grobelnik marko 01

Text-Mining Tutorial: 

Text-Mining Tutorial Marko Grobelnik, Dunja Mladenic J. Stefan Institute, Slovenia

What is Text-Mining?: 

What is Text-Mining? “…finding interesting regularities in large textual datasets…” (Usama Fayad, adapted) …where interesting means: non-trivial, hidden, previously unknown and potentially useful “…finding semantic and abstract information from the surface form of textual data…”

Which areas are active in Text Processing?: 

Which areas are active in Text Processing? Data Analysis Computational Linguistics Search & DB Knowledge Rep. & Reasoning

Tutorial Contents: 

Tutorial Contents Why Text is Easy and Why Tough? Levels of Text Processing Word Level Sentence Level Document Level Document-Collection Level Linked-Document-Collection Level Application Level References to Conferences, Workshops, Books, Products Final Remarks

Why Text is Tough? (M.Hearst 97): 

Why Text is Tough? (M.Hearst 97) Abstract concepts are difficult to represent “Countless” combinations of subtle, abstract relationships among concepts Many ways to represent similar concepts E.g. space ship, flying saucer, UFO Concepts are difficult to visualize High dimensionality Tens or hundreds of thousands of features

Why Text is Easy? (M.Hearst 97): 

Why Text is Easy? (M.Hearst 97) Highly redundant data …most of the methods count on this property Just about any simple algorithm can get “good” results for simple tasks: Pull out “important” phrases Find “meaningfully” related words Create some sort of summary from documents

Levels of Text Processing 1/6: 

Levels of Text Processing 1/6 Word Level Words Properties Stop-Words Stemming Frequent N-Grams Thesaurus (WordNet) Sentence Level Document Level Document-Collection Level Linked-Document-Collection Level Application Level

Words Properties: 

Words Properties Relations among word surface forms and their senses: Homonomy: same form, but different meaning (e.g. bank: river bank, financial institution) Polysemy: same form, related meaning (e.g. bank: blood bank, financial institution) Synonymy: different form, same meaning (e.g. singer, vocalist) Hyponymy: one word denotes a subclass of an another (e.g. breakfast, meal) Word frequencies in texts have power distribution: …small number of very frequent words …big number of low frequency words

Stop-words: 

Stop-words Stop-words are words that from non-linguistic view do not carry information …they have mainly functional role …usually we remove them to help the methods to perform better Natural language dependent – examples: English: A, ABOUT, ABOVE, ACROSS, AFTER, AGAIN, AGAINST, ALL, ALMOST, ALONE, ALONG, ALREADY, ALSO, ... Slovenian: A, AH, AHA, ALI, AMPAK, BAJE, BODISI, BOJDA, BRŽKONE, BRŽČAS, BREZ, CELO, DA, DO, ... Croatian: A, AH, AHA, ALI, AKO, BEZ, DA, IPAK, NE, NEGO, ...

Slide10: 

Original text Information Systems Asia Web - provides research, IS-related commercial materials, interaction, and even research sponsorship by interested corporations with a focus on Asia Pacific region. Survey of Information Retrieval - guide to IR, with an emphasis on web-based projects. Includes a glossary, and pointers to interesting papers. After the stop-words removal Information Systems Asia Web provides research IS-related commercial materials interaction research sponsorship interested corporations focus Asia Pacific region Survey Information Retrieval guide IR emphasis web-based projects Includes glossary pointers interesting papers

Stemming (I): 

Stemming (I) Different forms of the same word are usually problematic for text data analysis, because they have different spelling and similar meaning (e.g. learns, learned, learning,…) Stemming is a process of transforming a word into its stem (normalized form)

Stemming (II): 

Stemming (II) For English it is not a big problem - publicly available algorithms give good results Most widely used is Porter stemmer at http://www.tartarus.org/~martin/PorterStemmer/ E.g. in Slovenian language 10-20 different forms correspond to the same word: E.g. (“to laugh” in Slovenian): smej, smejal, smejala, smejale, smejali, smejalo, smejati, smejejo, smejeta, smejete, smejeva, smeješ, smejemo, smejiš, smeje, smejoč, smejta, smejte, smejva

Example cascade rules used in English Porter stemmer: 

Example cascade rules used in English Porter stemmer ATIONAL -> ATE relational -> relate TIONAL -> TION conditional -> condition ENCI -> ENCE valenci -> valence ANCI -> ANCE hesitanci -> hesitance IZER -> IZE digitizer -> digitize ABLI -> ABLE conformabli -> conformable ALLI -> AL radicalli -> radical ENTLI -> ENT differentli -> different ELI -> E vileli - > vile OUSLI -> OUS analogousli -> analogous

Rules automatically obtained for Slovenian language: 

