Back to Projects

Team Name:

Goldwyn


Team Members:


Evidence of Work

InsightsAI

Project Info

Goldwyn thumbnail

Team Name


Goldwyn


Team Members


Stan , Mangesh Gopale , Kang , Jasmine Guo , Timothy(Ziyang) Xu

Project Description


Insights AI

Governments and policymakers often face challenges in understanding the evolving needs of their communities, leading to inefficient resource allocation, delayed infrastructure upgrades, and slow responses to local developments.

This AI-powered platform addresses these issues by integrating multi-dimensional data, including historical population growth, current geo amenities (such as police stations, hospitals, and schools), and future forecasts for the next 10 to 20 years.

By analyzing these data points, the platform provides actionable insights and recommendations on critical infrastructure upgrades ranging from roads , agecare facilities, parks to educational facilities like schools and universities, healthcare services such as hospitals and clinics, and other essential amenities like petrol stations and train stations. It identifies optimal locations and timelines for these developments, ensuring they meet the future demands of the growing community effectively and efficiently.

This Platform also provides APIs for Real time Livability Index per location and Also Forecasts Livability Index and Corresponding SubIndices for User Selected Duration.

Problems We identified

  • Accurately predicting population growth using traditional methods that rely solely on birth and death rates proves to be increasingly challenging. For governments to allocate resources effectively and enhance overall quality of life, a more holistic and data-driven approach is crucial.
  • Governments and policymakers often struggle to fully understand the changing needs of their communities, which can result in inefficient resource distribution, delayed infrastructure improvements, and slow responses to local developments.
  • Without accurate insights and reliable forecasts, authorities are unable to identify areas that require attention within their jurisdictions, further complicating timely interventions.
  • Additionally, the general public lacks access to real-time and forecasted information about livability factors in areas where they wish to live or work, making informed decisions difficult.

Addressing these gaps is crucial to building more resilient, thriving communities.

How Insights AI solves these problems

Overall SubSytem View

Systems View

Language Model

For the language model, we have chosen RAG (Retrieval-Augmented Generation), which combines retrieval of relevant information from a database with a generative model, such as GPT from openAI or T5 from Google, to produce more accurate and contextually relevant responses.

Feature Selection Library

Feature Engineering creates and selects relevant data features, in our case the government datasets, to improve a model's performance. It uses domain knowledge to enhance accuracy.

Prediction Engine

The prediction engine utilizes our feature engineering library to analyse historical and current data, allowing it to make more acurate predictions about future trends. Supervised Learning is a viable candidate for the training of this prediction engine, and Continuous Learning, which involves updating the model incrementally as new data becomes available.

Example Applications

CommunityInsights

CommunityAI

AtlasAI

AtlasAPP

Livability Index Feed to Real Estate Apps.

Insights AI APIs

InsightsAPI

Livability Index

GET /api/v1/liveability

  • Query Params
    • lga (required)
    • year (required)
    • categories (optional)
    • unit (optional)
  • Response { "lga": "Whitehorse", "year": 2023, "index": 8.7, "categories": { "healthcare": 8.9, "education": 7.5, "infrastructure": 9.0 } } ### Population Forecast GET /api/v1/population-forecast

### Insights Api
GET /api/v1/insights/{domain}
* domain = healthcare, education, infrastructure, environment, housing

GenAi Prompt Api

POST /api/v1/generate-response

Benefits

Video Demonstration


#livability

Data Story


We utilize the following data sources to identify factors influencing the livability index:

ABS Data:
- ABS Region 1
- ABS Region 2
- ABS Region 3
- ABS Region 4
- ABS Region 5
- ABS Region 6

Additional Sources:
- Crime Statistics
- LGA Population Projections
- Australia's Transport Network
- State Emergency Services Facilities
- Police Stations
- Railway Stations
- Livability Index

By feeding this data into our LLM, InsightsAI forecasts population growth and livability index trends with new input.

Key Insights and Visualizations:

  1. Exploring Population Growth Indicators:

    We explored features beyond birth and death rates to better understand population growth.

    Population Growth Indicators

  2. LGA Population Growth (2017-2019):

    A visualization of the LGAs with the highest growth during this period to identify patterns.

    LGA Population Growth

  3. Unemployment Rates (Top 3 vs. Bottom 3 LGAs):

    Analyzing unemployment rates in top and bottom LGAs to assess any correlations.

    Unemployment Rates

  4. Crime Rates (Top 3 vs. Bottom 3 LGAs):

    Examining crime rates in these areas to identify any patterns.

    Crime Rates

  5. Findings and Correlations:

    Based on our analysis, we identified which features most strongly correlate with population growth.

    Findings

  6. Project Ideation:

    After exploring the data, we concluded that the livability index, driven by quality of life metrics, should be the core of our AI solution.

    Project Direction

    Livability Insights


Evidence of Work

Video

Homepage

Team DataSets

HouseHolds

Data Set

LGA Population HouseHold

Data Set

Victorian Population Data

Data Set

Region summary: Nillumbik

Description of Use Death/Birth Rate

Data Set

Region summary: Greater Geelong

Description of Use Death/Birth Rate

Data Set

Region summary: Hume

Description of Use Death/Birth Rate

Data Set

Region summary: South Gippsland

Description of Use Death Rate/Birth rate

Data Set

Region summary: Murrindindi

Description of Use Get the birth/death rate

Data Set

Population in VIC

Description of Use We use this dataset to indicate the community growth

Data Set

Crime Data in VIC

Description of Use Analyze the relationship between crime rate and population growth in each suburb

Data Set

Challenge Entries

Forecasting Community Evolution: Leveraging AI and Historical Planning Data

How might we predict future changes in community dynamics, such as population density, housing demand, traffic patterns, and the demand for public services or amenities?

Go to Challenge | 13 teams have entered this challenge.

Place-based insights unlocked: Generative AI x Digital Atlas of Australia

How can we use the Digital Atlas of Australia’s API and Generative AI to create innovative, user-friendly tools and visualisations that make geospatial data accessible to everyone, empower decision-making, and help all Australians better engage with their local and national environments?

Go to Challenge | 15 teams have entered this challenge.

Connecting Communities: Your Guide to local Information and Services

How can we assist local residents and visitors in easily locating local services and providing answers to frequently asked questions?

Go to Challenge | 29 teams have entered this challenge.