Value Added Courses

Title: Geospatial Data Analysis using Python (PU/GEOG/VACC-1/23-24)

Advertisement Date: 15/09/2023

NOTICE FOR SHORTLISTED PARTICIPANTS

Course Type: Certificate

Commencement Date and Time: 23 ^rd November 2023 (from 5 pm to 7 pm)

(Date and Time are subject to change)

Date of Completion: Tentatively by the last week of January 2024

Duration of the Course: 30 (Thirty) Hours

Schedule of the Classes: 1 hour/class; 2 classes/day (From 5 pm to 7 pm); 2 days/week

Mode of Course: Offline

Medium of Instruction: English

Course Fees: 10,300/- (Ten Thousand and Three Hundred only)

(Mode of Payment will be intimated subsequently to the shortlisted candidates)

Eligibility (Who can apply): Graduate in any discipline of Earth sciences

Course Coordinator Name: Dr. Priyank Pravin Patel and Soumendu Chatterjee

Email ID of Course Coordinator: priyank.geog@presiuniv.ac.in / soumendu.geog@presiuniv.ac.in

Course Objective:

Machine learning (ML), an important sub-field of Artificial Intelligence (AI), finds multifarious applications in Earth Systems Science. The high performance of ML techniques in handling classification and regression problems and for pattern recognition has drawn tremendous academic interests among the researchers and professionals. This course aims at imparting live training on basics of machine learning (ML) techniques and their implementation in Python using data related to the Earth.

Learning Outcome:

· Basics of Python programming and using libraries like numpy, pandas, skleran etc. for handing data.

· Build, train and evaluate ML models from data for prediction and classification purposes.

· Expertise in supervised ML algorithms including linear and logistic regression, decision tree, random forest, support vector, KNN and PCA.

Module content:

UNIT 1: PYTHON BASICS I

1.1 Programming Methodology: Algorithm and Flowchart

1.2 Introduction te Python: Installation of Python environment (Jupyter Notebook in ANACONDA); Variables and types, Operators and Operands, Statements, Input and Output, Modules and built-in functions; Conditional and Looping constructs (if-else, while and for loops with range method for simple and nested cases); Variable Scope; User defined functions.

1.3 Strings: Traversing, slicing, operations, methods and built-in functions, regular expression and pattern matching

1.4 Lists: Creating list, Accessing, traversing, adding, updating arid deleting elements, list functions and methods, list as an argument, matrix implementation.

UNIT 2: PYTHON BASICS II

2.1 Dictionaries: Key-value pair, Creating, initializing, accessing, traversing elements, dynamic allocation, appending value, merging dictionaries, removing items, dictionary functions and methods

Tuple: Creating tuple, appending element, assignment, slicing, tuple functions and methods.

2.2 Object Oriented Programming: Concept, Class and Object creation, Object attributes and class attributes, Methods

2.3 Numpy: Importing csv file, handling arrays, implementing mathematical operations on arrays using numpy.

2.4 Pandas: Accessing data, manipulating data in dataframe

2.5 Plotting of data using matplotlib

UNIT 3: BASIC SUPERVISED MACHINE LEARNING ALGORITHMS AND THEIR CODING I

3.1 Types of Machine Learning; Steps in supervised machine learning

3.2 Bivariate and Multiple Linear Regression: Mathematical background; Score, Cost function, gradient descent and learning rate; Coding Linear regression using sklearn.

3.3 Binary and Multi-class Logistic Regression: Cost function, gradient descent, classification measures and their coding in Python.

3.4 Decision Tree: Concept; Building Decision Tree for discrete and continuous data, accuracy assessment; Implementation in sklearn and pruning.

UNIT 4: BASIC SUPERVISED MACHIHE LEARNING ALGORITHMS AND THEIR CODING II

4.1 Random Forest: Concept and implementation in python.

4.2 K-Nearest Neighbour: KNN algorithm and its implementation in python.

4.3 Support Vector Machine: SVM algorithm; SVM cost function and accuracy; Non- linear decision boundary and kernels; Multi-class classification; Implementation in python.

4.4 Principal Component Analysis: Mathematical background; Application of PCA on 2D and 3D data; Coding for PCA; Application in Image classification.

Resource Persons:

Eminent professionals from relevant fields will take the classes

General terms and conditions:

· The aspiring candidates must fill in all the mandatory fields given in the Google form within the time frame from 15 ^th September 2023 to 8 ^th October 2023 mid-night (24.00 hrs).

· Once submitted, the candidates cannot change their responses in the form.

· The application form link is –

· The selection of candidates will be done on a first-come, first-serve basis.

· The maximum number of candidates will be decided based on the availability of resources; however, the minimum number will be 15 (fifteen).

· Selected participants will be informed about the payment method subsequently via email along with their Application Number.

· On successful submission of Course fees, participants will be notified about their enrollment number and commencement details of the course.

· Once a candidate gets enrolled, the fees will be non-refundable.

Sd/-
Dr. Priyank Pravin Patel
Prof. Soumendu Chatterjee
Course Coordinators