E.g., Discover travel patterns in an air flight database, orn find significant patterns from the sequences of actions in the
repair of automobilesn Method
n Attribute-oriented induction on sequence data
n A generalized travel plan: <small -big*-small>n Divide & conquer:Mine characteristics for each subsequence
n E.g., big*: same airline, small-big: nearby region
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
2
January 17, 2001 Data Mining: Concepts and Techniques 7
A Travel Database for Plan Mining
n Example: Mining a travel planbase
plan# action# departure depart_time arrival arrival_time airline …1 1 ALB 800 JFK 900 TWA …1 2 JFK 1000 ORD 1230 UA …1 3 ORD 1300 LAX 1600 UA …1 4 LAX 1710 SAN 1800 DAL …2 1 SPI 900 ORD 950 AA …. . . . . . . .. . . . . . . .. . . . . . . .
airport_code city state region airport_size …1 1 ALB 800 …1 2 JFK 1000 …1 3 ORD 1300 …1 4 LAX 1710 …2 1 SPI 900 …. . . . .. . . . .. . . . .
Travel plans table
Airport info table
January 17, 2001 Data Mining: Concepts and Techniques 8
Multidimensional Analysis
n Strategy
n Generalize the planbase in different directions
n Look for sequential patterns in the generalized plans
n Derive high-level plans
A multi-D model for the planbase
January 17, 2001 Data Mining: Concepts and Techniques 9
Multidimensional Generalization
Plan# Loc_Seq Size_Seq State_Seq 1 ALB - JFK - ORD - LAX - SAN S - L - L - L - S N - N - I - C - C2 SPI - ORD - JFK - SYR S - L - L - S I - I - N - N. . .. . .. . .
Multi-D generalization of the planbase
Plan# Size_Seq State_Seq Region_Seq …1 S - L+ - S N+ - I - C+ E+ - M - P+ …2 S - L+ - S I+ - N+ M+ - E+ …. . .. . .. . .
Merging consecutive, identical actions in plans
%]75[)()(),(_),(_),,(
yregionxregionLysizeairportSxsizeairportyxflight
=⇒∧∧
January 17, 2001 Data Mining: Concepts and Techniques 10
Generalization-Based Sequence Mining
n Generalize planbase in multidimensional way using dimension tables
n Use # of distinct values (cardinality) at each level to determine the right level of generalization (level-
“planning”)
n Use operators merge “+”, option “[]” to further
generalize patterns
n Retain patterns with significant support
January 17, 2001 Data Mining: Concepts and Techniques 11
Generalized Sequence Patterns
n AirportSize-sequence survives the min threshold (after applying merge operator):
S-L+-S [35%], L+-S [30%], S-L+ [24.5%], L+ [9%]
n After applying option operator:
[S] -L+-[S] [98.5%]
n Most of the time, people fly via large airports to get to
final destination
n Other plans: 1.5% of chances, there are other patterns:
S-S, L-S-L
January 17, 2001 Data Mining: Concepts and Techniques 12
Chapter 9. Mining Complex Types of Data
n Multidimensional analysis and descriptive mining of
complex data objects
n Mining spatial databases
n Mining multimedia databases
n Mining time-series and sequence data
n Mining text databases
n Mining the World-Wide Web
n Summary
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
3
January 17, 2001 Data Mining: Concepts and Techniques 13
Spatial Data Warehousing
n Spatial data warehouse: Integrated, subject-oriented, time-variant, and nonvolatile spatial data repository for data analysis and decision making
n Spatial data integration: a big issuen Structure-specific formats (raster- vs. vector-based,
OO vs. relational models, different storage and indexing, etc.)
n Vendor-specific formats (ESRI, MapInfo, Integraph, etc.)
n Spatial data cube: multidimensional spatial databasen Both dimensions and measures may contain spatial
components
January 17, 2001 Data Mining: Concepts and Techniques 14
Dimensions and Measures in Spatial Data Warehouse
n Dimension modelingn nonspatial
n e.g. temperature: 25-30 degrees generalizes to hot
n spatial-to-nonspatialn e.g. region “B.C.”
generalizes to description “western provinces”
n spatial-to-spatialn e.g. region “Burnaby”
generalizes to region “Lower Mainland”
n Measures
n numericaln distributive (e.g. count,
sum)
n algebraic (e.g. average)
n holistic (e.g. median, rank)
n spatialn collection of spatial
pointers (e.g. pointers to all regions with 25-30 degrees in July)
January 17, 2001 Data Mining: Concepts and Techniques 15
Example: BC weather pattern analysis
n Inputn A map with about 3,000 weather probes scattered in B.C.n Daily data for temperature, precipitation, wind velocity, etc.n Concept hierarchies for all attributes
n Outputn A map that reveals patterns: merged (similar) regions
n Goalsn Interactive analysis (drill-down, slice, dice, pivot, roll-up)n Fast response timen Minimizing storage space used
n Challengen A merged region may contain hundreds of “primitive” regions
(polygons)
January 17, 2001 Data Mining: Concepts and Techniques 16
Star Schema of the BC Weather Warehouse
n Spatial data warehouse
n Dimensionsn region_namen time
n temperaturen precipitation
n Measurementsn region_map
n arean count
Fact tableDimension table
January 17, 2001 Data Mining: Concepts and Techniques 17
Spatial Merge
è Precomputing all: too much storage space
è On-line merge: very expensive
January 17, 2001 Data Mining: Concepts and Techniques 18
Methods for Computation of Spatial Data Cube
n O n-line aggregation: collect and store pointers to spatial objects in a spatial data cuben expensive and slow, need efficient aggregation
techniquesn Precompute and store all the possible combinations
n huge space overheadn Precompute and store rough approximations in a spatial
data cuben accuracy trade-off
n Selective computation: only materialize those which will be accessed frequentlyn a reasonable choice
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
4
January 17, 2001 Data Mining: Concepts and Techniques 19
Spatial Association Analysis
n Spatial association rule: A ⇒ B [s%, c%]
n A and B are sets of spatial or nonspatial predicatesn Topological relations: intersects, overlaps, disjoint, etc.n Spatial orientations: left_of, west_of, under,etc.n Distance information: close_to, within_distance, etc.
n s% is the support and c% is the confidence of the rule
n Examplesis_a(x, large_town) ^ intersect(x, highway) → adjacent_to(x, water)
[7%, 85%]is_a(x, large_town) ^adjacent_to(x, georgia_strait) → close_to(x, u.s.a.)
