Lecture 7: Learning from Massive Datasets
Download as PPTX, PDF6 likes2,070 views
The lecture focuses on the challenges and methodologies of machine learning with massive datasets, emphasizing data mining and text mining concepts. It highlights the significance of domain knowledge, proper NLP techniques, and the importance of developing new algorithms to handle big data effectively. The presentation also discusses the pitfalls of current NLP tools and the necessity for more adaptable approaches to analyze unstructured text.
![NLP is pervasive
Ex: spell-checkers
Google Search
Google Mail
Facebook
Office Word
[…]
12
Lect. 7: Learning from Massive Datasets](https://image.slidesharecdn.com/lecture07ml4ltmarinasantini2013-131008062713-phpapp02/85/Lecture-7-Learning-from-Massive-Datasets-12-320.jpg)
![Actionable Textual Intelligence
Business Intelligence (BI) + Customer Analytics + Social
Network Analytics + Crisis Intelligence […] = Actionable
Intelligence
Actionable Intelligence is information that:
1.
2.
3.
4.
5.
6.
21
must be accurate and verifiable
must be timely
must be comprehensive
must be comprehensible
!!! give the power to make decisions and to act straightaway !!!
!!! must handle BIG BIG BIG UNSTRUCTURED TEXTUAL
DATA !!!
Lect. 7: Learning from Massive Datasets](https://image.slidesharecdn.com/lecture07ml4ltmarinasantini2013-131008062713-phpapp02/85/Lecture-7-Learning-from-Massive-Datasets-21-320.jpg)
![Big Data
BIG DATA [Wikipedia]:
Big data usually includes data sets with sizes beyond the ability of
commonly used software tools to capture, curate, manage, and
process the data within a tolerable elapsed time. Big data sizes
are a constantly moving target, as of 2012 ranging from a few
dozen terabytes to many petabytes of data in a single data
set. With this difficulty, new platforms of "big data" tools are being
developed to handle various aspects of large quantities of data.
Examples include Big Science, web logs, RFID, sensor networks,
social networks, social data (due to the social
data revolution), Internet text and documents,
Internet search indexing, call detail records,
astronomy, atmospheric science, genomics, biogeochemical,
biological, and other complex and often interdisciplinary scientific
research, military surveillance, medical records, photography
archives, video archives, and large-scale e-commerce.
22
Lect. 7: Learning from Massive Datasets](https://image.slidesharecdn.com/lecture07ml4ltmarinasantini2013-131008062713-phpapp02/85/Lecture-7-Learning-from-Massive-Datasets-22-320.jpg)
![Big Unstructured
TEXTUAL Data
Merrill Lynch is one of the world's leading
financial management and advisory companies,
providing financial advice.
―Merrill Lynch estimates that more than 85 percent of all
business information exists as unstructured data –
commonly appearing in e‐mails, memos, notes from call
centers and support operations, news, user groups,
chats, reports, letters, surveys, white papers,
marketing material, research, presentations and web
pages.‖ [DM Review Magazine, February 2003 Issue]
ECONOMIC LOSS!
23
Lect. 7: Learning from Massive Datasets](https://image.slidesharecdn.com/lecture07ml4ltmarinasantini2013-131008062713-phpapp02/85/Lecture-7-Learning-from-Massive-Datasets-23-320.jpg)
![Simple search is not enough…
Of course, it is possible to use simple search. But
simple search is unrewarding, because is based on
single terms.
