DATA SCIENCE Lesson 4 Data Science Predictive Method Parsing Process Topic Modeling conference Jean-Antoine Moreau
2 likes1,776 views
The document discusses predictive methods and analytics, focusing on data processing and modeling techniques to understand and predict behaviors. Key areas include natural language processing, data science applications in business, and the use of algorithms for predictive analysis across various fields. It emphasizes the importance of predictive models in understanding financial, social, and economic trends.
DATA SCIENCE Lesson 4 Data Science Predictive Method Parsing Process Topic Modeling conference Jean-Antoine Moreau
- 1. PPrreeddiiccttiivvee MMeetthhoodd –– AAllggoorriitthhmm MMeetthhooddoollooggyy –– PPrroocceessss DDaattaa PPrroocceessssiinngg DDaattaa MMiinniinngg GGooaallss Conference - Study : Jean-Antoine Moreau (Engineer - Lecturer) © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP.
- 2. CCoonntteenntt AAnnaallyyssiiss NNaattuurraall LLaanngguuaaggee PPrroocceessssiinngg TTooppiicc MMooddeelliinngg PPaarrssiinngg PPrroocceessss DDaattaa PPrroodduucctt Conference - Study : Jean-Antoine Moreau (Engineer - Lecturer) © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP.
- 3. NNaasshh eeqquuiilliibbrriiuumm uussee ttoo tthhee ssppeecciiffiiccaattiioonn SScchheelllliinngg MMooddeell MMaacchhiinnee lleeaarrnniinngg Conference - Study : Jean-Antoine Moreau (Engineer - Lecturer) © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP.
- 4. DDaattaa sscciieennccee UUssiinngg iinn BBuussiinneessss uussee pprroocceessss iinn tthhee ccoommppaannyy Conference - Study : Jean-Antoine Moreau (Engineer - Lecturer) © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP.
- 5. DDAATTAA SSCCIIEENNCCEE LLeeccttuurree 44 Conference - Study : Jean-Antoine Moreau (Engineer - Lecturer) © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP.
- 6. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 “Nothing in life is to be feared, it is only to be understood. Now is the time to understand more, so that we may fear less.” JAM 6 / 218 Contact http://www.jean-antoine-moreau.fr.nf Marie Curie
- 7. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 7 / 218 PPrreeddiiccttiivvee aannaallyyttiiccss Contact http://www.jean-antoine-moreau.fr.nf
- 8. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Measure the probability that event occurs JAM 8 / 218 Predictive analysis Retrieve a large amount of data To extract statistical purposes Contact http://www.jean-antoine-moreau.fr.nf
- 9. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 9 / 218 Predictive Analysis Process Contact http://www.jean-antoine-moreau.fr.nf
- 10. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 10 / Predictive Analysis Decision Tree Contact http://www.jean-antoine-moreau.fr.nf
- 11. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Using the Graph Theory to define a structure JAM 11 / Predictive Analysis Contact http://www.jean-antoine-moreau.fr.nf
- 12. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 12 / predict consumer behavior predict investor behavior predict the behavior of voters Contact http://www.jean-antoine-moreau.fr.nf
- 13. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 13 / adaptation of means the results of predictive analytics Contact http://www.jean-antoine-moreau.fr.nf
- 14. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 14 / • track financial trends; • track social trends; • track economic trends. Contact http://www.jean-antoine-moreau.fr.nf
- 15. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 individual intelligence collective intelligence JAM 15 / algorithms modeling Contact http://www.jean-antoine-moreau.fr.nf
- 16. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 16 / New algorithms are Method Contact http://www.jean-antoine-moreau.fr.nf
- 17. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 17 / predictive roles Contact http://www.jean-antoine-moreau.fr.nf
- 18. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 18 / Data • Social data; • Posts; • Comments; • Behavioral data; • Geographic data; • Environmental data. Contact http://www.jean-antoine-moreau.fr.nf
- 19. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 PredictiveDB is a predictive database, an online RDBMS to learn, classify or predict from SQL data without knowledge in Machine Learning or Natural Language/ JAM 19 / Exemple • http://blog.predictivedb.com/ Contact http://www.jean-antoine-moreau.fr.nf
- 20. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 constructing a predictive model JAM 20 / data mining Input | Output > Association Contact http://www.jean-antoine-moreau.fr.nf
- 21. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 environmental description Property that we want to predict or determine. JAM 21 / Association Input Ouput Contact http://www.jean-antoine-moreau.fr.nf
- 22. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 22 / Semantic Approach Semantic Analysis Machine Learning Contact http://www.jean-antoine-moreau.fr.nf
- 23. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • uses aggregated Google search data to estimate flu activity. JAM 23 / example Google Flu Trends : • Explore flu trends around the world http://www.google.org/flutrends/ Contact http://www.jean-antoine-moreau.fr.nf
- 24. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Uncertainty Management Strategy JAM 24 / Predictive Device Contact http://www.jean-antoine-moreau.fr.nf
- 25. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 25 / proactive assessment of the behaviors. Contact http://www.jean-antoine-moreau.fr.nf
- 26. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 participatory assessment of behavior statistical body of the individual (behaviors that could happen, JAM 26 / physical body of the individual Contact http://www.jean-antoine-moreau.fr.nf possible future)
- 27. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 27 / Behavioral detection facial expression motion recognition Patterns behavior patterns Contact http://www.jean-antoine-moreau.fr.nf
- 28. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 28 / Structures of behavior Algorithms analysis of mass data use of profiling framework Contact http://www.jean-antoine-moreau.fr.nf