[1%, 78%]
January 17, 2001 Data Mining: Concepts and Techniques 20
Progressive Refinement Mining of Spatial Association Rules
n Hierarchy of spatial relationship:n g_close_to: near_by , touch, intersect, contain, etc.n First search for rough relationship and then refine it
n Two-step mining of spatial association:n Step 1: Rough spatial computation (as a filter)
n Using MBR or R-tree for rough estimationn Step2: Detailed spatial algorithm (as refinement)
n Apply only to those objects which have passed the rough spatial association test (no less than min_support)
January 17, 2001 Data Mining: Concepts and Techniques 21
n Spatial classificationn Analyze spatial objects to derive classification
schemes, such as decision trees in relevance to certain spatial properties (district, highway, river, etc.)
n Example: Classify regions in a province into rich vs. poor according to the average family income
n Spatial trend analysisn Detect changes and trends along a spatial dimensionn Study the trend of nonspatial or spatial data changing
with spacen Example: Observe the trend of changes of the climate
or vegetation with the increasing distance from an ocean
Spatial Classification and Spatial Trend Analysis
January 17, 2001 Data Mining: Concepts and Techniques 22
Chapter 9. Mining Complex Types of Data
n Multidimensional analysis and descriptive mining of
complex data objects
n Mining spatial databases
n Mining multimedia databases
n Mining time-series and sequence data
n Mining text databases
n Mining the World-Wide Web
n Summary
January 17, 2001 Data Mining: Concepts and Techniques 23
Similarity Search in Multimedia Data
n Description-based retrieval systems
n Build indices and perform object retrieval based on image descriptions, such as keywords, captions, size, and time of creation
n Labor-intensive if performed manually
n Results are typically of poor quality if automated
n Content-based retrieval systems
n Support retrieval based on the image content, such as color histogram, texture, shape, objects, and wavelet transforms
January 17, 2001 Data Mining: Concepts and Techniques 24
Queries in Content-Based Retrieval Systems
n Image sample-based queries: n Find all of the images that are similar to the given
image sample
n Compare the feature vector (signature) extracted from the sample with the feature vectors of images that have already been extracted and indexed in the image database
n Image feature specification queries: n Specify or sketch image features like color, texture, or
shape, which are translated into a feature vector
n Match the feature vector with the feature vectors of the images in the database
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
5
January 17, 2001 Data Mining: Concepts and Techniques 25
Approaches Based on Image Signature
n Color histogram-based signature
n The signature includes color histograms based on color composition of an image regardless of its scale or orientation
n No information about shape, location, or texturen Two images with similar color composition may contain
very different shapes or textures, and thus could be completely unrelated in semantics
n Multifeature composed signaturen The signature includes a composition of multiple
features: color histogram, shape, location, and texture
n Can be used to search for similar images
January 17, 2001 Data Mining: Concepts and Techniques 26
Wavelet Analysis
n Wavelet-based signaturen Use the dominant wavelet coefficients of an image as its
signaturen Wavelets capture shape, texture, and location
information in a single unified frameworkn Improved efficiency and reduced the need for providing
multiple search primitivesn May fail to identify images containing similar in location
or size objectsn Wavelet-based signature with region-based granularity
n Similar images may contain similar regions, but a region in one image could be a translation or scaling of a matching region in the other
n Compute and compare signatures at the granularity of regions, not the entire image
January 17, 2001 Data Mining: Concepts and Techniques 27
C-BIRD: Content-Based Image Retrieval from Digital libraries
Search
nby image colors
nby color percentage
nby color layout
nby texture density
nby texture Layout
nby object model
nby illumination invariance
nby keywords
January 17, 2001 Data Mining: Concepts and Techniques 28
Multi-Dimensional Search in Multimedia Databases Color layout
January 17, 2001 Data Mining: Concepts and Techniques 29
Color histogram Texture layout
Multi-Dimensional Analysis in Multimedia Databases
January 17, 2001 Data Mining: Concepts and Techniques 30
Refining or combining searches
Search for “blue sky”(top layout grid is blue)
Search for “blue sky andgreen meadows”(top layout grid is blueand bottom is green)
Search for “airplane in blue sky”(top layout grid is blue and keyword = “airplane”)
Mining Multimedia Databases
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
6
January 17, 2001 Data Mining: Concepts and Techniques 31
Multidimensional Analysis of Multimedia Data
n Multimedia data cuben Design and construction similar to that of traditional
data cubes from relational datan Contain additional dimensions and measures for
multimedia information, such as color, texture, and shape
n The database does not store images but their descriptors n Feature descriptor: a set of vectors for each visual
characteristicn Color vector: contains the color histogramn MFC (Most Frequent Color) vector: five color centroidsn MFO (Most Frequent Orientation) vector: five edge orientation
centroidsn Layout descriptor : contains a color layout vector and an
edge layout vector
January 17, 2001 Data Mining: Concepts and Techniques 32
Mining Multimedia Databases in
January 17, 2001 Data Mining: Concepts and Techniques 33
REDWHITEBLUE
GIFJPEG
By Format
By Colour
Sum
Cross Tab
REDWHITEBLUE
Colour
Sum
Group By
Measurement
JPEGGIF
Small
Very Large
REDWHITEBLUE
By Colour
By Format & Colour
By Format & Size
By Colour & Size
By FormatBy Size
Sum
The Data Cube andthe Sub-Space Measurements
MediumLarge
• Format of image• Duration• Colors• Textures• Keywords• Size• Width• Height• Internet domain of image• Internet domain of parent pages• Image popularity
Mining Multimedia Databases
January 17, 2001 Data Mining: Concepts and Techniques 34
Classification in MultiMediaMiner
January 17, 2001 Data Mining: Concepts and Techniques 35
n Special features:n Need # of occurrences besides Boolean existence, e.g.,
n “Two red square and one blue circle” implies theme “air-show”
n Need spatial relationshipsn Blue on top of white squared object is associated
with brown bottomn Need multi-resolution and progressive refinement
miningn It is expensive to explore detailed associations
among objects at high resolutionn It is crucial to ensure the completeness of search at
multi-resolution space
Mining Associations in Multimedia Data
January 17, 2001 Data Mining: Concepts and Techniques 36
Spatial Relationships from Layout
property P1 next-to property P2property P1 on-top-of property P2
Different Resolution Hierarchy
Mining Multimedia Databases
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
7
January 17, 2001 Data Mining: Concepts and Techniques 37
From Coarse to Fine Resolution Mining
Mining Multimedia Databases
January 17, 2001 Data Mining: Concepts and Techniques 38
Challenge: Curse of Dimensionality
n Difficult to implement a data cube efficiently given a
large number of dimensions, especially serious in the case of multimedia data cubes
n Many of these attributes are set-oriented instead of single-valued
n Restricting number of dimensions may lead to the modeling of an image at a rather rough, limited, and
imprecise scale
n More research is needed to strike a balance between efficiency and power of representation
January 17, 2001 Data Mining: Concepts and Techniques 39
Chapter 9. Mining Complex Types of Data
n Multidimensional analysis and descriptive mining of
complex data objects
n Mining spatial databases
n Mining multimedia databases
n Mining time-series and sequence data
n Mining text databases
n Mining the World-Wide Web
n Summary
January 17, 2001 Data Mining: Concepts and Techniques 40
Mining Time-Series and Sequence Data
n Time-series database
n Consists of sequences of values or events changing with time
n Data is recorded at regular intervals
n Characteristic time-series components
n Trend, cycle, seasonal, irregular
n Applications
n Financial: stock price, inflation
n Biomedical: blood pressure
n Meteorological: precipitation
January 17, 2001 Data Mining: Concepts and Techniques 41
Mining Time-Series and Sequence Data
Time-series plot
January 17, 2001 Data Mining: Concepts and Techniques 42
Mining Time-Series and Sequence Data: Trend analysis
n A time series can be illustrated as a time-series graph which describes a point moving with the passage of time
n Categories of Time-Series Movements
n Long-term or trend movements (trend curve)
n Cyclic movements or cycle variations, e.g., business cycles
n Seasonal movements or seasonal variations
n i.e, almost identical patterns that a time series appears to follow during corresponding months of successive years.