Rules automatically obtained for Slovenian language Machine Learning applied on Multext-East dictionary (http://nl.ijs.si/ME/) Two example rules: Remove the ending “OM” if 3 last char is any of HOM, NOM, DOM, SOM, POM, BOM, FOM. For instance, ALAHOM, AMERICANOM, BENJAMINOM, BERLINOM, ALFREDOM, BEOGRADOM, DICKENSOM, JEZUSOM, JOSIPOM, OLIMPOM,... but not ALEKSANDROM (ROM -> ER) Replace CEM by EC. For instance, ARABCEM, BAVARCEM, BOVCEM, EVROPEJCEM, GORENJCEM, ... but not FRANCEM (remove EM)

Phrases in the form of frequent N-Grams: 

Phrases in the form of frequent N-Grams Simple way for generating phrases are frequent n-grams: N-Gram is a sequence of n consecutive words (e.g. “machine learning” is 2-gram) “Frequent n-grams” are the ones which appear in all observed documents MinFreq or more times N-grams are interesting because of the simple and efficient dynamic programming algorithm: Given: Set of documents (each document is a sequence of words), MinFreq (minimal n-gram frequency), MaxNGramSize (maximal n-gram length) for Len = 1 to MaxNGramSize do Generate candidate n-grams as sequences of words of size Len using frequent n-grams of length Len-1 Delete candidate n-grams with the frequency less then MinFreq

Generation of frequent n-grams for 50,000 documents from Yahoo: 

Generation of frequent n-grams for 50,000 documents from Yahoo # features 1.6M 1.4M 1.2M 1M 800 000 600 000 400 000 200 000 0 1-grams 2-grams 3-grams 4-grams 5-grams 318K->70K 1.4M->207K 742K->243K 309K->252K 262K->256K

Slide17: 

Original text on the Yahoo Web page: 1.Top:Reference:Libraries:Library and Information Science:Information Retrieval 2.UK Only 3.Idomeneus - IR \& DB repository - These pages mostly contain IR related resources such as test collections, stop lists, stemming algorithms, and links to other IR sites. 4.University of Glasgow - Information Retrieval Group - information on the resources and people in the Glasgow IR group. 5.Centre for Intelligent Information Retrieval (CIIR). 6.Information Systems Asia Web - provides research, IS-related commercial materials, interaction, and even research sponsorship by interested corporations with a focus on Asia Pacific region. 7.Seminar on Cataloging Digital Documents 8.Survey of Information Retrieval - guide to IR, with an emphasis on web-based projects. Includes a glossary, and pointers to interesting papers. 9.University of Dortmund - Information Retrieval Group Document represented by n-grams: 1."REFERENCE LIBRARIES LIBRARY INFORMATION SCIENCE (\#3 LIBRARY INFORMATION SCIENCE) INFORMATION RETRIEVAL (\#2 INFORMATION RETRIEVAL)" 2."UK" 3."IR PAGES IR RELATED RESOURCES COLLECTIONS LISTS LINKS IR SITES" 4."UNIVERSITY GLASGOW INFORMATION RETRIEVAL (\#2 INFORMATION RETRIEVAL) GROUP INFORMATION RESOURCES (\#2 INFORMATION RESOURCES) PEOPLE GLASGOW IR GROUP" 5."CENTRE INFORMATION RETRIEVAL (\#2 INFORMATION RETRIEVAL)" 6."INFORMATION SYSTEMS ASIA WEB RESEARCH COMMERCIAL MATERIALS RESEARCH ASIA PACIFIC REGION" 7."CATALOGING DIGITAL DOCUMENTS" 8."INFORMATION RETRIEVAL (\#2 INFORMATION RETRIEVAL) GUIDE IR EMPHASIS INCLUDES GLOSSARY INTERESTING" 9."UNIVERSITY INFORMATION RETRIEVAL (\#2 INFORMATION RETRIEVAL) GROUP"

WordNet – a database of lexical relations: 

WordNet – a database of lexical relations WordNet is the most well developed and widely used lexical database for English …it consist from 4 databases (nouns, verbs, adjectives, and adverbs) Each database consists from sense entries consisting from a set of synonyms, e.g.: musician, instrumentalist, player person, individual, someone life form, organism, being

WordNet relations: 

WordNet relations Each WordNet entry is connected with other entries in a graph through relations. Relations in the database of nouns:

Levels of Text Processing 2/6: 

Levels of Text Processing 2/6 Word Level Sentence Level Document Level Document-Collection Level Linked-Document-Collection Level Application Level

Levels of Text Processing 3/6: 

Levels of Text Processing 3/6 Word Level Sentence Level Document Level Summarization Single Document Visualization Text Segmentation Document-Collection Level Linked-Document-Collection Level Application Level

Summarization: 

Summarization

Summarization: 

