The Importance of Big Data in Environmental Monitoring

The Importance of Big Data in Environmental Monitoring

Before the term "data analytics" was coined, analytics were being used in the 1950's to give companies insights and trends based on their collected data. Big data means bigger opportunities for companies to grow by giving insights based on both future and immediate decisions. Reduction in cost, faster and better decision making, and innovative products are just some of the benefits that big data brings.

Ambience Data is on a mission to be the front leader in obtaining data and data analytics at a global scale in all environmental verticals. Our hardware devices (i.e. The BlueJay and the Starling) contain up-to-date sensor technology capable of monitoring pollutants like CO2, particulate matter, VOCs, and NOx in real-time. These devices can hook onto one another to create a network, capable of capturing multiple data points in different locations at the same time. Our third product, the Sparrow, serves as a communication hub between any standard legacy sensor and the cloud. 

From our devices, the data is sent to the cloud and onto our web-based dashboard. Our dashboard can display the information that the user requests along with insights that are created from combining different data points; this data is also available on our mobile app which features an alerting system if pollutant concentrations surpass threshold levels. 

Because of our devices, we have growing access to environmental pollution levels in major urban areas around the globe. Our interactive mobile app and online dashboard allows us to gather data from our users. Through IBM's Watson Analytics, we have access to global historical weather data and twitter feed data. Microsoft's Azure allows us access to important Bing Maps and Bing search engine data. Through Google, we have access to consumer location data and geospatial data. Additionally, we have access to public open source data such as government monitoring stations and traffic cam data.

Depending on our client, we use a variety of machine learning platforms including AzureML, Apache SparkML, and Google's TensorFlow. We primarily use supervised regression and cluster modelling to find relationships between various data sources, such as traffic data, weather data, pollution data, and hospital admission rate data, and predict future pollution levels, hospital admission rates, and other important information. Depending on the type of client, the types of models along with the input and output data vary. For outdoor pollution monitoring at a local level (for e.g.), our input data typically involves traffic volume, current air pollution levels, spatial coordinates, and weather data.

It is expected that the Worldwide Big Data and Business Analytics Market will grow to over $203 billion in 2020. We see value and growth potential in big data and data analytics, and hope to be at the forefront when providing valuable insights to our clients.



Pingbacks are closed.


Comments are closed.