n Irregular or random movements
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
8
January 17, 2001 Data Mining: Concepts and Techniques 43
Estimation of Trend Curve
n The freehand methodn Fit the curve by looking at the graph
n Costly and barely reliable for large-scaled data miningn The least-square method
n Find the curve minimizing the sum of the squares of the deviation of points on the curve from the corresponding data points
n The moving-average method
n Eliminate cyclic, seasonal and irregular patternsn Loss of end data
n Sensitive to outliers
January 17, 2001 Data Mining: Concepts and Techniques 44
Discovery of Trend in Time-Series (1)
n Estimation of seasonal variations
n Seasonal indexn Set of numbers showing the relative values of a variable during
the months of the yearn E.g., if the sales during October, November, and December are
80%, 120%, and 140% of the average monthly sales for the whole year, respectively, then 80, 120, and 140 are seasonal index numbers for these months
n Deseasonalized datan Data adjusted for seasonal variationsn E.g., divide the original monthly data by the seasonal index
numbers for the corresponding months
January 17, 2001 Data Mining: Concepts and Techniques 45
Discovery of Trend in Time-Series (2)
n Estimation of cyclic variations
n If (approximate) periodicity of cycles occurs, cyclic index can be constructed in much the same manner as seasonal indexes
n Estimation of irregular variations
n By adjusting the data for trend, seasonal and cyclic variations
n With the systematic analysis of the trend, cyclic, seasonal, and irregular components, it is possible to make long- or short-term predictions with reasonable quality
January 17, 2001 Data Mining: Concepts and Techniques 46
Similarity Search in Time-Series Analysis
n Normal database query finds exact match n Similarity search finds data sequences that differ only
slightly from the given query sequencen Two categories of similarity queries
n Whole matching: find a sequence that is similar to the query sequence
n Subsequence matching: find all pairs of similar sequences
n Typical Applicationsn Financial marketn Market basket data analysisn Scientific databasesn Medical diagnosis
January 17, 2001 Data Mining: Concepts and Techniques 47
Data transformation
n Many techniques for signal analysis require the data to be in the frequency domain
n Usually data-independent transformations are usedn The transformation matrix is determined a priori
n E.g., discrete Fourier transform (DFT), discrete wavelet transform (DWT)
n The distance between two signals in the time domain is the same as their Euclidean distance in the frequency domain
n DFT does a good job of concentrating energy in the first few coefficients
n If we keep only first a few coefficients in DFT, we can compute the lower bounds of the actual distance
January 17, 2001 Data Mining: Concepts and Techniques 48
Multidimensional Indexing
n Multidimensional indexn Constructed for efficient accessing using the first few
Fourier coefficients
n Use the index can to retrieve the sequences that are at most a certain small distance away from the query sequence
n Perform postprocessing by computing the actual distance between sequences in the time domain and discard any false matches
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
9
January 17, 2001 Data Mining: Concepts and Techniques 49
Subsequence Matching
n Break each sequence into a set of pieces of window with length w
n Extract the features of the subsequence inside the windown Map each sequence to a “trail” in the feature space
n Divide the trail of each sequence into “subtrails” and represent each of them with minimum bounding rectangle
n Use a multipiece assembly algorithm to search for longer sequence matches
January 17, 2001 Data Mining: Concepts and Techniques 50
Enhanced similarity search methods
n Allow for gaps within a sequence or differences in offsets or amplitudes
n Normalize sequences with amplitude scaling and offset translation
n Two subsequences are considered similar if one lies within an envelope of ε width around the other, ignoring outliers
n Two sequences are said to be similar if they have enough non-overlapping time-ordered pairs of similar subsequences
n Parameters specified by a user or expert: sliding window size, width of an envelope for similarity, maximum gap, and matching fraction
January 17, 2001 Data Mining: Concepts and Techniques 51
Steps for performing a similarity search
n Atomic matchingn Find all pairs of gap-free windows of a small length
that are similarn Window stitching
n Stitch similar windows to form pairs of large similar subsequences allowing gaps between atomic matches
n Subsequence Orderingn Linearly order the subsequence matches to
determine whether enough similar pieces exist
January 17, 2001 Data Mining: Concepts and Techniques 52
Query Languages for Time Sequences
n Time-sequence query languagen Should be able to specify sophisticated queries like
Find all of the sequences that are similar to some sequence in class A, but not similar to any sequence in class Bn Should be able to support various kinds of queries: range
queries, all-pair queries, and nearest neighbor queries
n Shape definition languagen Allows users to define and query the overall shape of time
sequences n Uses human readable series of sequence transitions or macrosn Ignores the specific details
n E.g., the pattern up, Up, UP can be used to describe increasing degrees of rising slopes
n Macros: spike, valley, etc.
January 17, 2001 Data Mining: Concepts and Techniques 53
Sequential Pattern Mining
n Mining of frequently occurring patterns related to time or other sequences
n Sequential pattern mining usually concentrate on symbolic patterns
n Examplesn Renting “Star Wars”, then “Empire Strikes Back”, then
“Return of the Jedi” in that ordern Collection of ordered events within an interval
n Applicationsn Targeted marketingn Customer retentionn Weather prediction
January 17, 2001 Data Mining: Concepts and Techniques 54
Mining Sequences (cont.)