- 29. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 29 / behavior using the graph theory initial behavior Contact http://www.jean-antoine-moreau.fr.nf
- 30. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Using a Finite State Machine and a Behavior Tree JAM 30 / Contact http://www.jean-antoine-moreau.fr.nf
- 31. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 31 / definition of profiling framework characteristics profiling database predictive behavior psychological profile predictive choices Contact http://www.jean-antoine-moreau.fr.nf
- 32. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 use the asymmetry of the information JAM 32 / Contact http://www.jean-antoine-moreau.fr.nf
- 33. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 33 / topological study allocation study Contact http://www.jean-antoine-moreau.fr.nf
- 34. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 34 / Contact http://www.jean-antoine-moreau.fr.nf Spaces of studies
- 35. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 35 / BBiigg DDaattaa DDaattaa SScciieennccee Creation of data product DDaattaa PPrroodduucctt Contact http://www.jean-antoine-moreau.fr.nf
- 36. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 36 / Model behavior error correction Feedback model Contact http://www.jean-antoine-moreau.fr.nf
- 37. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 37 / predictive analysis predictive logic Contact http://www.jean-antoine-moreau.fr.nf
- 38. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 mathematical models of conflict and cooperation between intelligent rational decision-makers JAM 38 / GGaammee TThheeoorryy (John von Neumann, Oskar Morgenstern) study of strategic decision making Contact http://www.jean-antoine-moreau.fr.nf
- 39. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 predictive assumptions about future events JAM 39 / Game Theory (John von Neumann, Oskar Morgenstern) analysis of present and past facts Contact http://www.jean-antoine-moreau.fr.nf
- 40. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 40 / Game Theory; (John von Neumann, Oskar Morgenstern); NNaasshh eeqquuiilliibbrriiuumm:: Simply reveal each person's strategy to all players. The Nash equilibrium existes if no players change their strategy, ... Contact http://www.jean-antoine-moreau.fr.nf
- 41. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 41 / Contact http://www.jean-antoine-moreau.fr.nf Data Pr operti es behavior predictive models
- 42. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 42 / improvement Contact http://www.jean-antoine-moreau.fr.nf Data Pr operti es behavior predictive models
- 43. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 43 / predictive mode opportunity risk method algorithm Contact http://www.jean-antoine-moreau.fr.nf
- 44. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 44 / Contact http://www.jean-antoine-moreau.fr.nf Data model method • Opportunity • Risk
- 45. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 45 / method Contact http://www.jean-antoine-moreau.fr.nf logical analysis relationship behavior issue condition factor
- 46. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 to estimate the probability of a behavior stimulus scenario JAM 46 / Analyze past performance simulate human behavior Contact http://www.jean-antoine-moreau.fr.nf
- 47. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. a little break, to ask questions. “ Whosoever desires constant success must change his conduct with the times.” Niccolò Machiavelli JAM 47 / Contact http://www.jean-antoine-moreau.fr.nf
- 48. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Descriptive models describe the relationships between data. JAM 48 / DDeessccrriippttiivvee mmooddeell Contact http://www.jean-antoine-moreau.fr.nf
- 49. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Descriptive models describe the relationships between JAM 49 / DDeessccrriippttiivvee mmooddeell data and information Contact http://www.jean-antoine-moreau.fr.nf
- 50. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Descriptive modeling is a mathematical process that describes real-world events and the relationships between factors responsible for them. JAM 50 / DDeessccrriippttiivvee mmooddeell Contact http://www.jean-antoine-moreau.fr.nf
- 51. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Descriptive models do not order the objects. JAM 51 / DDeessccrriippttiivvee mmooddeell Contact http://www.jean-antoine-moreau.fr.nf
- 52. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 52 / DDeessccrriippttiivvee mmooddeell mathematical modeling, the process Contact http://www.jean-antoine-moreau.fr.nf
- 53. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 53 / DDeessccrriippttiivvee mmooddeell Software / method > modeling Contact http://www.jean-antoine-moreau.fr.nf
- 54. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 54 / DDeessccrriippttiivvee mmooddeell for DDaattaa MMiinniinngg Contact http://www.jean-antoine-moreau.fr.nf
- 55. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Process of preparing data for the bbuussiinneessss mmooddeelliinngg JAM 55 / DDeessccrriippttiivvee mmooddeell Contact http://www.jean-antoine-moreau.fr.nf
- 56. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 the relationship between elements of a decision JAM 56 / decision models describe Contact http://www.jean-antoine-moreau.fr.nf
- 57. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 development of a logical / decision method data relation result / outcome JAM 57 / decision model method Contact http://www.jean-antoine-moreau.fr.nf
- 58. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 58 / decision models examples • Medical: help system for clinical decision; • Financial: recovery system; • Risk management system; • Transportation: driving assistance system. Contact http://www.jean-antoine-moreau.fr.nf
- 59. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 59 / Statistical techniques • regression model: • linear regression; • autoregression; • logistic regression; • multinomial logistic regression. Contact http://www.jean-antoine-moreau.fr.nf
- 60. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 60 / Graphical decision model Contact http://www.jean-antoine-moreau.fr.nf
