not approved

Chatbot AI for Project Catalyst

$40,124.00 Requested
Ideascale logo View on ideascale
Community Review Results (1 reviewers)
Impact / Alignment
Feasibility
Auditability
Solución

Building an AI chatbot that could promptly answer most of the Project Catalyst-related questions when users ask.

Examples:

• "How many DAO proposals are there?"

• "When are proposal drafts due?"

Problem:

The information involved in Project Catalyst is continuously updated and disseminated through many different channels and media, making it hard for users to find the neat info when needed.

Yes Votes:
₳ 72,268,214
No Votes:
₳ 105,363,331
Votes Cast:
286

  • Video cover image

[IMPACT] Please describe your proposed solution.

The Project Catalyst is globally growing at an amazing speed. It helps fund those with excellent ideas to deliver on the Cardano blockchain through communities with a completed decentralized and consensus voting cycle. Having participated in Catalyst for several funds, we also recognized that there are still some existing problems.

Problems:

  • Information involved in Project Catalyst (e.g., proposals, proposers, groups, voting, and funding) is continuously changing; it’s hard to stay caught up.
  • Information is updated and disseminated through many different channels, which may overwhelm users. Referencing information from the past is also difficult. For example, can you find "the official voting results of fund 6“ (which ended years ago)?
  • It is very hard to find the right information without investing lots of time in the search.

Our solution: A Project Catalyst chatbot. The Chatbot is able to answer frequently asked questions and provide Project Catalyst-related information (such as proposals, proposers, groups, voting, and funding).

How the chatbot is implemented:

The chatbot will be iOS and Android apps.

The benefits of the chatbot:

  • Save time and effort: Users will be provided with the right answer immediately, saving them time and effort in the search. The chatbot is supported 24/7 at any time.
  • Reliable information: the chatbot will provide reliable information. Users do not need to look for sources of information (e.g., asking someone in a group chat that may not be reliable).
  • Provide guidelines: How can you find a proposal in the past that you remember just a little bit of its information? Through QA, the chatbot will guide you to find the right answer.

Who will benefit?

Anyone outside or within the Project Catalyst communities, such as newcomers, can look at frequently asked questions to learn more about Project Catalyst roles like Proposal Assessor (PA), veteran Proposal Assessor (vPA), Proposers, voters, SPO, NFT, etc.

Anyone who wishes to convey their idea to become a proposal, go through the voting cycle and deploy on the Cardano blockchain if approved for funding but is limited in information and unsure where to start. The Chatbot AI will propose a range of past proposals similar to the ideas or asked keywords that the potential proposer is exploring. This is the ultimate assistance from Chatbot AI, so the proposer can possibly get on with the proposal forming quicker.

[IMPACT] Please describe how your proposed solution will address the Challenge that you have submitted it in.

This proposal is aligned with this Challenge's KPIs as follows:

"Increasing the number of dapps and products available for the community to use that help to enrich the ecosystem with new use case"

Answer: The chatbot is a free mobile app that is available for the whole community to use. Even a new user can understand the Project Catalyst ecosystem thoroughly just by querying the chatbot. Surely, it will increase users' engagement with the Catalyst community, thus enriching the Project Catalyst ecosystem.

"Increase the number of integrations that bring existing solutions together for a more seamless and connected experience between different product"

Answer: Currently, information is disseminated through many different channels, which makes it hard for users to follow. The chatbot will bring these pieces of information together in one place where people can query the information they need right away.

"Increased quality of existing products & integrations through suggested improvements that is supported by customer feedback or increased usage by the community"

Answer: With the help of the chatbot, it is certain that it will increase the usage of the whole community.

[IMPACT] What are the main risks that could prevent you from delivering the project successfully and please explain how you will mitigate each risk?

Risk 1: The biggest challenge is to understand the users' intentions behind their questions.

Solution: we will implement the state-of-the-art natural language processing AI algorithm to train our chatbot.

Risk 2: The database of the chatbot needs to be updated frequently to catch up with Project Catalyst's latest information.

Solution: we plan to hire a person who is responsible for keeping the database updated. We are also thinking about a feature that would allow an administrator to update the database through the chatbot itself.

[FEASIBILITY] Please provide a detailed plan, including timeline and key milestones for delivering your proposal.