CustId Video sequence1 {(C), (H)}2 {(AB), (C), (DFG)}3 {(CEG)}4 {(C), (DG), (H)}5 {(H)}
Customer-sequence
Sequential patterns with support > 0.25{(C), (H)}{(C), (DG)}
Map Large Itemsets
Large Itemsets MappedID(C) 1(D) 2(G) 3(DG) 4(H) 5
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
10
January 17, 2001 Data Mining: Concepts and Techniques 55
Sequential pattern mining: Cases and Parameters
n Duration of a time sequence Tn Sequential pattern mining can then be confined to the
data within a specified durationn Ex. Subsequence corresponding to the year of 1999n Ex. Partitioned sequences, such as every year, or every
week after stock crashes, or every two weeks before and after a volcano eruption
n Event folding window wn If w = T, time-insensitive frequent patterns are foundn If w = 0 (no event sequence folding), sequential
patterns are found where each event occurs at a distinct time instant
n If 0 < w < T, sequences occurring within the same period w are folded in the analysis
January 17, 2001 Data Mining: Concepts and Techniques 56
Sequential pattern mining: Cases and Parameters (2)
n Time interval, int , between events in the discovered pattern
n int = 0: no interval gap is allowed, i.e., only strictly consecutive sequences are foundn Ex. “Find frequent patterns occurring in consecutive weeks”
n min_int ≤ int ≤ max_int: find patterns that are separated by at least min_int but at most max_intn Ex. “If a person rents movie A, it is likely she will rent movie
B within 30 days” (int ≤ 30)
n int = c ≠ 0: find patterns carrying an exact intervaln Ex. “Every time when Dow Jones drops more than 5%, what
will happen exactly two days later?” (int = 2)
January 17, 2001 Data Mining: Concepts and Techniques 57
Episodes and Sequential Pattern Mining Methods
n Other methods for specifying the kinds of patterns
n Serial episodes: A → B
n Parallel episodes: A & B
n Regular expressions: (A | B)C*(D → E)
n Methods for sequential pattern mining
n Variations of Apriori-like algorithms, e.g., GSP
n Database projection-based pattern growth
n Similar to the frequent pattern growth without candidate generation
January 17, 2001 Data Mining: Concepts and Techniques 58
Periodicity Analysis
n Periodicity is everywhere: tides, seasons, daily power consumption, etc.
n Full periodicityn Every point in time contributes (precisely or
approximately) to the periodicityn Partial periodicit: A more general notion
n Only some segments contribute to the periodicityn Jim reads NY Times 7:00-7:30 am every week day
n Cyclic association rulesn Associations which form cycles
n Methodsn Full periodicity: FFT, other statistical analysis methodsn Partial and cyclic periodicity: Variations of Apriori-like
mining methods
January 17, 2001 Data Mining: Concepts and Techniques 59
Chapter 9. Mining Complex Types of Data
n Multidimensional analysis and descriptive mining of
complex data objects
n Mining spatial databases
n Mining multimedia databases
n Mining time-series and sequence data
n Mining text databases
n Mining the World-Wide Web
n Summary
January 17, 2001 Data Mining: Concepts and Techniques 60
Text Databases and IR
n Text databases (document databases) n Large collections of documents from various sources:
news articles, research papers, books, digital libraries, e-mail messages, and Web pages, library database, etc.
n Data stored is usually semi-structuredn Traditional information retrieval techniques become
inadequate for the increasingly vast amounts of text data
n Information retrievaln A field developed in parallel with database systemsn Information is organized into (a large number of)
documentsn Information retrieval problem: locating relevant
documents based on user input, such as keywords or example documents
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
11
January 17, 2001 Data Mining: Concepts and Techniques 61
Information Retrieval
n Typical IR systems
n Online library catalogs
n Online document management systems
n Information retrieval vs. database systems
n Some DB problems are not present in IR, e.g., update,
transaction management, complex objects
n Some IR problems are not addressed well in DBMS,
e.g., unstructured documents, approximate search
using keywords and relevance
January 17, 2001 Data Mining: Concepts and Techniques 62
Basic Measures for Text Retrieval
n Precision: the percentage of retrieved documents that are in fact relevant to the query (i.e., “correct” responses)
n Recall: the percentage of documents that are relevant to the query and were, in fact, retrieved
|}{||}{}{|
RelevantRetrievedRelevant
precision∩
=
|}{||}{}{|
RetrievedRetrievedRelevant
precision∩
=
January 17, 2001 Data Mining: Concepts and Techniques 63
Keyword-Based Retrieval
n A document is represented by a string, which can be identified by a set of keywords
n Queries may use expressions of keywordsn E.g., car and repair shop, tea or coffee, DBMS but
not Oraclen Queries and retrieval should consider synonyms,
e.g., repair and maintenancen Major difficulties of the model
n Synonymy : A keyword T does not appear anywhere in the document, even though the document is closely related to T, e.g., data mining
n Polysemy : The same keyword may mean different things in different contexts, e.g., mining
January 17, 2001 Data Mining: Concepts and Techniques 64
Similarity-Based Retrieval in Text Databases
n Finds similar documents based on a set of common keywords
n Answer should be based on the degree of relevance based on the nearness of the keywords, relative frequency of the keywords, etc.
n Basic techniquesn Stop list
n Set of words that are deemed “irrelevant”, even though they may appear frequently
n E.g., a, the, of, for, with, etc.
n Stop lists may vary when document set varies
January 17, 2001 Data Mining: Concepts and Techniques 65
Similarity-Based Retrieval in Text Databases (2)
n Word stemn Several words are small syntactic variants of each
other since they share a common word stemn E.g., drug, drugs, drugged
n A term frequency tablen Each entry frequent_table(i, j) = # of occurrences
of the word t i in document di
n Usually, the ratio instead of the absolute number of occurrences is used
n Similarity metrics: measure the closeness of a document to a query (a set of keywords)n Relative term occurrencesn Cosine distance:
|| ||),(
21
2121 vv
vvvvsim
⋅=
January 17, 2001 Data Mining: Concepts and Techniques 66
Latent Semantic Indexing
n Basic idean Similar documents have similar word frequenciesn Difficulty: the size of the term frequency matrix is very largen Use a singular value decomposition (SVD) techniques to reduce
the size of frequency tablen Retain the K most significant rows of the frequency table
n Methodn Create a term frequency matrix, freq_matrixn SVD construction: Compute the singular valued decomposition of
freq_matrix by splitting it into 3 matrices, U, S, Vn Vector identification: For each document d, replace its original
document vector by a new excluding the eliminated termsn Index creation: Store the set of all vectors, indexed by one of a
number of techniques (such as TV-tree)
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
12
January 17, 2001 Data Mining: Concepts and Techniques 67
Other Text Retrieval Indexing Techniques
n Inverted indexn Maintains two hash- or B+-tree indexed tables:
n document_table: a set of document records <doc_id, postings_list>
n term_table: a set of term records, <term, postings_list>n Answer query: Find all docs associated with one or a set of termsn Advantage: easy to implementn Disadvantage: do not handle well synonymy and polysemy, and
posting lists could be too long (storage could be very large)n Signature file
n Associate a signature with each documentn A signature is a representation of an ordered list of terms that
describe the documentn Order is obtained by frequency analysis, stemming and stop lists
January 17, 2001 Data Mining: Concepts and Techniques 68
Types of Text Data Mining
n Keyword-based association analysisn Automatic document classificationn Similarity detection
n Cluster documents by a common authorn Cluster documents containing information from a
common source n Link analysis: unusual correlation between entitiesn Sequence analysis: predicting a recurring eventn Anomaly detection: find information that violates usual
patterns n Hypertext analysis
n Patterns in anchors/linksn Anchor text correlations with linked objects
January 17, 2001 Data Mining: Concepts and Techniques 69
Keyword-based association analysis
n Collect sets of keywords or terms that occur frequently together and then find the association or correlation relationships among them
n First preprocess the text data by parsing, stemming, removing stop words, etc.