24
”a search is made on the term felony. In a simple search,
the term felony is used, and everywhere there is a
reference to felony, a hit to an unstructured document is
made. But a simple search is crude. It does not find
references to crime, arson, murder, embezzlement,
vehicular homicide, and such, even though these crimes
are types of felonies” [ Source: Inmon, B. & A. Nesavich,
"Unstructured Textual Data in the Organization" from
"Managing Unstructured data in the organization",
Prentice Hall 2008, pp. 1–13]
Lect. 7: Learning from Massive Datasets](https://image.slidesharecdn.com/lecture07ml4ltmarinasantini2013-131008062713-phpapp02/85/Lecture-7-Learning-from-Massive-Datasets-24-320.jpg)
Lecture 7: Learning from Massive Datasets
- 1. October 2013 Machine Learning for Language Technology Lecture 7: Learning from Massive Datasets Marina Santini, Uppsala University Department of Linguistics and Philology
- 2. Outline Watch the pitfalls Learning from massive datasets Data Mining Text Mining – Text Analytics Web Mining Big Data Programming Languages and Framework for Big Data Big Textual Data & Commercial Applications Events, MeetUps, Coursera 2 Lect. 7: Learning from Massive Datasets
- 3. Practical Machine Learning 3 Lect. 7: Learning from Massive Datasets
- 4. Data Mining Data mining is the extraction of implicit, previously unknown and potentially useful information from data (Witten and Frank, 2005) 4 Lect. 7: Learning from Massive Datasets
- 5. Watch out! Machine Learning is not just about: Finding data and blindly applying learning algorithms to it Blindly compare machine learning methods: 1. 2. Model complexity Representativeness of training data distribution Reliability of class labels 1. 2. 3. Remember: Practitioners’ expertise counts! 5 Lect. 7: Learning from Massive Datasets
- 6. Massive Datasets Space and Time Three ways to make learning feasible (the old way) Small subset Parallelization Data chunks The new way: 6 Develop new algorithms with lower computational complexity Increase background knowledge Lect. 7: Learning from Massive Datasets
- 7. Domain Knowledge Metadata Semantic relation Causal relation Functional dependencies 7 Lect. 7: Learning from Massive Datasets
- 8. Text Mining Actionable information Comprehensible information Problems 8 Text Analytics Lect. 7: Learning from Massive Datasets
- 9. Definition: Text Analytics A set of NLP techniques that provide some structure to textual documents and help identify and extract important information. 9 Lect. 7: Learning from Massive Datasets
- 10. Set of NLP (Natural Language Processing ) techniques Common components of a text analytic package are: 10 Tokenization Morphological Analysis Syntactic Analysis Named Entity Recognition Sentiment Analysis Automatic Summarization Etc. Lect. 7: Learning from Massive Datasets
- 11. NLP at Coursera (www.coursera.org) 11 Lect. 7: Learning from Massive Datasets
- 12. NLP is pervasive Ex: spell-checkers Google Search Google Mail Facebook Office Word […] 12 Lect. 7: Learning from Massive Datasets
- 13. NLP is parvasive Ex: Name Entity Recognition Opinion mining Brand Trends Conversation clouds on web magazines and online newspapers… 13 Lect. 7: Learning from Massive Datasets
- 14. Sentiment Analysis 14 Lect. 7: Learning from Massive Datasets
- 15. Text Analytics Products and Frameworks Commercial Products: Attensity Clarabridge Temis Lexalytics Texify SAS SPSS IBM Cognos etc. 15 Open Source Frameworks: • • • • • GATE NLTK UIMA openNLP etc. Lect. 7: Learning from Massive Datasets
- 16. However… (I) NLP tools and applications (both commercial and open source) are not perfect. Research is still very active in all NLP fields. 16 Lect. 7: Learning from Massive Datasets
- 17. Ex: Syntactic Parser Connexor What about parsing a tweet? “My son, Ky/o, asked me for the first time today how my DAY was . . . I about melted. Told him that I had pizza for lunch. Response? No fair “ (Twitter Tutorial 1: How to Tweet Well) 17
- 18. Why NLP and Text Analytics for Text Mining? Why is it important to know that a word is a noun, or a verb or the name of brand? Broadly speaking (Think about these as features for a classification problem!) 18 Nouns and verbs (a.k.a. content words): Nouns are important for topic detection; verbs are important if you want to identify actions or intentions. Adjectives = sentiment identification. Function words (a.k.a. stop words) are important for authorship attribution, plagiarism detection, etc. etc. Lect. 7: Learning from Massive Datasets
- 19. However… (II) At present, the main pitfall of many NLP applications is that they are not flexible enough to: Completly disambiguate language Identify how language is used in different types of documents (a.k.a. genres). For instance, in tweets langauge is used in a different way than an emails, language used in email is different from the language used in academic papers, etc. ) Often tweaking NLP tools to different types of text or solve language ambiguity in an ad-hoc manner is time-consuming, difficult and unrewarding… 19 Lect. 7: Learning from Massive Datasets
- 20. What for? Text summarization Document clustering Authorship attribution Automatic medadata extraction Entity extraction Information extraction Information discovery ACTIONABLE INTELLIGENCE 20 Lect. 7: Learning from Massive Datasets
