28 Pages

COMP3702 - AI applications

Course: COMP 3702, Fall 2009
School: Allan Hancock College
Rating:
 
 
 
 
 

Word Count: 538

Document Preview

of State the art Applications of AI Outline Speech Processing Natural Language Processing Image Processing Bioinformatics Diagnosis Drug Discovery Understanding of Biological Systems Cheminformatics Speech Processing Speech Synthesis Speech Synthesis/Recognition Speech Recognition Modelling Modelling Speaker Recognition Speaker Recognition Speakers gender/ age group To recognize speaker Identification...

Register Now

Unformatted Document Excerpt

Coursehero >> California >> Allan Hancock College >> COMP 3702

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.
of State the art Applications of AI Outline Speech Processing Natural Language Processing Image Processing Bioinformatics Diagnosis Drug Discovery Understanding of Biological Systems Cheminformatics Speech Processing Speech Synthesis Speech Synthesis/Recognition Speech Recognition Modelling Modelling Speaker Recognition Speaker Recognition Speakers gender/ age group To recognize speaker Identification Verification Text Dependent Text Independent Other Language Processing Applications Natural Language processing To analyse and process natural languages Text Mining Extracting reliable information from text Search Engines Context based search engines Spell checker in Google Face Recognition Face Recognition Other Image processing based applications Number plate extraction Optical Character Recognizers Image to text Google books Medical applications Diagnosis MRI data Computational Biology/Systems Biology/Bioinformatics Human Genome Project Goals: identify all the approximately 20,000-25,000 genes in human DNA, that make up human DNA, determine the sequences of the 3 billion chemical base pairs store this information in databases, improve tools for data analysis, transfer related technologies to the private sector, and address the ethical, legal, and social issues (ELSI) that may arise from the project. Diagnosis Intelligent systems to learn to discriminate between normal and patient data. Applicable for genetic disorders Will be learning in the second part of the course Protein patterns may be able to predict prostate cancer Diagnosis and Drug Discovery Vastly improved test for detecting ovarian cancer in 2002 by Correlogic Correlogics research is based on a novel approach that looks for subtle changes in protein and other serum molecule patterns, rather than simply an increase in an individual molecule. It employs patented intelligence-based artificial computer technology to identify these hidden patterns. http://www.correlogic.com/approach/index.php Finding new diseases for known cures. Biomedical image analysis Protein structure prediction Systems Approach S. E. Calvano et al., Nature Letter, 2007. Systems Approach S. E. Calvano et al., Nature Letter, 2007. Systems Approach Construct graph Search through the graph Find relationships between proteins and drugs S. E. Calvano et al., Nature Letter, 2007. Chem-informatics Use of computers for a variety of chemistry problems In-Silico drug discovery Small molecules Where its going we are within a few decades of building a computer, a simulacrum, that can experience the color red, savor the smell of a rose, feel pain and pleasure, and fall in love. It might be a robot with a body. Or it might just be softwarea huge, ever-changing cloud of bits that inhabit an immensely complicated and elaborately constructed virtual domain. IEEE Spectrum's SPECIAL REPORT: THE SINGULARITY, June, 08. People I work with ... Pfizer Research and Technology Centre, Cambridge,...

Find millions of documents on Course Hero - Study Guides, Lecture Notes, Reference Materials, Practice Exams and more. Course Hero has millions of course specific materials providing students with the best way to expand their education.

Below is a small sample set of documents:

Allan Hancock College - COMP - 3702
Tutorial 4 Minimax algorithm (an adversarial search) for Tic-tac-toe game:Tutorial 4 Two-ply search means that your resulting tree will contain three levels *including* the root node. The values represent expected utility:Tutorial 4 Estimatin
Allan Hancock College - COMP - 3702
Uncertain knowledgeRussell and Norvig, Chapter 13, + 7.1-7.2.Overview: aimshave a feel for the limitations of logic for dealing with uncertainties (logic is the theme for chapters 7-10) understand the basics of probability theory and know conce
Allan Hancock College - COMP - 3702
Assignment 2 26 letters: A-Z, encoded 0-25. 5605 examples in cleaned dataset (~200 of each). Assume that each letter is equally likely (not true in English or other natural languages). Notation in code is older form, e.g. d for desired output, y
Allan Hancock College - COMP - 3702
Machine learning Symbolic techniquesRussell and Norvig, chapter 18, 19 (section 19.1)Machine learning: Symbolic techniques Overview: aimsknow of several symbolic machine learning techniques and representations, decision tree learning current-be
Allan Hancock College - COMP - 3702
Statistical machine learningwhere Data are evidence (instantiations of random variables) Hypotheses are probabilistic theories of how the domain worksChapter 20 (only parts of sections 20.1 and 20.2 so far).Overview: aimsUnderstand the applic
Allan Hancock College - COMP - 3702
Non-symbolic machine learning Neural networks Russell and Norvig, Section 20.5Overview: aims know what a neuron / unit is understand single-layer neural networks understand multi-layer neural networks know what a learning error is un
Allan Hancock College - COMP - 3702
COMP3702 / COMP7702 Artificial IntelligenceReview LectureiCEVAL Institutional Course EvaluationCompetition 17 working classifiers 50 letters 1 bonus mark to all entrants 1 bonus mark to top 4 classifiersReview of Lecture Material1. 2
UCSB - MCDB - 112
MCDB 112, Winter 2007Lecture Set 1 - "First Principles" Diagram set 1Apage 1Lecture Diagram set #1A. The First Principles of Developmental Biology Figure 1-1. Frog Developmental life cycle.Figure 1-2. The 3 germ layersDiagram of evolutionar
UCSB - MCDB - 112
MCDB 112, Winter 2007Lecture Set #1 - First Principles Diagram Set #1Bpage 1Lecture set #1B Diagrams Figure 1-12 IntegrinsFigure 1-13. Canonical RTK PathwayMCDB 112, Winter 2007Lecture Set #1 - First Principles Diagram Set #1Bpage 2A
UCSB - MCDB - 112
MCDB112 W07Cell Death in Development (Diagram Set #10)1PROGRAMMED CELL DEATH (PCD) IN ANIMAL DEVELOPMENT Gilbert website 6.4 worth checking out See pages 1598-160 in text PPT Figure 6.24; diagram set #10-1. Apoptosis pathways in nematodes and m
UCSB - MCDB - 112
MCDB 112W07Diagram Set #11 Birth Defects1Lecture Set #11. Human development and birth defects Powerpoint Figure 21.3; Diagram set #11-1. Fate of 20 eggs in the US and western EuropeWe cannot experiment on human embryos and we cannot selecti
UCSB - MCDB - 112
MCDB 112. W07 (Foltz)Fertilization Lecture diagram set #21Text book Chapter 7Figure 2-1 (TEXT FIGURE 7.5) shows stages of oocyte maturation and when fertilizedFour Major Activities of Fertilization know them cold! 1. Contact and Recogniti
UCSB - MCDB - 112
MCDB 112 W2007Diagram Set #41Lecture Set 4: Fate Specification - Symmetry Breaks and Morphogen Gradients Outline 1. Asymmetric cell division as a way of creating breaks - the P granules & PAR genes of C. elegans 2. induction v lateral specifica
UCSB - MCDB - 112
MCDB 112 W07diagram set #61Set #6: Vertebrate embryonic axis specification Xenopus axis specification (Chapter 10) The side opposite of sperm entry will be the future DORSAL side while the site of sperm entry will be the VENTRAL side - The site
UCSB - MCDB - 112
MCDB 112 W071Diagram Set #7 - PATTERN FORMATION: THE HOMEOTICS Figure 9.35A; diagram set #7-1 expression map of Drosophila homeotic genes Antennapedia Complex contains the homeotic genes that specify the gene batteries of the head and thoracic s
UCSB - MCDB - 112
MCDB 112W07 Set #8page 1NEURULATION -Chapters 12 and 13 in textbook Figure 12.1; diagram set #8-1. Major derivatives of the ectoderm germ layerTwo basic types of cell movements used to form the neural tube: 1. Primary neurulation: cells surr