Summarization Task: the task is to produce shorter, summary version of an original document. Two main approaches to the problem: Knowledge rich – performing semantic analysis, representing the meaning and generating the text satisfying length restriction Selection based

Selection based summarization: 

Selection based summarization Three main phases: Analyzing the source text Determining its important points Synthesizing an appropriate output Most methods adopt linear weighting model – each text unit (sentence) is assessed by: Weight(U)=LocationInText(U)+CuePhrase(U)+Statistics(U)+AdditionalPresence(U) …a lot of heuristics and tuning of parameters (also with ML) …output consists from topmost text units (sentences)

Slide25: 

Selected units Selection threshold Example of selection based approach from MS Word

Visualization of a single document: 

Visualization of a single document

Why visualization of a single document is hard?: 

Why visualization of a single document is hard? Visualizing of big text corpora is easier task because of the big amount of information: ...statistics already starts working ...most known approaches are statistics based Visualization of a single (possibly short) document is much harder task because: ...we can not count of statistical properties of the text (lack of data) ...we must rely on syntactical and logical structure of the document

Simple approach: 

Simple approach The text is split into the sentences. Each sentence is deep-parsed into its logical form we are using Microsoft’s NLPWin parser Anaphora resolution is performed on all sentences ...all ‘he’, ‘she’, ‘they’, ‘him’, ‘his’, ‘her’, etc. references to the objects are replaced by its proper name From all the sentences we extract [Subject-Predicate-Object triples] (SPO) SPOs form links in the graph ...finally, we draw a graph.

Slide29: 

Clarence Thomas article

Slide30: 

Alan Greenspan article

Text Segmentation: 

Text Segmentation

Text Segmentation: 

Text Segmentation Problem: divide text that has no given structure into segments with similar content Example applications: topic tracking in news (spoken news) identification of topics in large, unstructured text databases

Algorithm for text segmentation: 

Algorithm for text segmentation Algorithm: Divide text into sentences Represent each sentence with words and phrases it contains Calculate similarity between the pairs of sentences Find a segmentation (sequence of delimiters), so that the similarity between the sentences inside the same segment is maximized and minimized between the segments …the approach can be defined either as optimization problem or as sliding window

Levels of Text Processing 4/6: 

Levels of Text Processing 4/6 Word Level Sentence Level Document Level Document-Collection Level Representation Feature Selection Document Similarity Representation Change (LSI) Categorization (flat, hierarchical) Clustering (flat, hierarchical) Visualization Information Extraction Linked-Document-Collection Level Application Level

Representation: 

Representation

Bag-of-words document representation: 

Bag-of-words document representation

Word weighting: 

Word weighting In bag-of-words representation each word is represented as a separate variable having numeric weight. The most popular weighting schema is normalized word frequency TFIDF: Tf(w) – term frequency (number of word occurrences in a document) Df(w) – document frequency (number of documents containing the word) N – number of all documents Tfidf(w) – relative importance of the word in the document The word is more important if it appears several times in a target document The word is more important if it appears in less documents

Example document and its vector representation: 

Example document and its vector representation TRUMP MAKES BID FOR CONTROL OF RESORTS Casino owner and real estate Donald Trump has offered to acquire all Class B common shares of Resorts International Inc, a spokesman for Trump said. The estate of late Resorts chairman James M. Crosby owns 340,783 of the 752,297 Class B shares. Resorts also has about 6,432,000 Class A common shares outstanding. Each Class B share has 100 times the voting power of a Class A share, giving the Class B stock about 93 pct of Resorts' voting power. [RESORTS:0.624] [CLASS:0.487] [TRUMP:0.367] [VOTING:0.171] [ESTATE:0.166] [POWER:0.134] [CROSBY:0.134] [CASINO:0.119] [DEVELOPER:0.118] [SHARES:0.117] [OWNER:0.102] [DONALD:0.097] [COMMON:0.093] [GIVING:0.081] [OWNS:0.080] [MAKES:0.078] [TIMES:0.075] [SHARE:0.072] [JAMES:0.070] [REAL:0.068] [CONTROL:0.065] [ACQUIRE:0.064] [OFFERED:0.063] [BID:0.063] [LATE:0.062] [OUTSTANDING:0.056] [SPOKESMAN:0.049] [CHAIRMAN:0.049] [INTERNATIONAL:0.041] [STOCK:0.035] [YORK:0.035] [PCT:0.022] [MARCH:0.011]

Feature Selection: 

Feature Selection

Feature subset selection: 

Feature subset selection

Feature subset selection : 

Feature subset selection Select only the best features (different ways to define “the best”-different feature scoring measures) the most frequent the most informative relative to the all class values the most informative relative to the positive class value,…

Scoring individual feature: 

Scoring individual feature InformationGain: CrossEntropyTxt: MutualInfoTxt: WeightOfEvidTxt: OddsRatio: Frequency:

Example of the best features: 

Example of the best features Odds Ratio feature score [P(F|pos), P(F|neg)] IR 5.28 [0.075, 0.0004] INFORMATION RETRIEVAL 5.13... RETRIEVAL 4.77 [0.075, 0.0007] GLASGOW 4.72 [0.03, 0.0003] ASIA 4.32 [0.03, 0.0004] PACIFIC 4.02 [0.015, 0.0003] INTERESTING 4.02[0.015, 0.0003] EMPHASIS 4.02 [0.015, 0.0003] GROUP 3.64 [0.045, 0.0012] MASSACHUSETTS 3.46 [0.015, ...] COMMERCIAL 3.46 [0.015,0.0005] REGION 3.1 [0.015, 0.0007] Information Gain feature score [P(F|pos), P(F|neg)] LIBRARY 0.46 [0.015, 0.091] PUBLIC 0.23 [0, 0.034] PUBLIC LIBRARY 0.21 [0, 0.029] UNIVERSITY 0.21 [0.045, 0.028] LIBRARIES 0.197 [0.015, 0.026] INFORMATION 0.17 [0.119, 0.021] REFERENCES 0.117 [0.015, 0.012] RESOURCES 0.11 [0.029, 0.0102] COUNTY 0.096 [0, 0.0089] INTERNET 0.091 [0, 0.00826] LINKS 0.091 [0.015, 0.00819] SERVICES 0.089 [0, 0.0079]

Document Similarity: 

Document Similarity

Cosine similarity between document vectors: 

Cosine similarity between document vectors Each document is represented as a vector of weights D = <x> Similarity between vectors is estimated by the similarity between their vector representations (cosine of the angle between vectors):

Representation Change: Latent Semantic Indexing: 

Representation Change: Latent Semantic Indexing

Latent Semantic Indexing: 

Latent Semantic Indexing LSI is a statistical technique that attempts to estimate the hidden content structure within documents: …it uses linear algebra technique Singular-Value-Decomposition (SVD) …it discovers statistically most significant co-occurences of terms

LSI Example: 

LSI Example Correlation matrix Original document-term mantrix Rescaled document matrix, Reduced into two dimensions High correlation although d2 and d3 don’t share any word

Text Categorization: 

Text Categorization

Document categorization: 

Document categorization labeled documents unlabeled document document category (label) ??? Machine learning

Automatic Document Categorization Task: 

Automatic Document Categorization Task Given is a set of documents labeled with content categories. The goal is: to build a model which would automatically assign right content categories to new unlabeled documents. Content categories can be: unstructured (e.g., Reuters) or structured (e.g., Yahoo, DMoz, Medline)

Algorithms for learning document classifiers: 

Algorithms for learning document classifiers Popular algorithms for text categorization: Support Vector Machines Logistic Regression Perceptron algorithm Naive Bayesian classifier Winnow algorithm Nearest Neighbour ....

Perceptron algorithm: 

Perceptron algorithm Input: set of pre-classified documents Output: model, one weight for each word from the vocabulary Algorithm: initialize the model by setting word weights to 0 iterate through documents N times classify the document X represented as bag-of-words predict positive class else predict negative class if document classification is wrong then adjust weights of all words occurring in the document sign(positive) = 1 sign(negative) =-1

Measuring success - Model quality estimation : 

Measuring success - Model quality estimation Classification accuracy Break-even point (precision=recall) F-measure (precision, recall = sensitivity) The truth, and ..the whole truth

Reuters dataset – Categorization to flat categories: 

Reuters dataset – Categorization to flat categories Documents classified by editors into one or more categories Publicly available set of Reuter news mainly from 1987: 120 categories giving the document content, such as: earn, acquire, corn, rice, jobs, oilseeds, gold, coffee, housing, income,... …from 2000 is available new dataset of 830,000 Reuters documents available fo research

Distribution of documents (Reuters-21578): 

Distribution of documents (Reuters-21578)

Example of Perceptron model for Reuters category “Acquisition”: 

Example of Perceptron model for Reuters category “Acquisition” Feature Positive Class Weight ----------------------------- STAKE 11.5 MERGER 9.5 TAKEOVER 9 ACQUIRE 9 ACQUIRED 8 COMPLETES 7.5 OWNERSHIP 7.5 SALE 7.5 OWNERSHIP 7.5 BUYOUT 7 ACQUISITION 6.5 UNDISCLOSED 6.5 BUYS 6.5 ASSETS 6 BID 6 BP 6 DIVISION 5.5 …

SVM, Perceptron & Winnow text categorization performance on Reuters-21578 with different representations: 

SVM, Perceptron & Winnow text categorization performance on Reuters-21578 with different representations

Comparison on using SVM on stemmed 1-grams with related results: 