n Then evoke association mining algorithmsn Consider each document as a transactionn View a set of keywords in the document as a set of
items in the transactionn Term level association mining
n No need for human effort in tagging documentsn The number of meaningless results and the execution
time is greatly reducedJanuary 17, 2001 Data Mining: Concepts and Techniques 70
Automatic document classification
n Motivationn Automatic classification for the tremendous number of
on-line text documents (Web pages, e-mails, etc.) n A classification problem
n Training set: Human experts generate a training data setn Classification: The computer system discovers the
classification rulesn Application: The discovered rules can be applied to
classify new/unknown documentsn Text document classification differs from the classification of
relational datan Document databases are not structured according to
attribute-value pairs
January 17, 2001 Data Mining: Concepts and Techniques 71
Association-Based Document Classification
n Extract keywords and terms by information retrieval and simple association analysis techniques
n Obtain concept hierarchies of keywords and terms using n Available term classes, such as WordNetn Expert knowledgen Some keyword classification systems
n Classify documents in the training set into class hierarchiesn Apply term association mining method to discover sets of associated
termsn Use the terms to maximally distinguish one class of documents from
othersn Derive a set of association rules associated with each document classn Order the classification rules based on their occurrence frequency
and discriminative powern Used the rules to classify new documents
January 17, 2001 Data Mining: Concepts and Techniques 72
Document Clustering
n Automatically group related documents based on their contents
n Require no training sets or predetermined taxonomies, generate a taxonomy at runtime
n Major stepsn Preprocessing
n Remove stop words, stem, feature extraction, lexical analysis, …
n Hierarchical clusteringn Compute similarities applying clustering algorithms,
…n Slicing
n Fan out controls, flatten the tree to configurable number of levels, …
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
13
January 17, 2001 Data Mining: Concepts and Techniques 73
Chapter 9. Mining Complex Types of Data
n Multidimensional analysis and descriptive mining of
complex data objects
n Mining spatial databases
n Mining multimedia databases
n Mining time-series and sequence data
n Mining text databases
n Mining the World-Wide Web
n Summary
January 17, 2001 Data Mining: Concepts and Techniques 74
Mining the World-Wide Web
n The WWW is huge, widely distributed, global information service center for n Information services: news, advertisements,
consumer information, financial management, education, government, e-commerce, etc.
n Hyper-link informationn Access and usage information
n WWW provides rich sources for data miningn Challenges
n Too huge for effective data warehousing and data mining
n Too complex and heterogeneous: no standards and structure
January 17, 2001 Data Mining: Concepts and Techniques 75
Mining the World-Wide Web
n Growing and changing very rapidly
n Broad diversity of user communitiesn Only a small portion of the information on the Web is truly relevant or
usefuln 99% of the Web information is useless to 99% of Web usersn How can we find high-quality Web pages on a specified topic?
Internet growth
0
5000000
10000000
15000000
20000000
25000000
30000000
35000000
40000000
Sep-6
9
Sep-7
2
Sep-7
5
Sep-7
8
Sep-8
1
Sep-8
4
Sep-8
7
Sep-9
0
Sep-9
3
Sep-9
6
Sep-9
9
Hosts
January 17, 2001 Data Mining: Concepts and Techniques 76
Web search engines
n Index-based: search the Web, index Web pages, and build and store huge keyword-based indices
n Help locate sets of Web pages containing certain keywords
n Deficienciesn A topic of any breadth may easily contain hundreds of
thousands of documents
n Many documents that are highly relevant to a topic may not contain keywords defining them (polysemy )
January 17, 2001 Data Mining: Concepts and Techniques 77
Web Mining: A more challenging task
n Searches for
n Web access patterns
n Web structures
n Regularity and dynamics of Web contents
n Problems
n The “abundance” problem
n Limited coverage of the Web: hidden Web sources, majority of data in DBMS
n Limited query interface based on keyword-oriented search
n Limited customization to individual users
January 17, 2001 Data Mining: Concepts and Techniques 78
Web Mining
Web StructureMining
Web ContentMining
Web PageContent Mining
Search ResultMining
Web UsageMining
General AccessPattern Tracking
CustomizedUsage Tracking
Web Mining Taxonomy
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
14
January 17, 2001 Data Mining: Concepts and Techniques 79
Web Mining
Web StructureMining
Web ContentMining
Web Page Content MiningWeb Page Summarization WebLog (Lakshmanan et.al. 1996),WebOQL(Mendelzon et.al. 1998) …:Web Structuring query languages; Can identify information within given web pages •Ahoy! (Etzioniet.al. 1997):Uses heuristics to distinguish personal home pages from other web pages•ShopBot (Etzioni et.al. 1997): Looks for product prices within web pages
Search ResultMining
Web UsageMining
General AccessPattern Tracking
CustomizedUsage Tracking
Mining the World-Wide Web
January 17, 2001 Data Mining: Concepts and Techniques 80
Web Mining
Mining the World-Wide Web
Web UsageMining
General AccessPattern Tracking
CustomizedUsage Tracking
Web StructureMining
Web ContentMining
Web PageContent Mining Search Result Mining
Search Engine Result Summarization•Clustering Search Result (Leouskiand Croft, 1996, Zamir and Etzioni, 1997): Categorizes documents using phrases in titles and snippets
January 17, 2001 Data Mining: Concepts and Techniques 81
Web Mining
Web ContentMining
Web PageContent Mining
Search ResultMining
Web UsageMining
General AccessPattern Tracking
CustomizedUsage Tracking
Mining the World-Wide Web
Web Structure MiningUsing Links•PageRank (Brin et al., 1998)•CLEVER (Chakrabarti et al., 1998)Use interconnections between web pages to give weight to pages.
Using Generalization•MLDB (1994), VWV (1998)Uses a multi-level database representation of the Web. Counters (popularity) and link lists are used for capturing structure.
January 17, 2001 Data Mining: Concepts and Techniques 82
Web Mining
Web StructureMining
Web ContentMining
Web PageContent Mining
Search ResultMining
Web UsageMining
General Access Pattern Tracking
•Web Log Mining (Zaïane, Xin and Han, 1998)Uses KDD techniques to understand general access patterns and trends.Can shed light on better structure and grouping of resource providers.
CustomizedUsage Tracking
Mining the World-Wide Web
January 17, 2001 Data Mining: Concepts and Techniques 83
Web Mining
Web UsageMining
General AccessPattern Tracking
Customized Usage Tracking
•Adaptive Sites (Perkowitz and Etzioni, 1997)Analyzes access patterns of each user at a time.Web site restructures itself automatically by learning from user access patterns.