- 61. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 61 / Graphical decision model Objective / Profit Contact http://www.jean-antoine-moreau.fr.nf
- 62. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 62 / Business Graphical decision model Contact http://www.jean-antoine-moreau.fr.nf
- 63. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • In statistics, a probit model is a type of regression where the dependent variable can only take two values, – for example married or not married . JAM 63 / probit model Contact http://www.jean-antoine-moreau.fr.nf
- 64. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 creation of mathematical equation models to represent the interactions between the different variables considered. JAM 64 / regression model Contact http://www.jean-antoine-moreau.fr.nf
- 65. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 to study, define, test, the representation of interactions between the variables considered. JAM 65 / regression model method Contact http://www.jean-antoine-moreau.fr.nf
- 66. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 66 / dependent variable response variables relationship Contact http://www.jean-antoine-moreau.fr.nf set of independent variables set of predictors Equation (adjusted and optimized)
- 67. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 The Gauss-Markov hypothesis must be verified. JAM 67 / Contact http://www.jean-antoine-moreau.fr.nf
- 68. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • Total power of prediction of the dependent variable explained by the variation of the independent variables. JAM 68 / Test the model Contact http://www.jean-antoine-moreau.fr.nf
- 69. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 autoregressive integrated moving average (ARIMA) model JAM 69 / Autoregressive model autoregressive moving average (ARMA) model. Contact http://www.jean-antoine-moreau.fr.nf
- 70. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 70 / DDiissccrreettee cchhooiiccee mmooddeellss have become an essential tool in modeling individual behavior Contact http://www.jean-antoine-moreau.fr.nf
- 71. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 71 / The different kinds of regression • Linear regression; • Multiple Linear Regression; • Polynomial regression; • Logistic regression; • Generalized linear model; • Nonparametric regression; • Multiple regression models postulated and not postulated; • Circular regression; • Elliptical regression; • Local regression. Contact http://www.jean-antoine-moreau.fr.nf
- 72. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Method transforming information on the binary dependent variable in an estimated unlimited continuous variable using a multivariate normal model analysis. JAM 72 / Logistic regression Logistic model Contact http://www.jean-antoine-moreau.fr.nf
- 73. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Whenever a relationship within or between data items can be expressed as a statistical model with parameters to be estimated from a sample, the Wald test can be used to test the true value of the parameter based on the sample estimate. JAM 73 / TThhee WWaalldd TTeesstt is a parametric statistical test (statistician Abraham Wald) Contact http://www.jean-antoine-moreau.fr.nf
- 74. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 74 / TThhee WWaalldd TTeesstt LLiikkeelliihhoooodd rraattiioo tteesstt MMeetthhoodd ooff lleeaasstt ssqquuaarreess ssttaattiissttiiccaall eevvaalluuaattiioonn Contact http://www.jean-antoine-moreau.fr.nf
- 75. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • In the probit model, the inverse standard normal distribution of the probability is modeled as a linear combination of the predictors. JAM 75 / Probit regression • also called a probit model, is used to model dichotomous or binary outcome variables. Contact http://www.jean-antoine-moreau.fr.nf
- 76. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 – R (R software - RapidMiner is a powerful visual tool for data mining, machine learning, and predictive analytics); – Weka (Waikato Environment for Knowledge Analysis) JAM 76 / PPrreeddiiccttiivvee aannaallyyttiiccss ttoooollss Commercial software: – SAS; – SPSS (IBM); – STATISTICA (statoff); Free software: • machine learning algorithms for data mining tasks. Contact http://www.jean-antoine-moreau.fr.nf
- 77. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. “A thinker sees his own actions as experiments and questions - as attempts to find out something. Success and failure are for him answers above all.” JAM 77 / a little break, to ask questions. Contact http://www.jean-antoine-moreau.fr.nf Friedrich Nietzsche
- 78. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 78 / PPaarrssiinngg Contact http://www.jean-antoine-moreau.fr.nf
- 79. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 tthhee pprroocceessss ooff aannaallyyzziinngg aa ssttrriinngg ooff ssyymmbboollss JAM 79 / PPaarrssiinngg ssyynnttaaccttiicc aannaallyyssiiss Contact http://www.jean-antoine-moreau.fr.nf
- 80. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 80 / ttaaxxoonnoommyy aallggoorriitthhmm tthhee ttaaxxoonnoommyy lleeaarrnniinngg aallggoorriitthhmm Contact http://www.jean-antoine-moreau.fr.nf
- 81. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 81 / taxonomy algorithm • Resolves issue: – Recursive parser; – Predictive parser; – Predictive recursive parser. Contact http://www.jean-antoine-moreau.fr.nf
- 82. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 82 / pprreeddiiccttiivvee ppaarrssiinngg Contact http://www.jean-antoine-moreau.fr.nf
- 83. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 83 / predictive parsing The goal of predictive parsing is to construct a top-down parser Contact http://www.jean-antoine-moreau.fr.nf
- 84. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 84 / • Parse Table entries; • Parse Table; • Predictive parsing; • Assignment. Contact http://www.jean-antoine-moreau.fr.nf
- 85. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 85 / PPrreeddiiccttiivvee ppaarrssiinngg aallggoorriitthhmm Contact http://www.jean-antoine-moreau.fr.nf • Use : – The parse table; • An input buffer containing a sequence of tokens; • A stack of grammar symbols.