Comparison on using SVM on stemmed 1-grams with related results

Text Categorization into hierarchy of categories: 

Text Categorization into hierarchy of categories There are several hierarchies (taxonomies) of textual documents: Yahoo, DMoz, Medline, … Different people use different approaches: …series of hierarchically organized classifiers …set of independent classifiers just for leaves …set of independent classifiers for all nodes

Yahoo! hierarchy (taxonomy): 

Yahoo! hierarchy (taxonomy) human constructed hierarchy of Web-documents exists in several languages (we use English) easy to access and regularly updated captures most of the Web topics English version includes over 2M pages categorized into 50,000 categories contains about 250Mb of HTML files

Slide62: 

Document to categorize: CFP for CoNLL-2000

Slide63: 

Some predicted categories

Slide64: 

labeled documents (from Yahoo! hierarchy) Feature construction unlabeled document document category (label) ?? System architecture vectors of n-grams Document Classifier Subproblem definition Feature selection Classifier construction Web

Content categories: 

Content categories For each content category generate a separate classifier that predicts probability for a new document to belong to its category

Considering promising categories only (classification by Naive Bayes): 

Considering promising categories only (classification by Naive Bayes) Document is represented as a set of word sequences W Each classifier has two distributions: P(W|pos), P(W|neg) Promising category: calculated P(pos|Doc) is high meaning that the classifier has P(W|pos)>0 for at least some W from the document (otherwise, the prior probability is returned, P(neg) is about 0.90)

Summary of experimental results: 

Summary of experimental results

Document Clustering: 

Document Clustering

Document Clustering: 

Document Clustering Clustering is a process of finding natural groups in data in a unsupervised way (no class labels preassigned to documents) Most popular clustering methods are: K-Means clustering Agglomerative hierarchical clustering EM (Gaussian Mixture) …

K-Means clustering: 

K-Means clustering Given: set of documents (e.g. TFIDF vectors), distance measure (e.g. cosine) K (number of groups) For each of K groups initialize its centroid with a random document While not converging Each document is assigned to the nearest group (represented by its centroid) For each group calculate new centroid (group mass point, average document in the group)

Visualization: 

Visualization

Why text visualization?: 

Why text visualization? ...to have a top level view of the topics in the corpora ...to see relationships between the topics in the corpora ...to understand better what’s going on in the corpora ...to show highly structured nature of textual contents in a simplified way ...to show main dimensions of highly dimensional space of textual documents ...because it’s fun!

Examples of Text Visualization: 

Examples of Text Visualization Text visualizations WebSOM ThemeScape Graph-Based Visualization Tiling-Based Visualization … … collection of approaches at http://nd.loopback.org/hyperd/zb/

WebSOM: 

WebSOM Self-Organizing Maps for Internet Exploration An ordered map of the information space is provided: similar documents lie near each other on the map …algorithm that automatically organizes the documents onto a two-dimensional grid so that related documents appear close to each other … based on Kohonen’s Self-Organizing Maps Demo at http://websom.hut.fi/websom/

WebSOM visualization: 

WebSOM visualization

ThemeScape: 

ThemeScape Graphically displays images based on word similarities and themes in text Themes within the document spaces appear on the computer screen as a relief map of natural terrain The mountains in indicate where themes are dominant - valleys indicate weak themes Themes close in content will be close visually based on the many relationships within the text spaces.  … similar techniques for visualizing stocks (http://www.webmap.com./trademapdemo.html)

ThemeScape Document visualization: 

ThemeScape Document visualization

Graph based visualization: 

Graph based visualization The sketch of the algorithm: Documents are transformed into the bag-of-words sparse-vectors representation Words in the vectors are weighted using TFIDF K-Means clustering algorithm splits the documents into K groups Each group consists from similar documents Documents are compared using cosine similarity K groups form a graph: Groups are nodes in graph; similar groups are linked Each group is represented by characteristic keywords Using simulated annealing draw a graph

Example of visualizing Eu IST projects corpora: 

Example of visualizing Eu IST projects corpora Corpus of 1700 Eu IST projects descriptions Downloaded from the web http://www.cordis.lu/ Each document is few hundred words long describing one project financed by EC ...the idea is to understand the structure and relations between the areas EC is funding through the projects ...the following slides show different visualizations with the graph based approach

Slide80: 

Graph based visualization of 1700 IST project descriptions into 2 groups

Slide81: 

Graph based visualization of 1700 IST project descriptions into 3 groups

Slide82: 

Graph based visualization of 1700 IST project descriptions into 10 groups

Slide83: 

Graph based visualization of 1700 IST project descriptions into 20 groups

How do we extract keywords?: 

How do we extract keywords? Characteristic keywords for a group of documents are the most highly weighted words in the centroid of the cluster ...centroid of the cluster could be understood as an “average document” for specific group of documents ...we are using the effect provided by the TFIDF weighting schema for weighting the importance of the words ...efficient solution

