My Canberra
How do I learn about, and connect with, my city, my suburb or my neighbourhood?
Go to Challenge | 13 teams have entered this challenge.
DIISplay
Trendhound lets users search a keyword of interest across Twitter, Google News and public datasets.
The dashboard displays relevant tweets across the country and relevant news articles.
It highlights facts at the states and territory level from relevant datasets. It allows users to quickly see trends, and access data for more detailed analysis (if desired).
Trendhound will benefit a variety of users, including policy makers, government service providers and ordinary citizens. Policy makers could use Trendhound to better understand community sentiment, and easily access related data to make better decisions for the community.
Government service providers could use results to analyse short and long term trends, or use community feedback on Twitter to predict sources of community stress or public health issues. This can influence decision-making in real-time, for example to make staffing allocation decisions at hospitals. This will allow service providers to efficiently allocate scarce healthcare resources. Citizens can use Trendhound to see the healthcare trends and issues in their communities and how they compare to the rest of the country.
In this proof of concept phase, data is at the state and territory level, focusing on healthcare issues. This could be expanded to more issues, data sets, and geographical levels.
Example use cases:
User Story 1
Rachel is the senior administrator of a major hospital in Canberra, responsible for, among other things, staffing allocations. A quick search of the terms “cold” and “flu” on Trendhound can tell Rachel if there is a spike of cold and flu cases in the areas surrounding the hospital.
This can help Rachel to plan for staffing and other resource allocations in the hospital to ensure that resources are efficiently allocated and service delivery is not impacted by sudden changes in demand.
Trendhound could be expanded to search for emergencies and other infectious disease outbreaks with similar benefits. Rachel could also search Trendhound for cancer data and trends, making it useful for longer term planning and decisions.
User Story 2
Monica is a senior advisor on health policy with the ACT Government. On Trendhound, Monica can search for a number of health related keywords and gain access to a range of trends and information from sources such as Twitter, Google News and public datasets, all in the one convenient place.
Monica can use the result to better understand trends and sentiments in different communities across Canberra, and easily access related data, for example community feedback through Twitter to predict sources of community stress or public health issues. Monica can use the information to make better, more targeted and efficient policy decisions that will positively affect Canberra’s communities and citizens.
Check out the project on the homepage URL to explore Trendhound and see user cases.
You can check out more user cases here:
https://drive.google.com/open?id=1mlh_-uk4FavEQQb6iXEzrLor8s5i98L1-tpjWrVN5Vw
Key Risks
Trendhound can search and return results for tweets from private individuals, so there is potential for text that includes explicit material, incorrect information or other unwanted results.
Information from Twitter and Google news could be wrong and lead to poor decisions being made.
Searches may not return the results that are useful to users or not provide enough information to make a sufficiently informed decision.
Users may misinterpret the data presented, leading to poor decisions and outcomes.
Measuring the impact of Trendhound is challenging, as it is not known how many searches lead to high-value or high-impact decisions.
We used data from the AWHI, ABS, ACTgovdata and data.gov.au with Twitter and Google News Services to create a dynamic interface showing key information and datasets on an area of interest for the user.
Input search term(s) (hayfever, cancer, obesity, drugs, smoking and alcohol) to start searching.
Trendhound:
Simultaneously searches Twitter, google news and relevant datasets.
Returns relevant results from the above sources for the search term(s).
Gives breakdown of percentages for search terms.
Links search terms to relevant datasets.
Our data tells the story of the regions and people of Australia. It tells people about the issues that impact them and their neighbours, and offers the location of local services to become more engaged. It helps people find the most relevant tweets, news and datasets, so they can understand the context of their query. Our data tells a story about how communities live their lives and how they can use public facilities to make them better. It is about learning about where you are or where you'd rather be!
Description of Use We used Data.gov.au as a dataset to provide users with links to the most relevant, available data when they make a key word search. Users would be able to access the most relevant data.gov.au datasets to help inform their decisions and provide the most accurate and reliable information. By searching all available datasets, a user can choose to derive their own results, build a great understanding of the context their query, and become more informed citizens. Incorporating a search of data.gov.au into our platform will also help to build the public's awareness of the datasets that are available and nudge citizens to consider how they can use it more in the future.
Description of Use We used this dataset to compare use of drugs, alcohol and smoking across the Australian states and territories compared to the national average. We used this dataset as it provided an understanding of a range of leading indicators to long term health issues, with the capability to compare different regions. It will help our users connect the data with their own experiences in their state, driving awareness of local issues and potential behavioural change to unexpected information. For example, the ACT has a significantly lower rate of daily alcohol consumption than the national average. If public policy makers can understand what drives this behaviour they may be able to use this information to nudge more at-risk jurisdictions to reduce their consumption of particular drugs.
Description of Use As a proof of concept, we used this dataset to provide a snapshot of "Services Near Me" for a user in the ACT. A user would be able to see all the locations of hospitals near them to inform the user of services available to them. In a full-scale product the map would adjust depending on the location of the user, creating a personalised experience and providing real-time information to support users that may be in crisis or require medical attention. It can help users identify the services available closest to them that they may not have been aware of.
Description of Use As a proof of concept, we used this dataset to provide a snapshot of "Services Near Me" for a user in the ACT. A user would be able to see all the locations of public basketball courts near them to nudge them to explore the facilities around them. In a full-scale product the map would adjust depending on the location of the user and would encourage them to use these facilities. Users could also engage their friends ad family to use the basketball courts as a group, providing broader community health benefits.
Description of Use We used this dataset to compare the rate of overweight and obese people across Australian states and territories compared to the national average. We used this dataset as it provided the most comprehensive view of this issues across Australia. For the user, it can help them understand how their demographics are contributing to their region’s overall health. It can move people to action, to not only get out and be active themselves but to motivate their friends and family to live healthy lifestyles. With sufficient support from government, this data can also build a level of healthy competition between different states to lower their rate of overweight and obese people.
Description of Use We used this dataset to compare the prevalence of hay fever across the Australian states and territories compared to the national average. We used this dataset as it provides information on a commonly underestimated illness in our community. With spring approaching, we thought it provided some interesting insights, including the particularly high prevalence of hay fever in Canberra, the home of the national flower show Floriade. Sufferers or carers can use this data to compare different regions and help them understand that their experiences may be more common than they think. It can help promote better hay fever management as sufferers in areas of high-incidence can prepare ahead of the spring season.
Description of Use As a proof of concept, we used this dataset to provide a snapshot of "Services Near Me" for a user in the ACT. A user would be able to see all the locations of public fitness equipment near them to nudge them to explore the facilities around them. In a full-scale product the map would adjust depending on the location of the user, creating a personalised experience and providing real-time information to encourage users to act on the information they receive from the dashboard.
Description of Use We used this dataset to compare the prevalence of different cancers across Australian states and territories compared to the national average. We used this dataset as it provided some interesting and compelling information, including the relatively low prevalence of cancer in the Northern Territory which could indicate a lack of diagnosis services or individuals travelling interstate to access the necessary services. It can move them to action, to share this information with their friends and family, or support cancer research in their community.
Go to Challenge | 13 teams have entered this challenge.
Go to Challenge | 34 teams have entered this challenge.
Go to Challenge | 35 teams have entered this challenge.
Go to Challenge | 9 teams have entered this challenge.
Go to Challenge | 34 teams have entered this challenge.
Go to Challenge | 8 teams have entered this challenge.
Go to Challenge | 7 teams have entered this challenge.
Go to Challenge | 5 teams have entered this challenge.
Go to Challenge | 58 teams have entered this challenge.
Go to Challenge | 49 teams have entered this challenge.