About This Course
Published by :
The telecom industry is going through a digital transformation across the world with a rigorous focus on providing personalized mobile services, understanding the consumer better, monitoring social media channels to identify customer needs, and supporting applications in various domains like smart cities and healthcare. This transformation is enabled by game-changing technologies such as Artificial Intelligence (AI) and Internet of Things (IoT).
Telecom companies are looking to leverage terabytes of data generated through their platforms and generate actionable insights. For this purpose, they are looking to expand their workforce by hiring people who can analyze and interpret data using techniques based on artificial intelligence and machine learning.
Objective: The course objective is to equip engineering students with introductory skills in analyzing data, finding relevant patterns and solving real-world problems, especially for the telecom industry. After the course, students will be able to:
1) Formulate problem statements based on an analysis of requirements.
2) Collect the data related to a given problem.
3) Analyze/visualize, clean and manipulate the collected data.
4) Identify the right techniques to build, validate and test the model.
5) Present the analysis to all the stakeholders.
The course will be taught in a blended style, combining online and instructor-led learning. The instructors will be industry practitioners with expertise in AI and data science. The course will be run over a 10 week period, and organized into different modules. Each module will run for a week with a workload of ~10 hours/week. There will be a live/Q&A session every week for 2-3 hours and 7 hours of self-learning using curated videos and programming assignments. The curated videos will feature world-class experts in AI, and will be made available on the award-winning VideoKen platform.
1. Introduction to Python
(Setting up of Environment, Basic Data Types, Variables, Use of Python Lists, Functions/Methods, Python Packages, NumPy Arrays)
2. Data Manipulation in Python
(Matplotlib, Pandas/Data Frames, Grouping data in logical pieces)
3. Basic Statistics & Data Analysis
(Exploratory Data Analysis, Descriptive Statistics, Population & Sample, Hypothesis Testing)
4. Supervised Learning
(Linear Regression, Logistic Regression, Decision Trees, Supervised Classification Techniques)
5. Unsupervised Learning
(K-means Clustering, Hierarchical Clustering, PCA)
6. Neural Networks
(Introduction to Deep Learning, Forward & backward propagation, Activation Functions, Optimization Functions, Creating & Running your own Neural Network)
7. Deep Learning for Computer Vision
(CNN, Deep Models, AlexNet/ImageNet, Visual Classification using pre-trained models)
8. Deep Learning for Text / Natural Language Processing
(Sequence Modelling, RNN/LSTM, Working with data sequences, Text Classification)
9. Two weeks of Capstone projects related to Telecom Analytics, some of the suggested topics are:
(Customer Churn Prediction, Social Media Sentiment Analysis, Customer Segmentation, Call Drop Analysis, Activity Detection & Classification on Mobile Devices)
What are the requirements?
Section 1: Getting Started