TFIDF words weighting in vector representation: 

TFIDF words weighting in vector representation In Information Retrieval, the most popular weighting schema is normalized word frequency TFIDF: Tf(w) – term frequency (number of word occurrences in a document) Df(w) – document frequency (number of documents containing the word) N – number of all documents Tfidf(w) – relative importance of the word in the document

Tiling based visualization: 

Tiling based visualization The sketch of the algorithm: Documents are transformed into the bag-of-words sparse-vectors representation Words in the vectors are weighted using TFIDF Hierarchical top-down two-wise K-Means clustering algorithm builds a hierarchy of clusters The hierarchy is an artificial equivalent of hierarchical subject index (Yahoo like) The leaf nodes of the hierarchy (bottom level) are used to visualize the documents Each leaf is represented by characteristic keywords Each hierarchical binary split splits recursively the rectangular area into two sub-areas

Slide87: 

Tiling based visualization of 1700 IST project descriptions into 2 groups

Slide88: 

Tiling based visualization of 1700 IST project descriptions into 3 groups

Slide89: 

Tiling based visualization of 1700 IST project descriptions into 4 groups

Slide90: 

Tiling based visualization of 1700 IST project descriptions into 5 groups

Slide91: 

Tiling visualization (up to 50 documents per group) of 1700 IST project descriptions (60 groups)

ThemeRiver: 

ThemeRiver System that visualizes thematic variations over time across a collection of documents The “river” flows through time, changing width to visualize changes in the thematic strength of documents temporally collocated Themes or topics are represented as colored “currents” flowing within the river that narrow or widen to indicate decreases or increases in the strength of a topic in associated documents at a specific point in time. Described in paper at http://www.pnl.gov/infoviz/themeriver99.pdf

ThemeRiver topic stream: 

ThemeRiver topic stream

Information Extraction: 

Information Extraction (slides borrowed from William Cohen’s Tutorial on IE)

Extracting Job Openings from the Web : 

Extracting Job Openings from the Web

Slide96: 

IE from Research Papers

What is “Information Extraction”: 

What is “Information Extraction” Filling slots in a database from sub-segments of text. As a task: October 14, 2002, 4:00 a.m. PT For years, Microsoft Corporation CEO Bill Gates railed against the economic philosophy of open-source software with Orwellian fervor, denouncing its communal licensing as a "cancer" that stifled technological innovation. Today, Microsoft claims to "love" the open-source concept, by which software code is made public to encourage improvement and development by outside programmers. Gates himself says Microsoft will gladly disclose its crown jewels--the coveted code behind the Windows operating system--to select customers. "We can be open source. We love the concept of shared source," said Bill Veghte, a Microsoft VP. "That's a super-important shift for us in terms of code access.“ Richard Stallman, founder of the Free Software Foundation, countered saying… NAME TITLE ORGANIZATION

What is “Information Extraction”: 

What is “Information Extraction” Filling slots in a database from sub-segments of text. As a task: October 14, 2002, 4:00 a.m. PT For years, Microsoft Corporation CEO Bill Gates railed against the economic philosophy of open-source software with Orwellian fervor, denouncing its communal licensing as a "cancer" that stifled technological innovation. Today, Microsoft claims to "love" the open-source concept, by which software code is made public to encourage improvement and development by outside programmers. Gates himself says Microsoft will gladly disclose its crown jewels--the coveted code behind the Windows operating system--to select customers. "We can be open source. We love the concept of shared source," said Bill Veghte, a Microsoft VP. "That's a super-important shift for us in terms of code access.“ Richard Stallman, founder of the Free Software Foundation, countered saying… NAME TITLE ORGANIZATION Bill Gates CEO Microsoft Bill Veghte VP Microsoft Richard Stallman founder Free Soft.. IE

What is “Information Extraction”: 

What is “Information Extraction” Information Extraction = segmentation + classification + clustering + association As a family of techniques: October 14, 2002, 4:00 a.m. PT For years, Microsoft Corporation CEO Bill Gates railed against the economic philosophy of open-source software with Orwellian fervor, denouncing its communal licensing as a "cancer" that stifled technological innovation. Today, Microsoft claims to "love" the open-source concept, by which software code is made public to encourage improvement and development by outside programmers. Gates himself says Microsoft will gladly disclose its crown jewels--the coveted code behind the Windows operating system--to select customers. "We can be open source. We love the concept of shared source," said Bill Veghte, a Microsoft VP. "That's a super-important shift for us in terms of code access.“ Richard Stallman, founder of the Free Software Foundation, countered saying… Microsoft Corporation CEO Bill Gates Microsoft Gates Microsoft Bill Veghte Microsoft VP Richard Stallman founder Free Software Foundation aka “named entity extraction”