Mining the World-Wide Web
Web StructureMining
Web ContentMining
Web PageContent Mining
Search ResultMining
January 17, 2001 Data Mining: Concepts and Techniques 84
Mining the Web's Link Structures
n Finding authoritative Web pages
n Retrieving pages that are not only relevant, but also of high quality, or authoritative on the topic
n Hyperlinks can infer the notion of authority
n The Web consists not only of pages, but also of hyperlinks pointing from one page to another
n These hyperlinks contain an enormous amount of latent human annotation
n A hyperlink pointing to another Web page, this can be considered as the author's endorsement of the other page
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
15
January 17, 2001 Data Mining: Concepts and Techniques 85
Mining the Web's Link Structures
n Problems with the Web linkage structuren Not every hyperlink represents an endorsement
n Other purposes are for navigation or for paid advertisements
n If the majority of hyperlinks are for endorsement, the collective opinion will still dominate
n One authority will seldom have its Web page point to its rival authorities in the same field
n Authoritative pages are seldom particularly descriptive
n Hub n Set of Web pages that provides collections of links to
authoritiesJanuary 17, 2001 Data Mining: Concepts and Techniques 86
HITS (Hyperlink-Induced Topic Search)
n Explore interactions between hubs and authoritative pages
n Use an index-based search engine to form the root setn Many of these pages are presumably relevant to the
search topicn Some of them should contain links to most of the
prominent authoritiesn Expand the root set into a base set
n Include all of the pages that the root-set pages link to, and all of the pages that link to a page in the root set, up to a designated size cutoff
n Apply weight-propagation n An iterative process that determines numerical
estimates of hub and authority weights
January 17, 2001 Data Mining: Concepts and Techniques 87
Systems Based on HITS
n Output a short list of the pages with large hub weights, and the pages with large authority weights for the given search topic
n Systems based on the HITS algorithmn Clever, Google: achieve better quality search results
than those generated by term-index engines such as AltaVista and those created by human ontologists such as Yahoo!
n Difficulties from ignoring textual contexts
n Drifting: when hubs contain multiple topicsn Topic hijacking: when many pages from a single Web
site point to the same single popular site
January 17, 2001 Data Mining: Concepts and Techniques 88
Automatic Classification of Web Documents
n Assign a class label to each document from a set of predefined topic categories
n Based on a set of examples of preclassified documents
n Examplen Use Yahoo!'s taxonomy and its associated
documents as training and test sets n Derive a Web document classification schemen Use the scheme classify new Web documents by
assigning categories from the same taxonomyn Keyword-based document classification methods
n Statistical models
January 17, 2001 Data Mining: Concepts and Techniques 89
Multilayered Web Information Base
n Layer0: the Web itselfn Layer1: the Web page descriptor layer
n Contains descriptive information for pages on the Webn An abstraction of Layer0: substantially smaller but still
rich enough to preserve most of the interesting, general information
n Organized into dozens of semistructured classesn document, person, organization, ads, directory,
sales, software, game, stocks, library_catalog, geographic_data, scientific_data, etc.
n Layer2 and up: various Web directory services constructed on top of Layer1n provide multidimensional, application-specific services
January 17, 2001 Data Mining: Concepts and Techniques 90
Multiple Layered Web Architecture
Generalized Descriptions
More Generalized Descriptions
Layer0
Layer1
Layern
...
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
16
January 17, 2001 Data Mining: Concepts and Techniques 91
Mining the World-Wide Web
Layer-0: Primitive data
Layer-1: dozen database relations representing types of objects (metadata)
document, organization, person, software, game, map, image,…
• document(file_addr, authors, title, publication, publication_date, abstract, language, table_of_contents, category_description, keywords, index, multimedia_attached, num_pages, format, first_paragraphs, size_doc, timestamp, access_frequency, links_out,...)
• person(last_name, first_name, home_page_addr, position, picture_attached, phone, e-mail, office_address, education, research_interests, publications, size_of_home_page, timestamp, access_frequency, ...)
• image(image_addr, author, title, publication_date, category_description, keywords, size, width, height, duration, format, parent_pages, colour_histogram, Colour_layout, Texture_layout, Movement_vector, localisation_vector, timestamp, access_frequency, ...)
January 17, 2001 Data Mining: Concepts and Techniques 92
Mining the World-Wide Web
•doc_brief(file_addr, authors, title, publication, publication_date, abstract, language, category_description, key_words, major_index, num_pages, format, size_doc, access_frequency, links_out)
•person_brief (last_name, first_name, publications,affiliation, e-mail, research_interests, size_home_page, access_frequency)
Layer-2: simplification of layer-1
Layer-3: generalization of layer-2
•cs_doc(file_addr, authors, title, publication, publication_date, abstract, language, category_description, keywords, num_pages, form, size_doc, links_out)
•doc_summary(affiliation, field, publication_year, count, first_author_list, file_addr_list)
•doc_author_brief(file_addr, authors, affiliation, title, publication, pub_date, category_description, keywords, num_pages, format, size_doc, links_out)
•person_summary(affiliation, research_interest, year, num_publications, count)
January 17, 2001 Data Mining: Concepts and Techniques 93
XML and Web Mining
n XML can help to extract the correct descriptors n Standardization would greatly facilitate information
extraction
n Potential problemn XML can help solve heterogeneity for vertical applications, but
the freedom to define tags can make horizontal applications on the Web more heterogeneous
<NAME> eXtensible Markup Language</NAME><RECOM>World-Wide Web Consortium</RECOM><SINCE>1998</SINCE>
<VERSION>1.0</VERSION><DESC>Meta language that facilitates more meaningful and precise declarations of document content</DESC><HOW>Definition of new tags and DTDs</HOW>
January 17, 2001 Data Mining: Concepts and Techniques 94
Benefits of Multi-Layer Meta-Web
n Benefits:
n Multi-dimensional Web info summary analysisn Approximate and intelligent query answeringn Web high-level query answering (WebSQL, WebML)
n Web content and structure miningn Observing the dynamics/evolution of the Web
n Is it realistic to construct such a meta-Web?n Benefits even if it is partially constructed
n Benefits may justify the cost of tool development, standardization and partial restructuring
January 17, 2001 Data Mining: Concepts and Techniques 95
Web Usage Mining
n Mining Web log records to discover user access patterns of Web pages
n Applications
n Target potential customers for electronic commercen Enhance the quality and delivery of Internet
information services to the end user
n Improve Web server system performancen Identify potential prime advertisement locations
n Web logs provide rich information about Web dynamicsn Typical Web log entry includes the URL requested, the
IP address from which the request originated, and a timestamp
January 17, 2001 Data Mining: Concepts and Techniques 96
Techniques for Web usage mining
n Construct multidimensional view on the Weblog databasen Perform multidimensional OLAP analysis to find the top
N users, top N accessed Web pages, most frequently accessed time periods, etc.