- 86. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 86 / PPrreeddiiccttiivvee ppaarrssiinngg • Example : Parse the string (id+num)*id Contact http://www.jean-antoine-moreau.fr.nf
- 87. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 87 / Predictive parsing Process and use Contact http://www.jean-antoine-moreau.fr.nf
- 88. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • Prédictive parcing, the front-end (exemple – compiler): – Parsing is the core of the front end of a compiler. – The predictive recursive-descent parsing technology is most widely used in a traditional compiler design. JAM 88 / Contact http://www.jean-antoine-moreau.fr.nf
- 89. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Front end Source code IR Machine code JAM 89 / Front end Back end Contact http://www.jean-antoine-moreau.fr.nf error management
- 90. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Source code tokens IR Source code Scanner lexical analysis, preprocessing, parsing, semantic analysis (Syntax-directed translation), code generation, and code optimization. JAM 90 / The Front-End-Scanner Contact http://www.jean-antoine-moreau.fr.nf Errors management
- 91. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 91 / UML modelisation an example Contact http://www.jean-antoine-moreau.fr.nf
- 92. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 92 / The Front-End-Scanner Source code tokens IR scanner Parser Contact http://www.jean-antoine-moreau.fr.nf Error management
- 93. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 – Checks the stream of words and their parts of speech (product by the scanner) for grammatical correctors; – Determines if the input is syndically well formed; – Guides checking at deeper levels than syntax; – Build an IR representative of the code. JAM 93 / • Parser (other example) : • ((IIntermediate / IIndependent RRepresentation) Contact http://www.jean-antoine-moreau.fr.nf
- 94. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • The parse table; • An Input buffer containing a sequence of tokens; • A stack of grammar symbols. JAM 94 / Predictive Parsing Algorithm Contact http://www.jean-antoine-moreau.fr.nf
- 95. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 95 / • Roadmap (where we are?) – Specifying tokens; – Specifying syntax; • Predictive top-down parsing : – First, Follow … – The condition; – Table-driven parsers. Contact http://www.jean-antoine-moreau.fr.nf
- 96. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. “Do what you can, with what you have, where you are.” Theodore Roosevelt JAM 96 / a little break, to ask questions. Contact http://www.jean-antoine-moreau.fr.nf
- 97. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 PPaarrssiinngg TTeecchhnniiqquueess JAM 97 / Contact http://www.jean-antoine-moreau.fr.nf
- 98. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 98 / PPaarrssiinngg TTrreeee first example Contact http://www.jean-antoine-moreau.fr.nf
- 99. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 99 / PPaarrssiinngg TTrreeee second example Contact http://www.jean-antoine-moreau.fr.nf
- 100. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 100 / PPaarrssiinngg TTrreeee Using the Graph Theory Contact http://www.jean-antoine-moreau.fr.nf
- 101. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 – Start at the root of the parse tree and grow toward leaves; – Pick a production & try to mach the input; JAM 101 / • Top Down parser: Contact http://www.jean-antoine-moreau.fr.nf
- 102. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • Bottom-up parsers : • Start of the leaves and grow toward root; • As input is consumed, encode, possibilities in a internal state; • Start in a state valid for legal first tokens; • Bottom-up parsers handle a large class of grammars. JAM 102 / – Operator procedure Contact http://www.jean-antoine-moreau.fr.nf
- 103. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 103 / • Top-down parsers: – Recursive descend. Contact http://www.jean-antoine-moreau.fr.nf
- 104. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 SSiimmppllee mmeetthhoodd Construct the root of the parse tree; Repeat until the fringe of the parse tree matches the input string; Select the product of each symbol and construct the child; Find each node to be expanded. JAM 104 / • Top-down parsing algorithm : Contact http://www.jean-antoine-moreau.fr.nf
- 105. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 105 / EExxaammpplleess ooff uussiinngg Contact http://www.jean-antoine-moreau.fr.nf
- 106. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 106 / Parallel Parsing Process Contact http://www.jean-antoine-moreau.fr.nf
- 107. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 PPaarrssiinngg AAnnaallyyssiiss PPrroocceessss Exemple de modélisation: le plan du gouvernement pour augmenter l’impôt sur le revenu. JAM 107 / Example of modelisation : the government's plan to increase the income tax. Contact http://www.jean-antoine-moreau.fr.nf
- 108. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 108 / PPaarrssiinngg TTrreeee ppaatttteerrnn rreeccooggnniittiioonn in this case the recognition of the shape of the furniture component Contact http://www.jean-antoine-moreau.fr.nf
- 109. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 109 / EExxaammppllee ooff TTrreeee TThhee HHTTMMLL ppaarrssee ssttrruuccttuurree Contact http://www.jean-antoine-moreau.fr.nf
- 110. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 110 / PPaarrssiinngg AAllggoorriitthhmm Contact http://www.jean-antoine-moreau.fr.nf
- 111. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 111 / PPaarrssiinngg aannaallyyssiiss tthhee ddeeccoommppoossiittiioonn ooff tthhee pprroocceessss Contact http://www.jean-antoine-moreau.fr.nf
- 112. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 112 / pphhrraassee ppaarrssiinngg PPrroocceessss // AAllggoorriitthhmm Contact http://www.jean-antoine-moreau.fr.nf
- 113. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 113 / PPhhrraassee PPaarrssiinngg EExxpprreessssiioonn SSttrreeaammiinngg AAllggoorriitthhmm Contact http://www.jean-antoine-moreau.fr.nf
- 114. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 HHooww ttoo ddeeffiinnee tthhee ssppaannnniinngg ttrreeee ppaarrssiinngg mmiinniimmuumm ?? JAM 114 / UUssiinngg tthhee GGrraapphhss TThheeoorryy Contact http://www.jean-antoine-moreau.fr.nf
- 115. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 PPaarrssiinngg ttoo ffiinndd tthhee ssttrruuccttuurree ooff aa sseenntteennccee JAM 115 / Contact http://www.jean-antoine-moreau.fr.nf