What is “Information Extraction”: 

What is “Information Extraction” Information Extraction = segmentation + classification + association + clustering As a family of techniques: October 14, 2002, 4:00 a.m. PT For years, Microsoft Corporation CEO Bill Gates railed against the economic philosophy of open-source software with Orwellian fervor, denouncing its communal licensing as a "cancer" that stifled technological innovation. Today, Microsoft claims to "love" the open-source concept, by which software code is made public to encourage improvement and development by outside programmers. Gates himself says Microsoft will gladly disclose its crown jewels--the coveted code behind the Windows operating system--to select customers. "We can be open source. We love the concept of shared source," said Bill Veghte, a Microsoft VP. "That's a super-important shift for us in terms of code access.“ Richard Stallman, founder of the Free Software Foundation, countered saying… Microsoft Corporation CEO Bill Gates Microsoft Gates Microsoft Bill Veghte Microsoft VP Richard Stallman founder Free Software Foundation

What is “Information Extraction”: 

What is “Information Extraction” Information Extraction = segmentation + classification + association + clustering As a family of techniques: October 14, 2002, 4:00 a.m. PT For years, Microsoft Corporation CEO Bill Gates railed against the economic philosophy of open-source software with Orwellian fervor, denouncing its communal licensing as a "cancer" that stifled technological innovation. Today, Microsoft claims to "love" the open-source concept, by which software code is made public to encourage improvement and development by outside programmers. Gates himself says Microsoft will gladly disclose its crown jewels--the coveted code behind the Windows operating system--to select customers. "We can be open source. We love the concept of shared source," said Bill Veghte, a Microsoft VP. "That's a super-important shift for us in terms of code access.“ Richard Stallman, founder of the Free Software Foundation, countered saying… Microsoft Corporation CEO Bill Gates Microsoft Gates Microsoft Bill Veghte Microsoft VP Richard Stallman founder Free Software Foundation

What is “Information Extraction”: 

What is “Information Extraction” Information Extraction = segmentation + classification + association + clustering As a family of techniques: October 14, 2002, 4:00 a.m. PT For years, Microsoft Corporation CEO Bill Gates railed against the economic philosophy of open-source software with Orwellian fervor, denouncing its communal licensing as a "cancer" that stifled technological innovation. Today, Microsoft claims to "love" the open-source concept, by which software code is made public to encourage improvement and development by outside programmers. Gates himself says Microsoft will gladly disclose its crown jewels--the coveted code behind the Windows operating system--to select customers. "We can be open source. We love the concept of shared source," said Bill Veghte, a Microsoft VP. "That's a super-important shift for us in terms of code access.“ Richard Stallman, founder of the Free Software Foundation, countered saying… Microsoft Corporation CEO Bill Gates Microsoft Gates Microsoft Bill Veghte Microsoft VP Richard Stallman founder Free Software Foundation * * * *

IE in Context: 

IE in Context Create ontology Segment Classify Associate Cluster Load DB Spider Query, Search Data mine IE Document collection Database Filter by relevance Label training data Train extraction models

Typical approaches to IE: 

Typical approaches to IE Hand-built rules/models for extraction Machine learning used on manually labeled data: Classification problem on sliding window …examples are taken from sliding window …models classify short segments of text such as title, name, institution, … …limitation of sliding window because it does not take into account sequential nature of text Training stochastic finite state machines (e.g. HMM) …probabilistic reconstruction of parsing sequence

Levels of Text Processing 5/6: 

Levels of Text Processing 5/6 Word Level Sentence Level Document Level Document-Collection Level Linked-Document-Collection Level Labelling unlabeled data Co-training Application Level

Labelling unlabeled data: 

Labelling unlabeled data

Using unlabeled data (Nigam et al., ML Journal 2000): 

Using unlabeled data (Nigam et al., ML Journal 2000) small number of labeled documents and a large pool of unlabeled documents, eg., classify an article in one of the 20 News groups, classify Web page as student, faculty, course, project,... approach description (EM + Naive Bayes): train a classifier with only labeled documents, assign probabilistically-weighted class labels to unlabeled documents, train a new classifier using all the documents iterate until the classifier remains unchanged

Using Unlabeled Data with Expectation-Maximization (EM): 

E-step: Estimate labels of unlabeled documents M-step: Use all documents to rebuild classifier Naive Bayes Initialize: Learn from labeled only Using Unlabeled Data with Expectation-Maximization (EM) Guarantees local maximum a posteriori parameters

Co-training: 

Co-training

Co-training: 

Co-training Better performance on labelling unlabeled data compared to EM approach

Bootstrap Learning to Classify Web Pages (co-training): 

Bootstrap Learning to Classify Web Pages (co-training) Given: set of documents where each document is described by two independent sets of attributes (e.g. text + hyperlinks) Document content Hyperlink, pointing to the document

