With the connected devices emerging in every Domain of Industry day by day. we are entering into the smart world where the ways of doing things are being changed,with sensors collecting the huge data set to measure the physical world and taking Analytical step to take intelligent decision is becoming the new business model.
To be a part of IoT ecosystem ,connected devices enables the system to sense the environment and cloud based analytic service make the smart decision. Here is the time now With Coursera, ebooks, Stack Overflow, and GitHub — all free and open — how can you afford not to take advantage of an open source education?
The dimensions (core themes) are
- Internet of Things
- Data Science
- Architecture ex Hadoop, Azure etc
- Algorithms/ Maths
- Programming (Python, SQL and Deep learning)
- Verticals(in our case IoT) but also spanning Telecoms, Supply Chain, Transport, Retail as applications of IoT
Image Credit : Dataaspirant
The Approach should cover Data science with an emphasis on IoT. Although it is recommended to cover the Mathematics from first principles, a basic understanding of Math and an aptitude for Math is useful. For the strategic version, you would cover the same ideas but would not be expected to write code.
Programming scope includes:
- Python:The basic Python ecosystem – from Data wrangling to Visualization.
- Python Time series:Python/ Pandas based code.
- Python scikit-learn:The Machine learning libraries for Python for prediction
- Real time: Spark and Storm
- Microsoft Azure:
- Hadoop, SQL and Visualization:Based on platforms like Cloudera impala and Tableau. Real time, distributed SQL on Hadoop.
- Sensor fusion (Complex event processing)
- Deep learning
Typical Data science algorithms with an emphasis on IoT datasets. Because it is recommended to follow a context based approach based on IoT, co-relate the maths to specific examples. Algorithms are encapsulated in libraries and APIs such as scikit-learn(for Python)
You will need an aptitude for maths. However, the mathematical foundations are necessary. These include: Linear Algebra including Matrix algebra, Bayesian Statistics, Optimization techniques (such as Gradient descent) etc.
In 2017, IoT is emerging but the impact is yet to be felt over the next five years. Today, we see IoT driven by Bluetooth 4.0 including iBeacons. Over the next five years, we will see IoT connectivity driven by the wide area network (with the deployment of 5G 2020 and beyond). We will also see entirely new forms of connectivity (ex from companies like Sigfox).
Enterprises (Renewables, Telematics, Transport, Manufacturing, Energy, Utilities etc) will be the key drivers for IoT. On the consumer side, Retail and wearables will play a part.
This tsunami of data will lead to an exponential demand for analytics since analytics is the key business model behind the data deluge. Most of this data will be Time series data but will also include other types of data. For example, our emphasis on IoT also includes Deep Learning since we treat video and images as sensors. IoT will lead to a Re-imagining of everyday objects.
Image Credit : Dataaspirant
Unique elements of IoT:
This should include: A greater emphasis on time series data, Edge computing, Real-time processing, Cognitive computing, In memory processing, Deep learning, Geospatial analysis for IoT, Managing massive geographic scale(ex for Smart cities), Telecoms datasets, Strategies for integration with hardware and Sensor fusion (Complex event processing). Note that we can also include video and images as sensors through cameras (hence the study of Deep learning)
The emphasis on IoT allows us to take a Problem solving approach through Use Cases. Taking a Use case approach allows us to identify as many complex use cases in each vertical (Wearables, Healthcare, Manufacturing, Retail, Supply chain, Smart energy, Smart cities, Smart Home) and drill down from there. We thus take an ‘engineering led approach’ – i.e. start with a problem and work back to understand the implementation. In contrast,Start with specific Algorithms. This has two disadvantages: · Firstly, Machine learning algorithms are often unfamiliar and hard for most participants. Secondly, most people are more familiar with Programming than Maths – and the platform often uses libraries which encapsulate the Algorithms. Thus, by taking a problem solving approach and working backwards – you are empowered to work across many verticals and technologies. This approach is similar to the ‘Context based learning’ in education
The Use case approach spans the understanding. We tie specific concepts to IoT use cases where possible. This includes the mathematical concepts. Each Use case starts with an IoT related problem. There may be more than one use case per concept. Themes for Use cases include (but not confined to): Complex event processing, Real time, Deep learning, Time series, Edge processing, Use cases by IoT verticals (ex healthcare), Spatial processing.
The IoT ecosystem, Unique considerations for the IoT ecosystem – Addressing IoT problems in Data science (time series data, enterprise IoT edge computing, real-time processing, cognitive computing, image processing, introduction to deep learning algorithms, geospatial analysis for IoT/managing massive geographic scale, strategies for integration with hardware, sensor fusion)
Here we learn Python for Data analysis packages including Numpy, Scipy, Matplotlib, Pandas, Scikit-learn and Bokeh plots. We use the iPython notebook, the Anaconda distribution, sci-kit learn, Pandas and Wakari.
An exploration of IoT datasets and APIs by application: Healthcare, Manufacturing, wearables, Energy etc
Developing for IoT devices (Unique considerations)
Supervised algorithms, unsupervised algorithms (classification, regression, clustering, dimensionality reduction etc) as applicable to IoT datasets
The person should emphasises the following unique elements for IoT
- Complex event processing
- Deep Learning and
- Real Time / Time series datasets includes architectures like Spark and Storm
Mathematics and Statistics
- Descriptive Statistics
- Inferential Statistics
- Linear Algebra’
- Structural Thinking
Process involved while working on Data Set
- Study Outliers
- Check Distribution
- Group similar valuesa
- Look for co-relation and missing data
- Try Clustering and classification algorithm
- Build data pipeline on cloud
- NLP library on unstructured text data
Want to know more about Data Science Courses for free visit Coursera
Featured Image Credit : Dataaspirant
Stay Blessed. Work Hard