Social impact using data science
SOCIAL DATA SCIENCE
Soda's mission is to solve social challenges using data science in collaboration with domain experts. Using a combination of data analytics, machine learning and data visualization, we bring your social mission a step forward.
Soda was founded from a passion for enabling health, happiness and sustainability using smart technologies. Many opportunities are yet to be explored in terms of improving society with data. Too few decisions are based on a careful analysis of facts. There is often an abundance of data, but extracting valuable insights from these data turns out to be challenging.
Together with your domain knowledge and our data science expertise, we achieve the next step in accomplishing your mission. We discover patterns and trends, predict future developments and make data intuitive and appealing. Personal contact is important at Soda. We happily come by to brainstorm about the possibilities. Contact us at email@example.com.
PROOF OF CONCEPT
"What do you need to hunt, capture and replace misconceptions? Data. You have to show the data and describe the reality behind it."
Hans Rosling; Factfulness
There is a lot of data available, but how do you use it in the benefit of your mission... Familiar question? To investigate the extent to which the data is adequate for answering your question, we work in a rapid, flexible and accessible way towards a first solution.
This is what we call the Proof of Concept. We first brainstorm about the question, making it as compact and valuable as possible. Then, we make a selection of the data and develop an appropriate solution.
Upon completion we discuss the results and the way forward. From there we can continue developing in an iterative matter, making sure that output matches needs and expectations.
The main goal of data science is converting large amounts of data (also called big data) to useful information. Applying the correct visualization techniques makes information tangible, and serves as a solid foundation for decision making. Adding interactivity allows the user to play around with the data, resulting in a better understanding of what is really going on. Interactivity makes every data solution more appealing and more useful in (1) your daily work, (2) communication with your team and (3) communication with the outside world. Ideally, we combine all valuable insights into one comprehensive overview. In our demo, we for example interactively visualize geographical and temporal trends in life expectancy.
The visualization below summarizes 6,600 numbers into one overview. You see 36 years of data from 200 countries. Data concerning population, income and life expectancy become clear for everyone. Pres 'Play' to start the animation.
MACHINE LEARNING & ANALYTICS
Valuable information in data is not always directly apparent. To extract this information we analyse the data using smart self-learning Machine Learning, or Artificial Intelligence (AI) algorithms
Analyzing patterns and relationships between data points enables prediction of results and future trends, and creating groups within data sets.
Where Machine Learning is also valuable on structured data sets such as CSV files, an example of application on unstructured data (video) is shown on the right. Deep learning techniques (subgroup of Machine Learning) allows us to automatize complex and labor intensive visual tasks. In this case, we have trained a Deep Learning Neural Network to quantify plastic debris in rivers.