- 21. Actionable Textual Intelligence Business Intelligence (BI) + Customer Analytics + Social Network Analytics + Crisis Intelligence […] = Actionable Intelligence Actionable Intelligence is information that: 1. 2. 3. 4. 5. 6. 21 must be accurate and verifiable must be timely must be comprehensive must be comprehensible !!! give the power to make decisions and to act straightaway !!! !!! must handle BIG BIG BIG UNSTRUCTURED TEXTUAL DATA !!! Lect. 7: Learning from Massive Datasets
- 22. Big Data BIG DATA [Wikipedia]: Big data usually includes data sets with sizes beyond the ability of commonly used software tools to capture, curate, manage, and process the data within a tolerable elapsed time. Big data sizes are a constantly moving target, as of 2012 ranging from a few dozen terabytes to many petabytes of data in a single data set. With this difficulty, new platforms of "big data" tools are being developed to handle various aspects of large quantities of data. Examples include Big Science, web logs, RFID, sensor networks, social networks, social data (due to the social data revolution), Internet text and documents, Internet search indexing, call detail records, astronomy, atmospheric science, genomics, biogeochemical, biological, and other complex and often interdisciplinary scientific research, military surveillance, medical records, photography archives, video archives, and large-scale e-commerce. 22 Lect. 7: Learning from Massive Datasets
- 23. Big Unstructured TEXTUAL Data Merrill Lynch is one of the world's leading financial management and advisory companies, providing financial advice. ―Merrill Lynch estimates that more than 85 percent of all business information exists as unstructured data – commonly appearing in e‐mails, memos, notes from call centers and support operations, news, user groups, chats, reports, letters, surveys, white papers, marketing material, research, presentations and web pages.‖ [DM Review Magazine, February 2003 Issue] ECONOMIC LOSS! 23 Lect. 7: Learning from Massive Datasets
- 24. Simple search is not enough… Of course, it is possible to use simple search. But simple search is unrewarding, because is based on single terms. 24 ”a search is made on the term felony. In a simple search, the term felony is used, and everywhere there is a reference to felony, a hit to an unstructured document is made. But a simple search is crude. It does not find references to crime, arson, murder, embezzlement, vehicular homicide, and such, even though these crimes are types of felonies” [ Source: Inmon, B. & A. Nesavich, "Unstructured Textual Data in the Organization" from "Managing Unstructured data in the organization", Prentice Hall 2008, pp. 1–13] Lect. 7: Learning from Massive Datasets
- 25. Programming languages and frameworks for big data 25 Lect. 7: Learning from Massive Datasets
- 26. http://www.r-project.org/ R R is a statistical programming language. It is a free software programming language and a software environment for statistical computing and graphics. The R language is widely used among statisticians and data miners for developing statistical software and data analysis. Polls and surveys of data miners are showing R's popularity has increased substantially in recent years (wikipedia) 26
- 27. 27 Lect. 7: Learning from Massive Datasets
- 28. MeetUps: R in Stockholm 28 Lect. 7: Learning from Massive Datasets
- 29. Can R help out? Can R help overcome NLP shortcomings and open a new direction in order to extract useful information from Big TEXTUAL Data? 29 Lect. 7: Learning from Massive Datasets
- 30. Existing literature for linguists Stefan Th. Gries (2013) Statistics for linguistics With R: A Practical Introduction. De Gruyter Mouton. New Edition. Stefan Th. Gries (2009) Quantitative corpus linguistics with R: a practical introduction. Routledge, Taylor & Francis Group (companion website). Harald R. Baayen (2008) Analyzing Linguistic Data: A Practical Introduction to Statistics using R. Cambridge. …. 30 Lect. 7: Learning from Massive Datasets
- 31. Companion website by Stefan Th. Gries BNC=British National Corpus (PoS tagged) 31 Lect. 7: Learning from Massive Datasets
- 32. BNC The British National Corpus (BNC) is a 100 million word collection of samples of written and spoken language from a wide range of sources, designed to represent a wide cross-section of British English from the later part of the 20th century, both spoken and written. The latest edition is the BNC XML Edition, released in 2007. The corpus is encoded according to the Guidelines of the Text Encoding Initiative (TEI) to represent both the output from CLAWS (automatic part-of-speech tagger) and a variety of other structural properties of texts (e.g. headings, paragraphs, lists etc.). Full classification, contextual and bibliographic information is also included with each text in the form of a TEI-conformant header. 32 Lect. 7: Learning from Massive Datasets
- 33. R & the BNC: Excerpt from Google Books 33 Lect. 7: Learning from Massive Datasets
- 34. What about Big Textual Data? Non standardized language Non standard texts Electronic documents of all kinds, eg. formal, informal, short, long, private, public, etc. 34 Lect. 7: Learning from Massive Datasets
- 35. Not distributed system Open Source The name Scala is a portmanteau of "scalable" and "language", signifying that it is designed to grow with the demands of its users. James Strachan, the creator of Groovy, described Scala as a possible successor to Java Commercial 35 R Scala (also distributed systems) Rapid Miner Weka … SPSS SAS MatLab … Lect. 7: Learning from Massive Datasets
- 36. From The Economist: The Big Data scenario 36 Lect. 7: Learning from Massive Datasets
