Day 1 - 28 November 2018
Data Analytics for AI and IoT: Chair’s Welcome and Opening Remarks
Solo: Edge Processing for Data Analytics and Training AI Algorithms
- How the huge influx of data will require fit-for-purpose architecture. What is the distance from the edge to your device and how to consider this during the creation of your IOT / AI architecture?
- Discussing how IoT / AI architectures need to be put in place to ensure increased compatibility across domains.
- Using cloud analytics platforms to derive value from IoT / AI data vs physical gateways -pros and cons.
Panel: IoT and AI Data analytics for intelligent decision making
- Identifying target-rich, high-value data that can be used to generate business intelligence
- Using cloud analytics platforms to derive value from IoT data
- Discussing the barriers to widespread IoT/ AI /Big Data value delivery and how these might be overcome.
- Real time data analytics in practice – examples of how IoT / AI data is creating business efficiency and revolutionising working practices
Principal Solutions Architect
10:30AM - Day 1
Keynote: Leading the way into the new era of IoT analytics with AI
Businesses today are looking to leverage all types of data to promote a data-driven decision making culture for their customers as well as own organizations. Specifically in the Internet of Things (IoT) domain, the amount of data being generated from sensors, devices, equipment, and infrastructure is on a very rapid incline. As a result, there is a tremendous need for the use of analytics algorithms and methodologies along with embedded AI to tackle, understand, and process IoT data to derive meaningful business insights. This session will focus on key aspects of AI as it pertains to IoT and a few customer stories across these domains.
11:40AM - Day 1
Industrial Machine Intelligence: The Golden Braid of Data Streams, AI, and Human Expertise
We are now more than a decade into the commercialization of “big data” and “data science,” but these technologies have yet to meet the needs of the businesses whose work exists outside the data center. The commodity stack of big data technologies are fundamentally flawed for use in the rising tide of data streaming from connected machines in industrial settings. There are many reasons for this, as challenges abound when embedding machine intelligence into a production industrial lifecycle. Perhaps the most challenging, however, is understanding how intelligent software systems will support normal business operations and what real benefits they will provide. In this talk I will present the concept of the “golden braid” of industrial machine intelligence: blending massive data, advanced machine intelligence, and human expertise. This approach enables both the human experts and the algorithms to leverage their comparative strengths. To support this, I will provide a case study demonstrating how I have done this in practice.
Raising the Bar – How Can IoT and AI Improve Our Performance and Fitness?
- Many fitness companies are borrowing techniques from sports teams and integrating them into consumer IoT/AI technologies.
- What are the issues surrounding data collection from these types of devices – who has access and ownership to this type of sensitive health data?
- How can predictive analytics can play a role in improving sports performance in personal and a team setting?
- Exploring innovations in smart textiles, sensors and platforms and examining which areas will provide future growth
Emerging Use Cases for IoT Data Analytics
Discussing new use cases for the data produced by IoT hardware, from video analytics to customer product usage data that can aid marketing.
What is the state of Hadoop today?
Hadoop has been one of the most important big data tools over several years – but if you look at the most recent report from Wikibon, vendors are not mentioning Hadoop as much as previously.
So why is this? The big cloud vendors have arguably been cannibalising Hadoop with their own storage layers while for many, object storage has become the de facto method of crunching big data. This session will explore the benefits and challenges facing Hadoop implementation, as well as trends in big data platforms, from object storage to stream computing
Chief Product Officer
AI Data Innovations
02:50PM - Day 1
Panel: The (Big) Data challenge
- How to collect human generated or human enhanced data
- Quality versus quantity
- The labelling challenge
NoSQL for big data analytics: Best practice and use cases
- NoSQL vs Hadoop vs SQL
- Enterprise implementations and use cases
- Advantages of horizontal vs vertical scalability
- Ensuring greater performance with larger data sets
Case study: How to get the most out of Apache Spark
- Moving from testing and proof-of-concept through to production applications
- The industries set to be impacted – financial, manufacturing, pharmaceutical
- Flexibility and adaptability in workloads
Head of Artificial Intelligence
04:40PM - Day 1