Day :
- Data Analytics
Location: London ,UK
Session Introduction
Tareq S. El-Hasan
PhD degree, Electrical Engineering from University of Hertfordshire, UK.
Title: Internet of Thing (IoT) Based Remote Labs in Engineering
Biography:
Tareq S. El-Hasan received his PhD degree in Electrical Engineering from University of Hertfordshire in the UK in Jan 2003. He has worked for Defense Industry at King Abdullah II Design and Development Bureau (KAADB) in Jordan from March 2003 until March 2010 in which he has held several senior technical and managerial positions. He has joined the electrical engineering department at Zarqa University in Jordan since Sep 2010 until now. As an associate professor of electrical engineering he has held the position of electrical engineering department chair since Sep 2011 until now and has held the position of vice dean for college of engineering from Sep 2014 – Sep 2015. His research interests are in High-Speed PM Machines, Electrical Drives and control, Pulsed Power, IoT and renewable energy. Dr. Tareq has published several scientific papers in International Journals and Conference proceedings.
Abstract:
This paper presents the development of a laboratory system at the electrical engineering department of Zarqa University based on IoT and mobile application technologies to remotely run, implement and monitor the experiments that are required in the course of work by the student. Throughout a custom designed dashboard on the mobile application, the system, allows the students to extract results on a real time bases and submit their reports once they finish the experiments. The system comprises all the necessary devices, such as sensors, controlling and interfacing kits, cameras and others. Initially, the work is implemented on a Separately Excited DC (SEDC) motor experiment as one of the examples related to Electrical Machines Laboratory (EML) course as part of the courses delivered to the Electrical Engineering Program at Zarqa University. The proposed work mainly intents to provide an off-campus easy accessibility to the assigned experiment through a user-friendly interactive dashboard. For example in this experiment, the student can switch the direction of rotation of the motor and can run the motor at preset conditions with the capability to measure the electrical quantities such as voltage, current and power as well as other required parameters, such as speed and torque. The preliminary results demonstrated a proof of concept for remotely running such experiments via mobile applications, thereby allowing the students to perform their experiments in a fixable yet reliable manner. The work is still in progress and it is anticipated to expand the system to cover other experiments once the work is finalized.
Puntis Palazzalo
Sr. Data Scientist, SAP LLC., Palo Alto, CA 94304, USA
Title: Young brains and wisdom of the elders — Big Data Technology and SAP
Biography:
Puntis Palazzolo is a Sr. Data Scientist at SAP where she manages the SAP Big Data solution, SAP Data Hub, Developer-focused and Advanced Analytics topics in her role in SAP Data Hub Product Management team. She has more than a decade of experience in software design and development, machine learning systems and database technologies in different industries such as Bioinformatics, Military and Health Care and applications such as Handwriting and Voice Recognition, Image Processing, Natural Language Processing and Recommendation Engines.
Puntis has several research publications in the field of Machine Learning and Data Science and has patented ideas in the field of Recommendation Engines.
Her academic background is in Computer and Electrical Engineering, Computer Science and Software Engineering.
Abstract:
In the Age of Insights where we use IoT for smart parking and wearable technology for health monitoring, customers should be able to access the right information at the right time and that requires access and processing of Big Data. With more than 40 years in the enterprise software market and almost 20 years of experience developing in-memory computing solutions, at SAP we have experienced the below shortcomings in the area of Big Data Analytics:
Companies struggle to bring various types of Big Data such as structured together with the semi or unstructured data such as graph, time series, spatial, and text together and be able to run machine learning algorithms across them seamlessly.
Data scientists have their preferred language such as R or Python and want to use both in the same piece of code. They are looking for a platform which allows them to experiment and implement and is language-agnostic.
Development of production-level systems are more difficult than developing a proof of concept and running experiments. Implementing the orchestration, scheduling, monitoring, and automation is still a big challenge.
In the area of Big Data what matters is how to elastically extend or reduce the computation and storage power as needed without sabotaging the work done.
Customers need to work and access data in the cloud, on-premise or both with a little effort.
There are so many great open-source libraries out there. The users’ concern is how to benefit from them without being worried about the compatibility of the libraries and how to handle the versioning and support.
Our goal is to address some of these obstacles and share some success stories in the area of Big Data Analytics. We will unravel our journey toward our Big Data solution and walk you through the examples of our platform and invites you to rethink the traditional means.