- 116. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 AAuuddiioo :: FFoouurriieerr uussee ttoo ddeeffiinnee aa ppaarrssiinngg ttrreeee.. JAM 116 / Contact http://www.jean-antoine-moreau.fr.nf
- 117. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 117 / PPaarrssiinngg ccllaassssiiffiiccaattiioonn Contact http://www.jean-antoine-moreau.fr.nf
- 118. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 118 / PPaarrssiinngg ccllaassssiiffiiccaattiioonn Contact http://www.jean-antoine-moreau.fr.nf
- 119. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. « If I am walking with two other men, each of them will serve as my teacher. I will pick out the good points of the one and imitate them, and the bad points of the other and correct them in myself. » JAM 119 / a little break, to ask questions. Contact http://www.jean-antoine-moreau.fr.nf Confucius
- 120. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 120 / PPrreeddiiccttiivvee ppaarrssiinngg Contact http://www.jean-antoine-moreau.fr.nf
- 121. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 121 / Predictive parser Contact http://www.jean-antoine-moreau.fr.nf
- 122. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 122 / Contact http://www.jean-antoine-moreau.fr.nf
- 123. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 123 / Contact http://www.jean-antoine-moreau.fr.nf
- 124. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 124 / The roles Contact http://www.jean-antoine-moreau.fr.nf
- 125. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Top-down parsing while scanning the input from left on right JAM 125 / Example : Contact http://www.jean-antoine-moreau.fr.nf
- 126. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 126 / Example: Using PLC algorithm. Contact http://www.jean-antoine-moreau.fr.nf decoder signal processing data processing PLC: • Packet Loss Concealment is a technique used to mask the effects of lost or discarded packets.
- 127. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 127 / parsing continued parsing continued algorithm Contact http://www.jean-antoine-moreau.fr.nf
- 128. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 128 / EExxaammppllee :: LLaanngguuaaggee ttrraannssllaattoorr Contact http://www.jean-antoine-moreau.fr.nf
- 129. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 129 / SSeenntteennccee aannaallyyssiiss Contact http://www.jean-antoine-moreau.fr.nf
- 130. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 130 / PPaarrssiinngg AAllggoorriitthhmm ddyynnaammiicc pprrooggrraammmmiinngg Contact http://www.jean-antoine-moreau.fr.nf
- 131. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 131 / the movement of the hand using tacking Contact http://www.jean-antoine-moreau.fr.nf
- 132. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 132 / BBeehhaavviioorr AAnnaallyyssiiss aanndd PPaarrssiinngg Contact http://www.jean-antoine-moreau.fr.nf
- 133. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 133 / Predictive parser • Top-down parser; • Implementations: – Recursive-descent parser; – Table-driven parser. Contact http://www.jean-antoine-moreau.fr.nf
- 134. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • Build parse tree from the tops (start symbol) down to leaves (terminals); JAM 134 / Top Down Parsing Contact http://www.jean-antoine-moreau.fr.nf
- 135. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 A grammar specifications may contain directives that are used to customize the generated parser or declare how some symbols will be used.. JAM 135 / Predictive parser • Predictive parser : top-down parser; • Implementations: – Recursive – descent parser: – Each non-terminal parsed by a procedure; – Call other procedure to parse sub-nonterminals recursively; • Table-driven parser: – Push-down automata; – Generate by a tool from a grammar specification. Contact http://www.jean-antoine-moreau.fr.nf
- 136. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 136 / PPaarrssiinngg pprroocceessss EExxaammppllee:: ffoorr aa sseenntteennccee ssttrruuccttuurree Step (1) Contact http://www.jean-antoine-moreau.fr.nf
- 137. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 137 / PPaarrssiinngg pprroocceessss EExxaammppllee:: ffoorr aa sseenntteennccee ssttrruuccttuurree Step (n+1) Contact http://www.jean-antoine-moreau.fr.nf
- 138. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 138 / Defining the predictive need Algorithm Method Tools Programming Contact http://www.jean-antoine-moreau.fr.nf
- 139. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • Construct parse tree for input; • Bottom-up parsing is more general than top-down JAM 139 / Bottom-up parsing parsing and just as efficient - generally preferred in practice. Contact http://www.jean-antoine-moreau.fr.nf
- 140. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 140 / LL(k) parsing LR(k) parsing • LL(k) parsing: Leftmost derivation • LR(k) parsing : – LLeft-to-RRight scan of input, • rightmost derivation; – K tokens of a look ahead. Contact http://www.jean-antoine-moreau.fr.nf
- 141. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 141 / Content analysis Topic modeling Contact http://www.jean-antoine-moreau.fr.nf
- 142. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. not final, failure is not fatal: it is the courage to continue that Winston JAM 142 / a break, you can ask your questions. Contact http://www.jean-antoine-moreau.fr.nf
- 143. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Search words that regularly come together in a document; Gather these words in a "theme« (area); Measure the frequency of each themes in each documents. JAM 143 / Topic modeling Contact http://www.jean-antoine-moreau.fr.nf
- 144. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 144 / Topic modeling Process's example Contact http://www.jean-antoine-moreau.fr.nf
- 145. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 145 / Topic modeling Machine learning Contact http://www.jean-antoine-moreau.fr.nf
- 146. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 146 / Machine learning Contact http://www.jean-antoine-moreau.fr.nf
- 147. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 147 / Machine learning Artificial intelligence Contact http://www.jean-antoine-moreau.fr.nf
- 148. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 148 / Artificial intelligence Modifiable Algorithm Contact http://www.jean-antoine-moreau.fr.nf
- 149. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 statistics analysis multidisciplinary modeling JAM 149 / Topic Modeling Contact http://www.jean-antoine-moreau.fr.nf
