Back to Projects

Team Name:

KaVa


Team Members:


Evidence of Work

Weave

Project Info

Team Name


KaVa


Team Members


Vanessa S , Audi Firdauz , Kazu and 1 other member with an unpublished profile.

Project Description


Often, citizens feel more comfortable and connected when they can communicate in their own language, fostering familiarity and stronger engagement. To address this, we propose Weave, a multilingual community forum application designed to strengthen civic participation by breaking down language barriers, fostering inclusive dialogue, and transforming online conversations into offline community action.

Each state and territory has its own localized version such as WeaveSA, WeaveNT, WeaveVIC, and so on. Ensuring that discussions are tailored to local issues and communities. The name Weave reflects both the idea of “we have” and “belonging,” symbolizing inclusivity and the interconnection of diverse voices in Australian society.

Through AI-powered translation, users can post in their own language while others read and respond in theirs, creating a truly multilingual space for engagement. The platform integrates automated news summarisation of government updates and local event listings, ensuring that citizens and immigrants alike can access, understand, and participate in public life. Community engagement tools, such as upvoting, allow users to highlight important discussions, which can gain visibility with the Department of Home Affairs, bridging everyday voices with government responsiveness.

Beyond facilitating dialogue, Weave’s AI-driven agent analyzes discussions to identify topics that are gaining genuine traction in the community, not just in terms of popularity, but in their relevance to real-world challenges. By filtering and highlighting these issues, the AI ensures that trending discussions reflect matters of public significance rather than simple virality. Priority topics are then flagged to the Department of Home Affairs and other relevant agencies, where they can be assessed, discussed, and acted upon.

In addition, insights from community interactions can help the Department enhance its own AI agents, making them more knowledgeable about relevant rules, regulations, and procedures. This improves the government’s capacity to provide accurate, timely responses to trivial or frequently asked questions, creating a more useful and responsive digital interface for citizens.

To promote ongoing interaction, the app includes profile picture customization rewards. The more discussions users start or contribute to, the more options they unlock to personalize their profile, making participation both meaningful and fun.

By combining localized focus, inclusivity, and gamified engagement, Weave builds stronger connections within communities, fostering transparency, belonging, and active civic participation across Australia.


#language #multicultural #community #connecting-mind #diversity #ai-agent #data-in-one-language

Data Story


Evidence from multiple statistics from the Australian Bureau of Statistics suggests that outcomes related to discrimination, social participation, trust, and perceptions of government are consistently linked to cultural and linguistic diversity. These findings are particularly relevant to migrant and immigrant populations, who often face barriers to inclusion due to language proficiency and cultural differences.

Table 1 (provided in GitHub) demonstrates that a quarter of the population prefers to speak in their own language at home. This suggests that while many individuals may be comfortable using English in everyday contexts, they often feel more at ease and authentic when expressing their true feelings and opinions in their first language.

Table 2 and Table 3 shows a decline in cultural attendance, there’s an increase in participation in cultural activities, a possible cause is that people are more inclined to engage in events and activities organized by members of their own ethnic or cultural group. Such settings provide a sense of familiarity, comfort, and belonging that may not always be present in broader, mainstream cultural events.

Problems Identified

  • Language Barriers Effects & Barriers to Social Cohesion: People are often more inclined to join events organized by those from the same ethnicity or cultural background, as shared language and familiarity make participation feel easier. Fragmented Civic Dialogue: There is no simple, inclusive channel for citizens to start conversations on topics that matter to them and turn those discussions into real-world action.
  • Government Related Concerns: Weak Channels for Government Response: Citizens’ concerns rarely reach authorities in an organized way, limiting government accountability and responsiveness. Government data and announcements are often published in English only, scattered across multiple platforms, and presented in formats that are difficult for non-native speakers to navigate.
  • Limited Accessibility of Government Services: Government services cannot reach everyone at the same time, creating delays and gaps in access to information.
  • Hidden Community Issues: Many local challenges and concerns remain invisible because there is no accessible platform for people to surface and share them.

Data Integration Workflow

  • Ingestion Automated scripts fetch datasets via API or scheduled downloads.
  • Processing When data is cleaned, normalized, and stored in a modular backend (Dockerised services).
  • Enrichment AI translation and summarization applied to text-based datasets.
  • Delivery Forum modules display data in user’s preferred language, with event and news feeds updated in real time.
  • Feedback Loop Engagement metrics inform future data prioritization.

Compliance and Licensing

  • All datasets are sourced from open government portals under Creative Commons or equivalent open licences.
  • Attribution is provided in the forum’s “Data Sources” page.
  • No personal or sensitive data is stored or processed.

Anticipated Social Impact

  • Inclusion Language barriers reduced for CALD (Culturally and Linguistically Diverse) communities.
  • Engagement Increased participation in local events and civic discussions.
  • Transparency Easier access to official information in plain language.
  • Resilience Strengthened community networks through both online and offline interaction.

Evidence of Work

Video

Homepage

Project Image

Team DataSets

Cultural Diversity Census Dataset

Description of Use Useful for analysing demographic patterns, informing inclusive government policies, planning multicultural services, and supporting research into migration and cultural diversity.

Data Set

Census Dictionary: Language Used at Home (LANP)

Description of Use Used as a reference standard for analysing Census cultural diversity data, ensuring consistency in reporting of language statistics. Supports alignment between Census outputs, migration datasets, and research into language use in Australian households.

Data Set

Cultural Diversity Data Summary

Description of Use Useful for analysing demographic trends, informing inclusive government policies, planning multicultural services, and supporting academic and community research into cultural diversity.

Data Set

Challenge Entries

Community AI Agents: Bridging Service Access Gaps

How can we design Agentic AI solutions that autonomously assist residents in discovering, accessing, and engaging with local government and community services?

#Empower #Connect #Include

Eligibility: Open to all. Your solution will be measured against its relevance to the theme, practicality and scalability, whether it follows ethical and inclusive design, and its innovation. You must use at least one Government dataset.

Go to Challenge | 29 teams have entered this challenge.

Bridging Social Divides: Bringing People Together to Strengthen Social Connections

How can we bring people together from diverse backgrounds to communicate respectfully, even when they hold opposing views?

#Strengthening-social-connections-for-community

Eligibility: Open to all. At least one government dataset must be used.

Go to Challenge | 20 teams have entered this challenge.