Alan Rezazadeh
PhD, Southern Alberta Institute of Technology Calgary, Alberta, Canada
Title: A New Paradigm for Successful Implementation of Big Data Predictive Analytics and Forecasting
Biography:
Alan Rezazadeh has been working for more than two decades in Alberta, Canada as a data analytics and governance consultant within public, oil & gas, and education sectors. His recent big data engagements include IoT, business process improvements, predictive analytics and forecasting. Alan earned a PhD degree in Computer Science from University of Regina, Canada in 1995.
Abstract:
The main objective of this presentation is to discuss a new paradigm for successful implementation of big data analytics. Big data analytics may contain multiple sources of corporate, proprietary vendors, and public data portals (e.g. governments) to support complex business processes. Amalgamation of the required data for predictive analytics may present challenges for data transfers, integrations, updates and quality issues such as missing and inaccurate readings. Adding to this complexity managing outliers, due to natural disasters such as forest fires may cause abnormal measurements and readings.
Predictive analytics may require tens of years of business process consumable data comprised of complex modelling and integration prior to analysis. The availability of large amount of required data demands effective and efficient data models for enabling business operations utilizing near real-time predictive analytics. The ultimate goal is avoiding retrieval of any unnecessary data in order to achieve the desired performance.
This presentation discusses two big data analytics initiatives, integrating multiple public and corporate data sources supporting decision makers for planning and monitoring business operations. One of the lessons learned was utilizing data virtualization for an effective implementation of data integration layer within a big data analytics initiative. Data virtualization methodology provided significant improvements during extraction, transformation, analysis as well as run-time performance, by eliminating duplication and loading only relevant data. The authors will address the lessons learned for effective design of analytics output to be understandable and consumable by business users to support their operational requirements.
Ching Suen
Director, Centre for Pattern Recognition and Machine Intelligence, Concordia University in Montreal, Canada.
Title: Enormous Varieties of Font Data for the License Plates of Automobiles
Biography:
Prof. Ching Suen is currently the Director of the Centre for Pattern Recognition and Machine Intelligence at Concordia University in Montreal, Canada. He is the Co-Editor of a recent book entitled "Digital Fonts and Reading" published by World Scientific. He has presented papers at ICPR, ICDAR, ICFHR and Big Data conferences and published several books and many papers in journals like Visible Language, Information Display, Pattern Recognition, Handwriting Recognition, and Document Analysis, and Image Processing. He has supervised more than 220 graduate students and visiting scientists from all over the world. He has been the Editor-in-Chief of two international journals. Currently he is an editor of several journals related to pattern recognition and image processing. He has founded numerous international conferences such as ICDAR, ICFHR, Vision Interface, and has chaired many conferences including a new one: International Conference on Pattern Recognition and Artificial Intelligence, http://www.icprai2018.com in Montreal.
Abstract:
License plates contain the alphanumeric identity numbers of automobiles. These numbers provide pertinent information about the cars and their owners. They can be detected by radar or by canners to check for speeding and thefts, etc. They are also important to security because they can help the police to identify a vehicle. However, some alphanumeric characters may not be easily read when the cars are moving or at a distance. As we march towards the modern age of having driverless cars on the roads, it is imperative that the license plates of vehicles can be read correctly by the human eyes as much as by intelligent machines. With that goal in mind, we started our research by studying the fonts of the current license plates to determine their legibility and suitability. We began by examining the regulations governing the designs of these alphanumeric numbers as published in government documents. Since then, we have also conducted an extensive survey and collected big data on the formats and fonts used in the license
plates of many countries in North America, Europe, and Asia. We have found that quite a few of the alphanumeric characters are susceptible to errors leading to the wrong reading of their identity. Typical examples include confusions between numerals 5 and 6, between B and 8, G and 6, D and 0, I and 1, K and X, Q and 0, V and Y, and Z and 2. Having identified the
confusing characters, we designed a new font with appropriate openings of the counters, X-heights, slants, distinctive features, angles and lengths of certain strokes and overhangs. Our new font has been judged to be more legible than all the designs found in our survey. This presentation will include the following findings:
1) The big font data used in the license plates found in different countries on several continents,
2) The design of our new font aiming at high legibility,
3) A subjective experiment comparing the legibility of our new font with the current fonts found on current license plates in different regions around the globe, based on computer simulation of moving vehicles.