Title: Working with Big Data: Tools, Techniques, and Practical Implementation
Md Manirul Islam
Director, Institute of Continuing Education Associate Professor, Faculty of Science and Technology American International University Bangladesh, Dhaka, Bangladesh
With the emergence of new technologies, chin-deep indulgence into social media platforms, and the extensive expansion of Internet of Things, more and more data are being created each day and growing exponentially. For instance, a jet engine can generate 10 Terabytes of data for only 30 minutes of flying. Facebook generates 4 petabytes of data per day. This massive load of data termed as Big Data has become an indispensable part of our lives and is the key basis for productivity growth, business innovation and intelligence in today’s competitive business world.
In this tutorial, we will discuss the tools and techniques that are being used to harness the power of Big data. Further we will dig deep into Cassandra, and Hadoop which are great tools for superior processing of voluminous data sets in clusters of hardware using effective programming models.Biography:
Md Manirul Islam is an Associate Professor of Computer Science and Director of Institute of Continuing Education and IT at the American International University-Bangladesh (AIUB). He is the lead architect of his university’s Data Center and Network Infrastructure. Mr. Islam is a Member of the Cisco Networking Academy’s Global Advisory Board. He holds several industry certifications in the track of networking and system administration. He holds a CCNP Certification and an award-winning Instructor Trainer for IT Essentials, CCNA, CCNA Security, Cybersecurity Operations, DevNet, IoT Security, IoT and Big Data Analytics. Mr. Islam has several journal publications, and his research interest lies in the areas of Quantum Networking, IoT, and Big Data Analytics.
Title: Implementing and Optimizing Deep Learning Models for Human Activity Recognition (Hands-on with Python)
Anindya Das Antar,
PhD Research Fellow, Computer Science and Engineering,Computational HCI Lab, University of Michigan, Ann Arbor, USA
Human activity recognition, or HAR, is a challenging time series classification task. It involves predicting the movement of a person based on sensor data and traditionally involves deep domain expertise and methods from signal processing to correctly engineer features from the raw data in order to fit a machine learning model. Recently, deep learning methods such as convolutional neural networks and recurrent neural networks have shown capable and even achieve state-of-the-art results by automatically learning features from the raw sensor data. This workshop will cover mostly the theoretical background of deep learning with an example tutorial on building an RNN-LSTM model for sensor-based Human Activity Recognition (HAR). How to optimize a model from a researcher’s perspective will be presented in brief at the end of the talk.Overview:
Background of Human Activity Recognition (HAR)
Evaluation of Deep Learning in Sensor-based HAR
Why Deep Learning?
Basics of a Deep Neural Network
Basics of RNN and LSTM
Tutorial on building an RNN-LSTM model in Python
Overview of hybrid deep neural architectures
How to debug and optimize an ML algorithm
Moderate knowledge of Python (Should be able to install and run common ML and DL libraries)
Jupyter Notebook (codes will be provided with explanations in a Jupyter Notebook (.ipynb) file and python file (.py))
Installations: TensorFlow (TensorFlow-CPU/GPU anything will be fine) (GPU is not mandatory), Keras, and other basic python libraries (scikit learn, numpy, matplotlib, etc.)
Basics of applied machine learning and data processing
Basics of Linear Algebra and Statistics
Basics of Multivariable calculus: partial derivatives, gradients, chain rule
My research concentrates on understanding and modeling human behavior in the healthcare domain. I believe that there is an opening to push the boundaries of computational modeling in Human-Computer Interaction (HCI) by modeling behaviors of people based on the data collected from people’s mobile, smart wearable devices, and their instrumented environments. I envision modeling human behavior, which can acquaint the design of prospective user interfaces that reason about and act in response to those behaviors to help people improve the quality of their life and wellbeing.
I have finished my B.Sc. in Electrical and Electronic Engineering (EEE) from the University of Dhaka. Before joining the University of Michigan as a CSE Ph.D. student, I have worked as a visiting researcher in Yagi Lab, Osaka University, Dept. of ISIR in the field of computer vision-based medical behavior monitoring. I also worked as an exchange student in Computer Science and Intelligent Systems department, Osaka Prefecture University under Japan–Asia Youth Exchange Program in Science (SAKURA) in the field of student behavior monitoring using eye trackers and wearable sensors.