- 150. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 150 / Data Science Contact http://www.jean-antoine-moreau.fr.nf Data Mining Predictive analytical processing Machine Learning
- 151. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 151 / Natural Language Processing Contact http://www.jean-antoine-moreau.fr.nf
- 152. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 152 / Data Science Natural Language Processing Parsing Scanning Contact http://www.jean-antoine-moreau.fr.nf
- 153. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Machine learning Natural language JAM 153 / Process Topic Generative Model Contact http://www.jean-antoine-moreau.fr.nf
- 154. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 – There are many more complex kinds of language models, such as bigram language models , which condition on the previous term. JAM 154 / Unigram model • Types of language models • How do we build probabilities over sequences of terms? – The simplest form of language model simply throws away all conditioning context, and estimates each term independently. Contact http://www.jean-antoine-moreau.fr.nf
- 155. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • Probabilistic Latent’ Semantic Analysis : – is a novel statistical technique for the analysis of two--mode and co-occurrence data, which has applications in information retrieval and filtering, natural language processing, machine learning from text, and in related areas. JAM 155 / Contact http://www.jean-antoine-moreau.fr.nf
- 156. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 156 / Contact http://www.jean-antoine-moreau.fr.nf
- 157. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • Latent Dirichlet allocation (LDA), a generative probabilistic model for collections of discrete data such as text corpora. • LDA is a three-level hierarchical Bayesian model, in which each item of a collection is modeled as a finite mixture over an underlying set of topics. Each topic is inturn, modeled as an infinite mixture over an underlying set of topic probabilities. JAM 157 / LDA Model Contact http://www.jean-antoine-moreau.fr.nf
- 158. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 158 / Contact http://www.jean-antoine-moreau.fr.nf LDA Model
- 159. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 159 / three-level hierarchical Bayesian model Contact http://www.jean-antoine-moreau.fr.nf
- 160. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 160 / three-level hierarchical Bayesian model Example: the case of a image Contact http://www.jean-antoine-moreau.fr.nf
- 161. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 161 / three-level hierarchical Bayesian model Example: a scene of life Contact http://www.jean-antoine-moreau.fr.nf
- 162. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 162 / three-level hierarchical Bayesian model Example: a local Network Contact http://www.jean-antoine-moreau.fr.nf
- 163. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 a topic model for text or other discrete data that models correlation between the occurrence of, different topics in a document. For the CTM model, you can use the Value Estimation Model JAM 163 / CTM Model • The Correlated Topic Model (CTM): Contact http://www.jean-antoine-moreau.fr.nf
- 164. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Model of interacting decision-makers JAM 164 / Game Theory Strategic Game Contact http://www.jean-antoine-moreau.fr.nf
- 165. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 165 / Game Theory Nash Equilibrium Strategic Game Model of interacting decision-makers Contact http://www.jean-antoine-moreau.fr.nf
- 166. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • A sseett ooff ppllaayyeerrss (actors); • For each player, a sseett ooff aaccttiioonnss; • For each player, preferences over the sseett ooff aaccttiioonnss pprrooffiilleess. JAM 166 / SSttrraatteeggiicc ggaammee Consist of : Contact http://www.jean-antoine-moreau.fr.nf
- 167. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • As in the model of rational choice by a single decision-maker; • Time is absent from the model. JAM 167 / Contact http://www.jean-antoine-moreau.fr.nf
- 168. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • The idea is that each player chooses her action once and for all, – and the players choose their actions « simultaneously » in the sense that no player is informed, when the action is chosen by any other player. • However an action may involve activities that extend over time, and • But the fact that time is absent from the complications, that may arise, if a player is a allowed to change her plan as events unfold : we assume that actions an chosen once and for all. JAM 168 / may take into account an unlimited number of contingencies. Contact http://www.jean-antoine-moreau.fr.nf
- 169. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 169 / SSttrraatteeggyy ggaammee For each situation define : define Contact http://www.jean-antoine-moreau.fr.nf Players (actors) Actions preferences Specifications
- 170. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 170 / Specifications Matrix Contact http://www.jean-antoine-moreau.fr.nf Players (actors) Players (actors) Action 1 Action 2 Action 1 Action 2 Preference (Action 1, Action 1) Preference (Action 1, Action 2) Preference (Action 1, Action 2) Preference (Action 2, Action 2)
- 171. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 171 / Matrix Use case Contact http://www.jean-antoine-moreau.fr.nf Players (actors) Players (actors) Action 1 Action 2 Action 1 Action 2 Preference (Action 1, Action 1) Preference (Action 1, Action 2) Preference (Action 1, Action 2) Preference (Action 2, Action 2) Use case / set
- 172. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Use case Specification JAM 172 / Contact http://www.jean-antoine-moreau.fr.nf Players (actors) Players (actors) Action 1 Action 2 Action 1 Action 2 Preference (Action 1, Action 1) Preference (Action 1, Action 2) Preference (Action 1, Action 2) Preference (Action 2, Action 2) Use case / set Specification / set
- 173. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 173 / With the Decision Tree Contact http://www.jean-antoine-moreau.fr.nf
- 174. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 174 / With the Decision Tree Contact http://www.jean-antoine-moreau.fr.nf
- 175. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 175 / With the Decision Tree Contact http://www.jean-antoine-moreau.fr.nf