- 37. Commercial applications for Big Textual Data Recorded Future web intelligence (anticipating emerging threats, future trends, anticipating competitors’ actions, etc.) Gavagai large-scale textual analysis (prediction and future trends) 37 Lect. 7: Learning from Massive Datasets
- 38. Thanks to Staffan Truffe’ for the ff slides 38 Lect. 7: Learning from Massive Datasets
- 39. Size 39 Lect. 7: Learning from Massive Datasets
- 40. In a few pictures… 40 Lect. 7: Learning from Massive Datasets
- 41. Metrics, structure and time 41 Lect. 7: Learning from Massive Datasets
- 42. Metric 42 Lect. 7: Learning from Massive Datasets
- 43. Structure 43 Lect. 7: Learning from Massive Datasets
- 44. Time 44 Lect. 7: Learning from Massive Datasets
- 45. Facts 45 Lect. 7: Learning from Massive Datasets
- 46. Pipeline 46 Lect. 7: Learning from Massive Datasets
- 47. Multi-Language 47 Lect. 7: Learning from Massive Datasets
- 48. Text Analytics 48 Lect. 7: Learning from Massive Datasets
- 49. Predictions 49 Lect. 7: Learning from Massive Datasets
- 50. Gavagai Jussi Karlgren (PhD in Stylistics in Information Retrieval) Magnus Sahlgren (PhD thesis in distributional semantics) Fredrick Olsson (PhD thesis in Active Learning) (co-workers at SICS) The indeterminacy of translation is a thesis propounded by 20thcentury American analytic philosopher W. V. Quine. Quine uses the example of the word "gavagai" uttered by a native speaker of the unknown language Arunta upon seeing a rabbit. A speaker of English could do what seems natural and translate this as "Lo, a rabbit." But other translations would be compatible with all the evidence he has: "Lo, food"; "Let's go hunting"; "There will be a storm tonight" (these natives may be superstitious)… (wikipedia) 50 Lect. 7: Learning from Massive Datasets
- 51. Ethersource presented Thanks to F. Olsson for the ff slides 51 Lect. 7: Learning from Massive Datasets
- 52. Associations 52 Lect. 7: Learning from Massive Datasets
- 53. Language is flux 53 Lect. 7: Learning from Massive Datasets
- 54. Learning from use 54 Lect. 7: Learning from Massive Datasets
- 55. Scope 55 Lect. 7: Learning from Massive Datasets
- 56. Architecture 56 Lect. 7: Learning from Massive Datasets
- 57. Web vs printed world 57 Lect. 7: Learning from Massive Datasets
- 58. Noise… 58
- 59. Multi-linguality 59 Lect. 7: Learning from Massive Datasets
- 60. SICS 60 Watch the videos! Lect. 7: Learning from Massive Datasets
- 61. Big Data MeetUp, Stockholm 61 Lect. 7: Learning from Massive Datasets
- 62. BIG DATA communities 62 Lect. 7: Learning from Massive Datasets
- 63. Future Directions in Machine Learning for Language Technology Deluge of data Little linguistic analysis in the realm of big-data realworld platforms and applications Top-down systems cannot efficiently deal with irregularity and unpredictability of big textual data Data-driven systems can make it. However, 63 …we know that computers are not at ease with natural languages used by humans, unless they learn how to learn linguistic structure underlying natual language from data… Lect. 7: Learning from Massive Datasets
- 64. For a data-driven approach… Annotated datasets that are needed for complete supervised machine learning are costly, timecomsuming and require specialist expertise. Is complete supervision even thinkable when we talk about tera-, peta- or yottabytes? How big should then be the training set? Alternative solutions: 64 Semi-supervised methods (combination of labelled and unlabelled data) Weakly supervised methods (human-constructed rules are typically used to guide the unsupervised learner) Unsupervised learning results cannot still compete with suprevised learning in many tasks… Lect. 7: Learning from Massive Datasets
- 65. A new way to explore: Incomplete Supervision Relies on partially labelled data: 65 ‖ Human experts — or possibly a crowd of laymen — annotate text with some linguistic structure related to the structure that one wants to predict. This data is then used for partially supervised learning with a statistical model that exploits the annotated structure to infer the linguistic structure of interest.‖ p. 4 Lect. 7: Learning from Massive Datasets
- 66. Example ”…it is possible to construct accurate and robust part-of-speech taggers for a wide range of languages, by combining (1) manually annotated resources in English, or some other language for which such resources are already available, with (2) a crowd-sourced target-language specific lexicon, which lists the potential parts of speech that each word may take in some context, at least for a subset of the words. Both (1) and (2) only provide partial information for the part-ofspeech tagging task. However, taken together they turn out to provide substantially more information than either taken alone. “ p. 46 Oscar Täckström “Predicting Linguistic Structure with Incomplete and Cross-Lingual Supervision” PhD Thesis, Uppsala University, 2013 (http://soda.swedish-ict.se/5513/) 66 Lect. 7: Learning from Massive Datasets
- 67. Conclusions This course is an introduction to Machine leaning for Language Technology”. You get a flavour of the problems we come across when devising models for enabling machines to analyse and make sense of natural human language. The next big big big step is to bring as much linguistic awareness as possible into big data. 67 Lect. 7: Learning from Massive Datasets
- 68. Reading Witten and Frank (2005) Ch. 8 68 Lect. 7: Learning from Massive Datasets
- 69. Thanx for your attention! 69 Lect. 7: Learning from Massive Datasets
Editor's Notes
- #10: Weneedtools toanalyse this huge amont of textual data and extract the information weneed.
