45
Internet of Things and Artificial Intelligence
3.3.1.3 Device Management
In an IoT scenario, even though a large number of devices are interconnected (not all)
among each other with servers for a wide region and can share the data effectively, still
managing a device not connected to network but somehow involved in data communication process may pose some kind of data linking issues.
3.3.1.4 Device Diversity
As many companies develop their products in a different way with different standards,
making such devices in IoT to perform together is really a challenge.
3.3.1.5 Flexibility
An IoT scenario shall be developed in such a way that the new device and technology
improvements may suitably be taken care without much hindrance.
3.4 IoT and AI
Machine learning as a part of AI along with IoT finds lots of application in both research
and industry (Poniszewska-Maranda and Kaczmarek, 2015).
3.4.1 IoT Challenges and Capabilities
As IoT is gaining everybody’s attention, there is an emergent need to understand the role
of AI methods to gain insights into the market scenario and the readiness of the competitors to address the situation by then. In an article published by Harvard Business Review,
it has been well said that IoT should be capable of addressing the following four basic
issues such as monitoring, control, optimization, and autonomy for making it more sensible for the use of the customer in a smart connected environment. While monitoring
is needed for the effective operation of sensor nodes in the working environment with
utmost control, optimization is needed for improving the performance based on the feedback received from first two steps. Finally, autonomy makes the IoT work independently
with self- diagnosis and repair.
3.4.2 IoT Won’t Work without AI
As IoT produces big data where city traffic data can be used to predict the accidents and
crime, it helps in building smart homes by digitally connected household appliances
and much more. The information extraction from the sheer volume of data that are being
collected from such IoT scenarios is a real challenge in order to see that IoT meets our
expectations. It is also envisaged that dealing with such a huge amount of data (even if a
sample of it) with traditional methods is too much of a time-consuming process. Hence,
one needs to use AI methods incorporated to IoT data for improving the speed and accuracy. The negative consequences like, in home applications, all connected devices not
working together will certainly annoy the customer; similarly, traffic can be mishandled
Internet of Things and Artificial Intelligence
3.3.1.3 Device Management
In an IoT scenario, even though a large number of devices are interconnected (not all)
among each other with servers for a wide region and can share the data effectively, still
managing a device not connected to network but somehow involved in data communication process may pose some kind of data linking issues.
3.3.1.4 Device Diversity
As many companies develop their products in a different way with different standards,
making such devices in IoT to perform together is really a challenge.
3.3.1.5 Flexibility
An IoT scenario shall be developed in such a way that the new device and technology
improvements may suitably be taken care without much hindrance.
3.4 IoT and AI
Machine learning as a part of AI along with IoT finds lots of application in both research
and industry (Poniszewska-Maranda and Kaczmarek, 2015).
3.4.1 IoT Challenges and Capabilities
As IoT is gaining everybody’s attention, there is an emergent need to understand the role
of AI methods to gain insights into the market scenario and the readiness of the competitors to address the situation by then. In an article published by Harvard Business Review,
it has been well said that IoT should be capable of addressing the following four basic
issues such as monitoring, control, optimization, and autonomy for making it more sensible for the use of the customer in a smart connected environment. While monitoring
is needed for the effective operation of sensor nodes in the working environment with
utmost control, optimization is needed for improving the performance based on the feedback received from first two steps. Finally, autonomy makes the IoT work independently
with self- diagnosis and repair.
3.4.2 IoT Won’t Work without AI
As IoT produces big data where city traffic data can be used to predict the accidents and
crime, it helps in building smart homes by digitally connected household appliances
and much more. The information extraction from the sheer volume of data that are being
collected from such IoT scenarios is a real challenge in order to see that IoT meets our
expectations. It is also envisaged that dealing with such a huge amount of data (even if a
sample of it) with traditional methods is too much of a time-consuming process. Hence,
one needs to use AI methods incorporated to IoT data for improving the speed and accuracy. The negative consequences like, in home applications, all connected devices not
working together will certainly annoy the customer; similarly, traffic can be mishandled