- 176. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 176 / Nash equilibrium Contact http://www.jean-antoine-moreau.fr.nf
- 177. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 What actions will be chosen by the players (actors) in a strategic game ? We wish to assume as in the theory a rational decision-maker, that each player chooses the best available action. In a game, the best actions for any given player depends, in general; on the other players’ actions. JAM 177 / Contact http://www.jean-antoine-moreau.fr.nf
- 178. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • We assume that each player has experience playing the game, we assume that he view each play of the game in isolation; • He does not become familiar with the behavior of specific opponents. JAM 178 / Contact http://www.jean-antoine-moreau.fr.nf
- 179. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • It is helpful to think of the following idealizing circumstances: – For each player in the game there is a population of the game, there is a population of many decision-makers who many, on many occasions, take that player's role; – Thus each player engages in the game repeatedly, against ever-varying opponents; – Her experience leads her to beliefs about the actions of "typical" opponents. JAM 179 / Contact http://www.jean-antoine-moreau.fr.nf
- 180. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 NNaasshh eeqquuiilliibbrriiuumm A NNaasshh eeqquuiilliibbrriiuumm is an aaccttiioonn pprrooffiill (a) with the property that no player (I) can do better by choosing an action different from (a), given that every other player (j) adheres to (a). A Nash equilibrium ccoorrrreessppoonndd ttoo aa sstteeaaddyy ssttaattee. JAM 180 / Contact http://www.jean-antoine-moreau.fr.nf
- 181. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 The second component of the theory of Nash equilibrium, that the player's beliefs about each other's actions are correct-implies, in particular, that two players' belief about a third player's action are the same. For this reason, the condition is sometimes said that the players' "expectations are coordinated". JAM 181 / Contact http://www.jean-antoine-moreau.fr.nf
- 182. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. JAM 182 / a break, you can ask your questions. « What we play is life. » Louis Armstrong of course Contact http://www.jean-antoine-moreau.fr.nf
- 183. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 a Nash equilibrium designed to model a steady state among experienced players. JAM 183 / Contact http://www.jean-antoine-moreau.fr.nf
- 184. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 How to know? How to understand ? an alternative approach to understanding players' actions in strategic games assumes that the players know each others' preferences, and considers what each player can deduce about the other players' actions from their inclination or tastes and their knowledge of each other's rationality. JAM 184 / Contact http://www.jean-antoine-moreau.fr.nf
- 185. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 185 / Specification define de requirement use a Nash equilibrium Contact http://www.jean-antoine-moreau.fr.nf
- 186. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Use the Nash equilibrium Specification JAM 186 / Contact http://www.jean-antoine-moreau.fr.nf Players (actors) Players (actors) Action 1 Action 2 Action 1 Action 2 (1,1) Preference (Action 1, Action 2) Preference (Action 1, Action 2) Preference (Action 2, Action 2) to Use case / set Specification / set The Nash Equilibrium
- 187. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. “Patience is the companion of wisdom. » JAM 187 / a break, you can ask your questions. Contact http://www.jean-antoine-moreau.fr.nf Saint Augustine
- 188. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 188 / BBuussiinneessss UUssiinngg Contact http://www.jean-antoine-moreau.fr.nf
- 189. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 189 / How to do ? How to use ? Contact http://www.jean-antoine-moreau.fr.nf
- 190. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 190 / • Data crunchers; • Data modeling; • Data modeler; • Data distillers; Contact http://www.jean-antoine-moreau.fr.nf
- 191. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Computer Science JAM 191 / • Data crunchers; • Data modeling; • Data modeler; • Data distillers; Contact http://www.jean-antoine-moreau.fr.nf
- 192. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Computer science Business Analysis JAM 192 / • Data crunchers; • Data modeling; • Data modeler; • Data distillers; Contact http://www.jean-antoine-moreau.fr.nf
- 193. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 193 / Tools for • Human Resources • Financial • Prospective / Perspective • Marketing Contact http://www.jean-antoine-moreau.fr.nf
- 194. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Get new clients Get new markets proactivity JAM 194 / Tools for • Human Resources • Financial • Prospective / Perspective • Marketing Contact http://www.jean-antoine-moreau.fr.nf
- 195. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Get new clients Get new markets proactivity WWiinnnniinngg tthhee iinntteerrnnaattiioonnaall ccoommppeettiittiioonn JAM 195 / Tools for • Human Resources • Financial • Prospective / Perspective • Marketing Contact http://www.jean-antoine-moreau.fr.nf
- 196. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 196 / RRiisskk MMaannaaggeemmeenntt • Performance management; • Cost control; • Risk Analysis; • Fraud detection; • … Contact http://www.jean-antoine-moreau.fr.nf
- 197. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 197 / OOppeerraattiioonnss aanndd pprroodduuccttiioonn • Supply chain optimisation; • Resource allocation; • Human Resource analysis; • Asset Management; • Customer Relationship’ Management; • Marketing approach; • ... Contact http://www.jean-antoine-moreau.fr.nf
- 198. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 198 / SSeerrvviicceess • Data science consulting; • Business analytics; • Business intelligence; • Data driven applications; • … Contact http://www.jean-antoine-moreau.fr.nf
- 199. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 199 / Data Contact http://www.jean-antoine-moreau.fr.nf • Web browsing; • Social networks; • Commercial transactions; • Financial transaction; • Industrial sensors; • Smart grids; • Geolocation; • Sensors; • Medical equipment; • Competitive intelligence; • Public data; • Pooled data; • Open data; • Data offered.
