Vai al contenuto principale
You are currently viewing this course as Ospite.
Marcello MASCINI

Food science and technology - 1st year

Indice degli argomenti

  • Prof Mascini

    Marcello Mascini is associate professor in analytical chemistry. In his academic career Prof Mascini has published more than 100 papers (70% as corresponding author) on international scientific journals (peer-reviewed) with an average Impact Factor of 4.2.

    He actively participated in national and international scientific meetings with more than 200 posters or oral presentations (abstracts). Prof. Mascini research area was focused on the development of screening methods for fast and real time detection of analytes important for health and quality control analysis. The research interests were with a particular focus on new methods to develop bio-synthetic systems (biomimetic or bioinspired) in analytical application using molecular modeling and advanced multivariate systems. 

    INFO ABOUT THE COURSE 

    OBJECTIVES OF THE COURSE

     The course aims to increase the knowledge of pre and post processing experimental data with multivariate statistical techniques applied to the analysis of foods.

    This course will emphasize computer approaches to multivariate statistical analysis. We will discuss how to design, conduct, and analyze experiments in food sciences Various designs will be discussed and their respective differences, advantages, and disadvantages will be noted. We will examine techniques for data reduction (principal components, factor analysis, and cluster analysis) and for discrimination and classification (cluster analysis, discriminant analysis).  In the first part the course will examine how to design experiments, carry them out, and analyze the data they yield. In the second part it will be compared univariate and multivariate statistical techniques (PCA and PLS). Case studies related to research projects will be taken as practical examples and they will be carried out by using academic free software

    PREREQUISITE AND PREPARATORY

    - Prerequisite: elementary mathematics for learning statistical basis. In particular, it is necessary to have knowledge about the normal distribution of Gauss, variance and standard deviation calibration, regression, least squares, chi-square, measurement error, precision and accuracy

    Preparatory: No

    UNIT 1: Univariate analysis

    Data, information, models, data types, analytical representation of data 

    Calibration and regression, Introduction to Statistics

    Average & Variance

    The Normal distribution, theory of measurement errors, the central limit theorem and the theorem of Gauss

    Maximum likelihood, method of least squares, Generalization of the method of least squares

    Polynomial regression, non-linear regression, the χ2 method, Validation of the model

    UNIT 2: Multivariate analysis

    Correlation

    Multiple linear regression

    Principal component analysis (PCA) 

    Principal component regression (PCR) and Partial least squares regression - (PLS)



    UNIT 3: Design of Experiments

    Basic design of experiments and analysis of the resulting data

    Analysis of variance, blocking and nuisance variables

    Factorial designs

    Fractional factorial designs

    Overview of other types of experimental designs (Plackett–Burman designs, D-optimal designs, Supersaturated designs, Asymmetrical designs) 

    Response surface methods and designs

    Applications of designed experiments from various fields of food science

    UNIT 4: Elements of Pattern recognition

    cluster analysis

    normalization

    The space representation (PCA) 

    Examples of PCA

    Discriminant analysis (DA) 

    PLS-DA

    Examples of PLS-DA


    WEEKLY LESSONS 

    COURSE BOOKS

    Because of the practical application nature of this course there is no mandatory textbook. Instead, you should purchase a text that suits your needs (e.g., practical application versus mathematical statistics). Recommended texts are: 

    Johnson, Dallas E. (1998). Applied multivariate methods for data analysis. Pacific Grove, CA: Duxbury Press. Good balance between theory and practice. 

    Tabachnick, B. G. & Fideii,L.S. (2000). Using Multivariate Statistics, 4th Ed. New York: Allyn & Bacon. A traditional and popular text that focuses on practical applications. 

    Oehlert, Gary W.  (2010). A first course in design and analysis of experiments. (http://users.stat.umn.edu/~gary/book/fcdae.pdf

    Barrentine Larry B. (1999) An Introduction to Design of Experiments: A Simplified Approach Amer Society for Quality

    RESEARCH MATERIAL 

    Slides of the lessons

    Author: Prof Mascini, Prof Sacchetti

    INTERMEDIATE TESTS 

    Date: at the end of the unit (please see the planning)

    Test type: Multiple choice questions (MCQs)

    EVALUATION 

    The tests are held during the semester at the end of the units and are a series of multiple choice questions, related to the specific arguments of the units. The correct answer to each question is 1 point. Wrong answer or no answer  is 0 points. The maximum score evaluation is 23/30. It should be noted that to have 30/30 it is mandatory to do the report (see below).

    In case of you can not do the tests during the course, you can do a final multiple choice quiz  at the end of the semester. 

    The score obtained in the tests will be kept up for one year.

    A presentation is requested in form of report which will highlight potential, limitations and possible developments of the work performed. Non-attending students are asked to submit a report online evaluated by the teacher and the attending students. The report is not mandatory but it allows to have an evaluation of 30/30 (with honors). If you don't do the report you will have a maximum evaluation of 23/30.

    During the lessons, students can check their learning, through the online test simulator, similar to the examination tests.

    The Teacher is available to answer questions at the end of the lesson, or on request by e-mail (mmascini@unite.it)


     

  • Data, information, models, data types, analytical representation of data

    Calibration and regression, Introduction to Statistics

    Average & Variance

    The Normal distribution, theory of measurement errors, the central limit theorem and the theorem of Gauss

    Maximum likelihood, method of least squares, Generalization of the method of least squares

    Polynomial regression, non-linear regression, the χ2 method, Validation of the model


  • Correlation

    Multiple linear regression

    Principal component analysis (PCA)

    Principal component regression (PCR) and Partial least squares regression - (PLS)

     

  • Basic design of experiments and analysis of the resulting data

    Analysis of variance, blocking and nuisance variables

    Factorial designs

    Fractional factorial designs

    Overview of other types of experimental designs (Plackett–Burman designs, D-optimal designs, Supersaturated designs, Asymmetrical designs)

    Response surface methods and designs

    Applications of designed experiments from various fields of food science


  • Cluster analysis

    Normalization

    The space representation (PCA)

    Examples of PCA

    Discriminant analysis (DA)

    PLS-DA

    Examples of PLS-DA