Levels of Text Processing 6/6: 

Levels of Text Processing 6/6 Word Level Sentence Level Document Level Document-Collection Level Linked-Document-Collection Level Application Level Question-Answering Mixing Data Sources (KDD Cup 2003)

Question-Answering: 

Question-Answering

Question Answering: 

Question Answering QA Systems are returning short and accurate replies to the well-formed natural language questions such as: What is the hight of Mount Everest? After which animal is the Canary Island named? How many liters are there in to a gallon? QA Systems can be classified into following levels of sophistication: Slot-filling – easy questions, IE technology Limited-Domain – handcrafted dictionaries & ontologies Open-domain – IR, IE, NL parsing, inferencing

Question Answering Architecture: 

Question Answering Architecture Parse and classify question Generate keyword query Retrieve documents from IR system Segment responses Rank segments Parse top segments Match segment with question Rank and prepare answer Answers Question Question taxonomy and supervised learner WordNet expansion, verb transformation, noun phrase identification Identification of sentence and paragraph boundaries, finding density of query terms in segment, TextTiling

Question Answering Example: 

Question Answering Example Example question and answer: Q:What is the color of grass? A: Green. …the answer may come from the document saying: “grass is green” without mentioning “color” with the help of WordNet having hypernym hierarchy: green, chromatic color, color, visual property, property

Mixing Data Sources (KDD Cup 2003): 

Mixing Data Sources (KDD Cup 2003) borrowed from Janez Brank & Jure Leskovec

The Dataset on KDD Cup 2003: 

The Dataset on KDD Cup 2003 Approx. 29000 papers from the “high energy physics – theory” area of arxiv.org For each paper: Full text (TeX file, often very messy) Avg. 60 KB per paper. Total: 1.7 GB. Metadata in a nice, structured file (authors, title, abstract, journal, subject classes) The citation graph Task: How many times have certain papers been downloaded in the first 60 days since publication in the arXiv?

Solution: 

Solution Textual documents have traditionally been treated as “bags of words” The number of occurrences of each word matters, but the order of the words is ignored Efficiently represented by sparse vectors We extend this to include other items besides words (“bag of X”) Most of our work was spent trying various features and adjusting their weight (more on that later) Use support vector regression to train a linear model, which is then used to predict the download counts on test papers Submitted solution was based on the model trained on the following representation: AA + 0.005 in-degree + 0.5 in-links + 0.7 out-links + 0.3 journal + 0.004 title-chars. + 0.6 (year – 2000) + 0.15 ClusDlAvg

A Look Back…: 

A Look Back…

References to some of the Books: 

References to some of the Books

References to Conferences: 

References to Conferences Information Retrieval: SIGIR, ECIR Machine Learning/Data Mining: ICML, ECML/PKDD, KDD, ICDM, SCDM Computational Linguistics: ACL, EACL, NAACL Semantic Web: ISWC, ESSW

References to some of the TM workshops (available online): 

References to some of the TM workshops (available online) ICML-1999 Workshop on Machine Learning in Text Data Analysis (TextML-1999) (http://www-ai.ijs.si/DunjaMladenic/ICML99/TLWsh99.html) at International Conference on Machine Learning, Bled 1999 KDD-2000 Workshop on Text Mining (TextKDD-2000) (http://www.cs.cmu.edu/~dunja/WshKDD2000.html) at ACM Conference on Knowledge Discovery on Databases, Boston 2000 ICDM-2001 Workshop on Text Mining (TextKDD-2001) (http://www-ai.ijs.si/DunjaMladenic/TextDM01/), at IEEE International Conference on Data Mining, San Jose 2001 ICML-2002 Workshop on Text Learning (TextML-2002) (http://www-ai.ijs.si/DunjaMladenic/TextML02/) at International Conference on Machine Learning, Sydney 2002 IJCAI-2003 Workshop on Text-Mining and Link-Analysis (Link-2003) (http://www.cs.cmu.edu/~dunja/TextLink2003/), at International Joint Conference on Artificial Intelligence, Acapulco 2003 KDD-2003 Workshop on Workshop on Link Analysis for Detecting Complex Behavior (LinkKDD2003) (http://www.cs.cmu.edu/~dunja/LinkKDD2003/) at ACM Conference on Knowledge Discovery on Databases, Washington DC 2003

Some of the Products: 

Some of the Products Authonomy ClearForest Megaputer SAS/Enterprise-Miner SPSS - Clementine Oracle - ConText IBM - Intelligent Miner for Text

Final Remarks: 

Final Remarks In the future we can expect stronger integration and bigger overlap between TM, IR, NLP and SW… …the technology and it’s solutions will try to capture deeper semantics within the text, …integration of various data sources (including text) is becoming increasingly important.