n Perform data mining on Weblog records n Find association patterns, sequential patterns, and
trends of Web accessingn May need additional information,e.g., user browsing
sequences of the Web pages in the Web server buffern Conduct studies to
n Analyze system performance, improve system design by Web caching, Web page prefetching, and Web page swapping
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
17
January 17, 2001 Data Mining: Concepts and Techniques 97
Mining the World-Wide Web
n Design of a Web Log Miner
n Web log is filtered to generate a relational databasen A data cube is generated form database
n OLAP is used to drill-down and roll-up in the cuben OLAM is used for mining interesting knowledge
1Data Cleaning
2Data CubeCreation
3OLAP
4Data Mining
Web log Database Data Cube Sliced and dicedcube
Knowledge
January 17, 2001 Data Mining: Concepts and Techniques 98
Chapter 9. Mining Complex Types of Data
n Multidimensional analysis and descriptive mining of
complex data objects
n Mining spatial databases
n Mining multimedia databases
n Mining time-series and sequence data
n Mining text databases
n Mining the World-Wide Web
n Summary
January 17, 2001 Data Mining: Concepts and Techniques 99
Summary (1)
n Mining complex types of data include object data, spatial data, multimedia data, time-series data, text data, andWeb data
n Object data can be mined by multi-dimensional generalization of complex structured data, such as plan mining for flight sequences
n Spatial data warehousing, OLAP and mining facilitates multidimensional spatial analysis and finding spatial associations, classifications and trends
n Multimedia data mining needs content-based retrievaland similarity search integrated with mining methods
January 17, 2001 Data Mining: Concepts and Techniques 100
Summary (2)
n Time-series/sequential data mining includes trend analysis, similarity search in time series, mining sequential patterns and periodicity in time sequence
n Text mining goes beyond keyword-based and similarity -based information retrieval and discovers knowledge from semi-structured data using methods like keyword-based association and document classification
n Web mining includes mining Web link structures to identify authoritative Web pages, the automatic classification of Web documents, building a multilayered Web information base, and Weblog mining
January 17, 2001 Data Mining: Concepts and Techniques 101
References (1)n R. Agrawal, C. Faloutsos , and A. Swami. Efficient similarity search in sequence databases.
In Proc. 4th Int. Conf. Foundations of Data Organization and Alg orithms, Chicago, Oct. 1993.
n R. Agrawal, K.-I. Lin, H.S. Sawhney, and K. Shim. Fast similarity search in the presence of noise, scaling, and translation in time -series databases. VLDB'95, Zurich, Switzerland, Sept. 1995.
n G. Arocena and A. O. Mendelzon. WebOQL : Restructuring documents, databases, and webs. ICDE'98, Orlando, FL, Feb. 1998.
n R. Agrawal, G. Psaila, E. L. Wimmers, and M. Zait. Querying shapes of histories. VLDB'95, Zurich, Switzerland, Sept. 1995.
n R. Agrawal and R. Srikant. Mining sequential patterns. ICDE'95, Taipei, Taiwan, Mar. 1995.n S. Brin and L. Page. The anatomy of a large-scale hypertextual web search engine.
WWW'98, Brisbane, Australia, 1998.n C. Bettini, X. Sean Wang, and S. Jajodia. Mining temporal relationships with multiple
granularities in time sequences. Data Engineering Bulletin, 21:3 2-38, 1998.n R. Baeza-Yates and B. Ribeiro-Neto. Modern Information Retrieval. Addison-Wesley, 1999.n S. Chakrabarti, B. E. Dom, and P. Indyk. Enhanced hypertext classification using hyper-
links. SIGMOD'98, Seattle, WA, June 1998.n S. Chakrabarti, B. E. Dom, S. R. Kumar, P. Raghavan, S. Rajagopalan, A. Tomkins , D.
Gibson, and J. M. Kleinberg. Mining the web's link structure. COMPUTER, 32:60-67, 1999.
January 17, 2001 Data Mining: Concepts and Techniques 102
References (2)n J. Chen, D. DeWitt, F. Tian, and Y. Wang. NiagraCQ: A scalable continuous query system for
internet databases. SIGMOD'00, Dallas, TX, May 2000.n C. Chatfield. The Analysis of Time Series: An Introduction, 3rd ed. Chapman and Hall, 1984.n S. Chakrabarti. Data mining for hypertex: A tutorial survey. SIGKDD Explorations, 1:1 -11,
2000.n S. Deerwester, S. Dumais , G. Furnas, T. Landauer, and R. Harshman. Indexing by latent
semantic analysis. J. American Society for Information Science, 41:391 -407, 1990.n M. Ester, A. Frommelt, H.-P. Kriegel, and J. Sander. Algorithms for claracterization and trend
detection in spatial databases. KDD'98, New York, NY, Aug. 1998.n M.J. Egenhofer. Spatial Query Languages. UMI Research Press, University of Maine, Portland,
Maine, 1989.n M. Ester, H.-P. Kriegel, and J. Sander. Spatial data mining: A database approach. SSD'9 7,
Berlin, Germany, July 1997.n C. Faloutsos . Access methods for text. ACM Comput. Surv., 17:49 -74, 1985.n U. M. Fayyad, S. G. Djorgovski, and N. Weir. Automating the analysis and cataloging of sky
surveys. In U.M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, editors, Advances in Knowledge Discovery and Data Mining, AAAI/MIT Press, 1996.
n R. Feldman and H. Hirsh. Finding associations in collectionds of text. In R. S. Michalski, I.Bratko, and M. Kubat, editors, "Machine Learning and Data Mining: Methods and Applications", John Wiley Sons, 1998.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
18
January 17, 2001 Data Mining: Concepts and Techniques 103
References (3)
n C. Faloutsos and K.-I. Lin. FastMap: A fast algorithm for indexing, data -mining and visualization of traditional and multimedia datasets. SIGMOD'95, San Jose, CA, May 1995.
n D. Florescu, A. Y. Levy, and A. O. Mendelzon. Database techniques for the world -wide web: A survey. SIGMOD Record, 27:59-74, 1998.
n U. M. Fayyad, G. Piatetsky -Shapiro, P. Smyth, and R. Uthurusamy (eds.). Advances in Knowledge Discovery and Data Mining. AAAI/MIT Press, 1996.
n C. Faloutsos , M. Ranganathan, and Y. Manolopoulos . Fast subsequence matching in time-series databases. SIGMOD'94, Minneapolis, Minnesota, May 1994.
n M. Flickner, H. Sawhney, W. Niblack, J. Ashley, B. Dom, Q. Huang, M. Gorkani, J. Hafner, D. Lee, D. Petkovic , S. Steele, and P. Yanker. Query by image and video content: The QBIC system. IEEE Computer, 28:23 -32, 1995.
n S. Guha, R. Rastogi, and K. Shim. Rock: A robust clustering algorithm for categoric al attributes. ICDE'99, Sydney, Australia, Mar. 1999.
n R. H. Gueting. An introduction to spatial database systems. The VLDB Journal, 3:357-400, 1994.
n J. Han, G. Dong, and Y. Yin. Efficient mining of partial periodic patterns in time series database. ICDE'99, Sydney, Australia, Apr. 1999.
n J. Han, K. Koperski, and N. Stefanovic . GeoMiner: A system prototype for spatial data mining. SIGMOD'97, Tucson, Arizona, May 1997.
