How Scientists Use Machine Learning To Track Health Trends On Twitter?

Yes, you read it right.  Just a decade ago, most of us didn’t know what twitter is and today, it is helping the researchers and data analysts change the world. Ironic, isn’t it?

Some scientists from the University of Alberta developed a machine learning tool that can traverse through millions of tweets from people living in Alberta to understand the wellness and health trends across Canada.

One of the scientists from the team, Osmar Zaiane, explained that they used a machine learning algorithm to find out the location a particular tweet is referring to and then assess health particulars and emotions expressed in the tweet. He believes that if this tool is executed properly, it is possible to predict and understand a place, city, or state. It can help scientists evaluate what living is like in a certain area in terms of wellness and health.

The Machine Learning Tool: Grebe

Scientists call this tool Grebe, which uses the power of machine learning to help the healthcare centers of Canada take control of health conditions. It uses various other data sources such as doctors, hospitals, etc. to extract data, analyze it, and assist in the prevention of health and wellness issues.

It is believed that health experts want to know the trends of wellness and health in a particular city or state. This is so because it allows healthcare stakeholders to control a destructive situation sooner than later.

While surveys allow gathering of useful information, reports generated by humans can be often inaccurate and unreliable. This tool aids scientists to assess the situation with concrete information analysis.

Aspects Of Grebe

The machine learning tool identifies health and wellness trends based on six aspects: emotional, physical, spiritual, occupational, social, and intellectual. The machine also considers the emotions attached to the tweet and the location.

The tool further empowers scientists to go to another route with the trends found through different mediums such as surveys.

Conclusion

Imagine the potential of this machine learning tool. It can help healthcare stakeholders assess the health trends in various cities and take appropriate actions accordingly, if necessary. Similarly, it is possible to analyze what it is like to live in a city or state as far as wellness and health conditions are concerned. All these factors can change the way we travel, how we move job locations, or simply how we live in a city.