- 200. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 200 / Parallel programming model Distributed File System Information management tools Contact http://www.jean-antoine-moreau.fr.nf MapReduce Hadoop HDFS Hadoop • NoSQL database; • Data mining; • Machine Learning; • Query Tool;
- 201. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 201 / OOppeerraattiioonnaall aarreeaass of use: • Analytical; • Action; • Decision; • Information; • Data; • Financial. Contact http://www.jean-antoine-moreau.fr.nf
- 202. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 202 / UUssee CCaassee • Industrial asset management; • Target marketing; • Price transparency; • Remote diagnostics; • Carrier mobility; • Production management; • Risk management. Contact http://www.jean-antoine-moreau.fr.nf
- 203. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Data science, use process in the company measure understand predict decide acting Innovate JAM 203 / Contact http://www.jean-antoine-moreau.fr.nf
- 204. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 external data composite data JAM 204 / Data Contact http://www.jean-antoine-moreau.fr.nf data segment data on individuals internal data transversal data
- 205. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 205 / PPrroocceessss 1. Formalization of the issues; 2. Data collection; 3. Formalization of the issues; 4. Model development; 5. Actions of automation; 6. Decisions; 7. Commitment decisions; 8. Implementation. Contact http://www.jean-antoine-moreau.fr.nf
- 206. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 206 / Logical architecture Contact http://www.jean-antoine-moreau.fr.nf Transactional data Machine Data Text - Image - Video Data Business Application Data Lab Business Intelligence Actors - Players Clients Data Scientist Business Manager Others …
- 207. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 207 / DDaattaa SScciieennttiisstt jjoobb aanndd rroolleess ffoorr tthhee bbuussiinneessss • Correlation analysis; • Causal study; • Predictive modeling; • Influence of measurement; • Detecting Patterns. Contact http://www.jean-antoine-moreau.fr.nf
- 208. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • Strategic planning; • Tactical analysis; • Operational management; • Study the customers behavior or set of clients. JAM 208 / BBuussiinneessss MMaannaaggeerr Contact http://www.jean-antoine-moreau.fr.nf
- 209. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 209 / SSkkiillllss • Statistics; • Machine Learning; • Marketing; • Network; • Operations; • Architecture; • Algorithms; • Application; • Computer science application. Contact http://www.jean-antoine-moreau.fr.nf
- 210. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Thomas Crombie Schelling’s Model JAM 210 / Contact http://www.jean-antoine-moreau.fr.nf
- 211. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 Improve our understanding of the mechanisms of conflict and cooperation, through the analysis of game theory. " JAM 211 / Contact http://www.jean-antoine-moreau.fr.nf
- 212. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • The game is played on a checkerboard, which is a city; • The checkerboard is filled of dimes and nickel representing two differents types of agents; • The different types of agents can be thought of: different sexes, different consumers or others ... • Each agent evaluates its current position based on a " happiness rule ", which depends on • Unhappy agents are allowed to switch places with the other (there are various ways); • The game continues until all the agents are happy, then it represents the equilibrium outcome. • Then all the agents found the place where they are happy. They are next to those who have the JAM 212 / Thomas Crombie Schelling’s Model, Example : adjacent squares (happiness rule, example : I love ice cream); same “happiness rule” than them. Equilibrium is reached . Contact http://www.jean-antoine-moreau.fr.nf
- 213. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 example: common rules, consolidation area and area of influence. Area ( b) JAM 213 / Contact http://www.jean-antoine-moreau.fr.nf Rule (a) Rule ( c ) Rule (b)
- 214. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 JAM 214 / The checkerboard can be: – A store chain; – Business services; – Financial services; – International business; – … Contact http://www.jean-antoine-moreau.fr.nf
- 215. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 If the checkerboard is a concert hall. The outcome primarily depends on the happiness rule. One example of a happiness rule would be "I want all my neighbors are musicians". Not surprisingly this rule give the outcome of complete orchestra. JAM 215 / Contact http://www.jean-antoine-moreau.fr.nf
- 216. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • To implement this model, you only need information about the individual preferences of each of the players (actors), and a dichotomy for defining distinct groups. JAM 216 / Contact http://www.jean-antoine-moreau.fr.nf
- 217. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • The model Schelling shows with a very simple simulation, that a strong differentiation may be the collective result of individual decisions, which are not intended to make this differentiation. JAM 217 / Contact http://www.jean-antoine-moreau.fr.nf
- 218. © Jean-Antoine Moreau copying and reproduction prohibited Managing my copyright ADAGP. DDAATTAA SSCCIIEENNCCEE -- LL44 • End of part of this lecture published on the internet. JAM 218 / Contact http://www.jean-antoine-moreau.fr.nf