January 17, 2001 Data Mining: Concepts and Techniques 104
References (4)n J. Han, S. Nishio, H. Kawano, and W. Wang. Generalization -based data mining in object-
oriented databases using an object-cube model. Data and Knowledge Engineering, 25:55-97, 1998.
n J. Han, J. Pei, B. Mortazavi -Asl, Q. Chen, U. Dayal, and M.-C. Hsu. Freespan: Frequent pattern-projected sequential pattern mining. KDD'00, Boston, MA, Aug. 2000.
n J. Han, N. Stefanovic , and K. Koperski. Selective materialization: An efficient method for spatial data cube construction. PAKDD'98. Melbourne, Australia, Apr. 1998.
n J. Han, Q. Yang, and E. Kim. Plan mining by divide-and-conquer. DMKD'99, Philadelphia, PA, May 1999.
n K. Koperski and J. Han. Discovery of spatial association rules in geographic information databases. SSD'95, Portland, Maine, Aug. 1995.
n J. M. Kleinberg. Authoritative sources in a hyperlinked environment. Journal of ACM, 46:604 -632, 1999.
n E. Knorr and R. Ng. Finding aggregate proximity relationships and commonalities in spatial data mining. IEEE Trans. Knowledge and Data Engineering, 8:884-897, 1996.
n J. M. Kleinberg and A. Tomkins . Application of linear algebra in information retrieval and hypertext analysis. PODS'99. Philadelphia, PA, May 1999.
n H. Lu, J. Han, and L. Feng. Stock movement and n -dimensional inter-transaction association rules. DMKD'98, Seattle, WA, June 1998.
January 17, 2001 Data Mining: Concepts and Techniques 105
References (5)n W. Lu, J. Han, and B. C. Ooi. Knowledge discovery in large spatial databases. In Proc.
Far East Workshop Geographic Information Systems, Singapore, June 1993.n D. J. Maguire, M. Goodchild, and D. W. Rhind. Geographical Information Systems:
Principles and Applications. Longman, London, 1992.n H. Miller and J. Han. Geographic Data Mining and Knowledge Discovery. Taylor and
Francis, 2000.n A. O. Mendelzon, G. A. Mihaila, and T. Milo. Querying the world -wide web. Int. Journal of
Digital Libraries, 1:54-67, 1997.n H. Mannila, H Toivonen, and A. I. Verkamo. Discovery of frequent episodes in event
sequences. Data Mining and Knowledge Discovery, 1:259-289, 1997.n A. Natsev, R. Rastogi, and K. Shim. Walrus: A similarity retrieval algorithm for image
databases. SIGMOD'99, Philadelphia, PA, June 1999.n B. Ozden, S. Ramaswamy, and A. Silberschatz. Cyclic association rules. ICDE'98, Orlando,
FL, Feb. 1998.n M. Perkowitz and O. Etzioni. Adaptive web sites: Conceptual cluster mining. IJCAI'99,
Stockholm, Sweden, 1999.n P. Raghavan. Information retrieval algorithms: A survey. In Proc. 1997 ACM-SIAM Symp.
Discrete Algorithms, New Orleans, Louisiana, 1997.
January 17, 2001 Data Mining: Concepts and Techniques 106
References (6)n D. Rafiei and A. Mendelzon. Similarity-based queries for time series data. SIGMOD'97,
Tucson, Arizona, May 1997.n G. Salton. Automatic Text Processing. Addison-Wesley, 1989.
n J. Srivastava, R. Cooley, M. Deshpande, and P. N. Tan. Web usage mining: Discovery and applications of usage patterns from web data. SIGKDD Explorations, 1:12-23, 2000.
n P. Stolorz and C. Dean. Quakefinder: A scalable data mining system for detecting earthquakes from space. KDD'96, Portland, Oregon, Aug. 1996.
n G. Salton and M. McGill. Introduction to Modern Information Retrieval. McGraw-Hill, 1983.
n V. S. Subrahmanian. Principles of Multimedia Database Systems. Morgan Kaufmann, 1998.n C. J. van Rijsbergen. Information Retrieval. Butterworth, 1990.
n K. Wang, S. Zhou, and S. C. Liew. Building hierarchical classifiers using class proximity. VLDB'99, Edinburgh, UK, Sept. 1999.
n B.-K. Yi, H. V. Jagadish, and C. Faloutsos . Efficient retrieval of similar time sequences under time warping. ICDE'98, Orlando, FL, Feb. 1998.
n C. T. Yu and W. Meng. Principles of Database Query Processing for Advanced Applications. Morgan Kaufmann, 1997.
January 17, 2001 Data Mining: Concepts and Techniques 107
References (7)n B.-K. Yi, N. Sidiropoulos , T. Johnson, H. V. Jagadish, C. Faloutsos , and A. Biliris . Online
data mining for co -evolving time sequences. ICDE'00, San Diego, CA, Feb. 2000.n C. Zaniolo, S. Ceri, C. Faloutsos , R. T. Snodgrass, C. S. Subrahmanian, and R. Zicari.
Advanced Database Systems. Morgan Kaufmann, 1997.n O. R. Za"iane and J. Han. Resource and knowledge discovery in global information
systems: A preliminary design and experiment. KDD'95, Montreal, Canada, Aug. 1995.n O. R. Za"iane and J. Han. WebML : Querying the world-wide web for resources and
knowledge. WIDM'98, Bethesda, Maryland, Nov. 1998.n O. R. Za"iane, J. Han, Z. N. Li, J. Y. Chiang, and S. Chee. MultiMedia-Miner: A system
prototype for multimedia data mining. SIGMOD'98, Seattle, WA, June 1998.n O. R. Za"iane, J. Han, and H. Zhu. Mining recurrent items in multimedia with progressive
resolution refinement. ICDE'00, San Diego, CA, Feb. 2000.n M. J. Zaki, N. Lesh, and M. Ogihara. PLANMINE: Sequence mining for plan failures.
KDD'98, New York, NY, Aug. 1998.n X. Zhou, D. Truffet, and J. Han. Efficient polygon amalgamation methods for spatial OLAP
and spatial data mining. SSD'99. Hong Kong, July 1999.n O. R. Za"iane, M. Xin, and J. Han. Discovering Webaccess patterns and trends by applying
OLAP and data mining technology on Web logs. ADL'98, Santa Barbara, CA, Apr. 1998.
January 17, 2001 Data Mining: Concepts and Techniques 108
http://www.cs.sfu.ca/~han/dmbook
Thank you !!!Thank you !!!
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net