- #13: Orthographic check: is somethingwrittencorrectly or not? Vital for searching
- #14: What is a namedentity?names of persons, organizations, locations, expressions of times, quantities, monetary values, percentages,
- #15: If you try with longer texts or with another genre, results are not reliable
- #16: Business intelligence (BI) is the ability of an organization to collect, maintain, and organize data. This produces large amounts of information that can help develop new opportunities. Identifying these opportunities, and implementing an effective strategy, can provide a competitive market advantage and long-term stability. BI technologies provide historical, current and predictive views of business operations.Customer Experience Management (CEM) is the practice of actively listening to the Voice of the Customer through a variety of listening posts, analyzing customer feedback to create a basis for acting on better business decisions and then measuring the impact of those decisions to drive even greater operational performance and customer loyalty. Through this process, a company strategically organizes itself to manage a customer's entire experience with its product, service or company. Companies invest in CEM to improve customer retention
- #18: A tweet: My son, 6y/o, asked me for the first time today how my DAY was . . . I about melted. Told him that I had pizza for lunch. Response? No fairLanguage is highty ambiguous. Fair =reasonable and acceptable//treatingeveryoneequallyFair=a form of outdoor entertainment, at which there are large machines to ride on and games in which you can win prizes//an event at which people or businesses show and sell their productsplay fair: to do something in a fair and honest way
- #22: Informationdiscovery is toovague
- #23: Problem of size + a problem of diverse data! = heterogeneos dataRadio-frequencyidentification (RFID )
- #24: Mucheffort hasbeenallocate to improvebig native data numeric data: balancesheets, income reports, financial and business reports, etc.Merrill Lynch – financial management and advisorywww.ml.com/Merrill Lynch is one of the world's leading financial management and advisory companies, providing financial advice and investment banking services.e‐mails, memos, notes from call centers and support operations, news, user groups, chats, reports, letters, surveys, white papers, marketing material, research, presentations , etc are different genres, ie different types of text. For example, emails and white papers are both textual genres but they differ a lot from each other. They might deal with the same topic, but in a complete different way. So the type of information related to the same topic can vary according to genre.
- #25: felony= any grave crimes, such as murder, rape, or burglary…
- #31: Professor of Linguistics, Department of Linguistics, University of California, Santa Barbara
- #32: N-gramsAveragesentence and wordlengthIndexingSplit infinitives
- #33: Stockholm –umeÅcorpus (joakim)
- #34: DescriptivestatisticsAnalyticalstatisticsMultifactorialmethodsToken/typeratio=The type-token ratio (TTR) is a measure of vocabulary variation within a written text or a person’s speech. The type-token ratios of two real world examples are calculated and interpreted. The type-token ratio is shown to be a helpful measure of lexical variety within a text. It can be used to monitor changes in children and adults with vocabulary difficulties.Tokens are the number of words. several of these tokens are repeated. For example, the token again occurs two times, the token are occurs three times, and the token and occurs five times. the total of 87 tokens in this text there are 62 so-called types. The relationship between the number of types and the number of tokens is known as the type-token ratio (TTR). For Text 1 above we can now calculate this as follows:Type-Token Ratio = (number of types/number of tokens) * 100= (62/87) * 100 = 71.3%The more types there are in comparison to the number of tokens, then the more varied is the vocabulary, i.e. it there is greater lexical variety.http://www.speech-therapy-information-and-resources.com/type-token-ratio.html
- #37: http://youtu.be/qqfeUUjAIyQ