UCSB - MCDB - 112
MCDB 112 W07Diagram Set #9 Limb Development1TETRAPOD LIMB DEVELOPMENT (Chapter 16 in Gilbert) FIGURE 16.1; diagram set #9-1. Skeletal pattern of chick wing.TIME PROXIMAL-DISTAL ELEMENT STYLOPOD ZEUGOPOD AUTOPOD Forelimb: humerus radius, uln
ECCD - MATH - 211
MATH 211 LECTURE 7TJ HITCHMANThe Theory of Dierential Equations, Part II We continue our study of the theory of dierential equations. Today we focus on two questions: Is it possible to have more than one solution to an initial value problem? What
ECCD - MATH - 211
211 LECTURE 25 Today we study inhomogeneous equations x = Ax + f . The solution set always has the form xp + xg where xp is a particular solution and xg is the general solution to the associated homogeneous problem. As usual, this means that the key
ECCD - MATH - 211
Math 211, Spring 2005 Exam II SolutionsProblem 1 Eulers method is an algorithm for approximating solutions to initial value problems numerically. Suppose we have an initial value problem of the form x =f (x, t) x0 . x(t0 ) =Then Eulers method is t
ECCD - MATH - 211
211 LECTURE 19Systems of Dierential Equations, part II A note on something I should have stressed last time. A system needs an initial condition, too. But for a system, an initial condition has more information in it because it is a vector equation
Michigan - ENGR - 100
StatisticsWhy havent we needed statistics so far in this class?Maximum clock frequencyIs there a limit to how fast you can set the clock? clockregisterplus1plus1_outENGR 1001Peter ChenSpeed binning of CPU chipsCPU chips vary in the
University of Florida - CNT - 4007
CNT 4007 Computer Network Fundamentals, Spring 2009 Assignment 3given by Jonathan C.L. Liu Out: Mar. 25 (Wednesday), 2009 Due: Beginning of the lecture on April 01 (Wednesday), 2009 The problem sets are based on the textbook "Computer Networking" b
University of Florida - CAP - 4621
Homework 3This assignment is due on Friday, November 19, 2004. The assignment is again somewhat longer than the previous two assignments; therefore, you are advised to start working on it immediately. Each problem is equally weighted. As always, sho
University of Florida - CNT - 5106
TCP Vegas: New Techniques for Congestion Detection and Avoidance Lawrence S. BrakmoSean W. OMalley Department of Computer Science University of Arizona Tucson, AZ 85721Larry L. PetersonAbstractVegas is a new implementation of TCP that achie
University of Florida - YX - 5106
TCP Vegas: New Techniques for Congestion Detection and Avoidance Lawrence S. BrakmoSean W. OMalley Department of Computer Science University of Arizona Tucson, AZ 85721Larry L. PetersonAbstractVegas is a new implementation of TCP that achie
University of Florida - CIS - 6930
Problem 1: Combine the weight and height attributes in to a ratio. This will not reduce the dimensionality of the data, but also will eliminate the need to scale and normalize the continuous attributes. Here is the modified dataset: Name Kristina Jim
ECCD - MATH - 209
MATH 209: FINDING SECOND SOLUTIONSSTEVEN J MILLER, SJM1@WILLIAMS.EDU Abstract. In studying second order linear constant coecient homogenous dierential or dierence equations, we saw that our divine inspiration guess only generates one solution if the
ECCD - MATH - 209
Solving Polynomial EquationsPart IIIFerrari and the BiquadraticFerrari's solution of the quartic (biquadratic) equation involved the introduction of a new variable and then specializing this variable to put the equation into a form that could eas
University of Florida - CGS - 3063
GOOD, BAD, & INFORMATIONHISTORICAL COSTS PER TRANSACTION, AS VOLUME INCREASES1201008060x$/trans40200 volume of transactions manual computer computer 2 manual 2xVolume of transactionsWhy do weproce data? ssTo produce inform
University of Florida - CIS - 3022
CIS 3022 Prog for CIS Majors I _ March 26, 2007 Exam IIPrint Your Name Your Section # _Total Score _ Your work is to be done individually. The exam is worth 100 points (although 104 points are available) and it has eight questions. Use these meth