The chatbot's development is separated into 2 phases. At the end of each phase, the chatbot is able to

Phase 1:

  • Answer frequently asked questions (e.g., what is Project Catalyst? how is process funding on Project Catalyst? Schedule and voting results for each Fund? Provide proposals-related information such as proposals' content, proposers' names and groups, their CA assessment, funding results, etc. )
  • Users can request the chatbot to search for various resources (e.g. search for proposals by name, search by Project Catalyst username, etc).

Phase 2:

  • Chatbot is able to support state-aware functions (i.e., remember things users have said, link questions together)
  • Support for local languages other than English (e.g., Vietnamese, Japanese)
  • Support QA by voice

This proposal is for Phase 1 only. Phase 2 will be implemented after Phase 1 ends. The timeline for Phase 1 is as follows:

Total duration: 5 months

Month 1:

  • Building landing page
  • Design the main functions and UI of the chatbot.
  • Collect Project Catalyst-related data (e.g., frequently asked questions, useful links and resources).

Month 2:

  • Process the collected data.
  • Build chatbot databases.

Month 3:

  • Train AI models (i.e., a natural language processing model to handle users' questions)
  • Evaluate chatbot answers (internal testing and QA)

Month 4:

  • Build iOS and Android apps
  • Run a trial with a small group of users (around 30 users).

Month 5:

  • Revise based on feedback after the trial run
  • Marketing and advertising the app
  • Launch the official app

How the team will manage the workload?

Since we are also working on other proposals, the above plan is calculated based on the time, effort, and resources we can spend on this project. Moreover, our team has 2 more engineers who can support us to be able to deliver this project successfully.

[FEASIBILITY] Please provide a detailed budget breakdown.

(1) Development cost:

  • Create a landing page: 3 working days (24 hours)
  • Collect Project Catalyst-related data: 4 working days (32 hours)
  • Process collected data: 16 working days (128 hours)
  • Train and evaluate chatbot model: 20 working days ( 160 hours)
  • Build iOS and Android app: 20 working days ( 160 hours)
  • Marketing, and advertising the app: 5 working days (40 hours)
  • Test and collect evaluations from 30 users: 4 working days (32 hours)
  • Cost for team management and meetings: 40 hours (2 hours/week*4 weeks/month*5 months)
  • The cost to hire 1 person to update the database for 1 year: 104 hours.

Total working hours: 720 hours

The cost is 720* $50/h= $36000.

(2) Infrastructure and deployment costs (Total $3124)

  • AWS infrastructure for development and trial run: $3000
  • Deploy to AppStore and Playstore: $99 + $25 = $124

The cost is $3124.

(3) Contingencies: $1000

Total cost: $36000 + $3124 + $1000 = 40124$.

[FEASIBILITY] Please provide details of the people who will work on the project.

Cao Thinh:

Ph.D. holder in Computer Science. A senior developer with 7 years of experience working in the information technology and services industry. Skilled in data analysis, databases, networking, web and application development. Strong programmer with Python, R, C++, C#.

Linkedin: <https://www.linkedin.com/in/thinh-cao-93350b18>

Darlington Wleh:

I am a co-dreamer at Lidonation, a growing collective of Cardano enthusiasts who believe in providing insightful education about Cardano in plain English, Swahili, and Spanish. Most online blockchain content is written by and for developers and finance geeks. We break down complicated technical ideas for everyday people, and provide relevant news in quick bites, and build tools that everyone can use.

Our website is a resource for rich Catalyst data for anyone wanting to glean insights from all the happenings within the catalyst. We want to make these insights accessible to as many people as possible.

Darlington is an engineer by day and at all times a dancer, humanitarian, idealist, and entrepreneur. Darlington has deep knowledge about blockchain technology and broad experience in the Cardano ecosystem.

<https://www.linkedin.com/in/profd2004>

<https://www.linkedin.com/company/lido-nation>

[FEASIBILITY] If you are funded, will you return to Catalyst in a later round for further funding? Please explain why / why not.

Surely, yes. As stated in the plan, there are 2 development phases. This proposal is for Phase 1. There are features that we would like to develop in Phase 2. We will continue to ask for funding next Funds to make our product close to perfect. Our purpose is to bring the best product to serve the community.

[AUDITABILITY] Please describe what you will measure to track your project's progress, and how will you measure these?

Our KPIs are:

  • The number of total downloads on AppStore and Playstore >500
  • The star ratings on AppStore and PlayStore are over 4.5

We also measure the following indicators:

  • The number of interactions the chatbot has weekly, monthly, yearly. This will indicate whether users’ interest in the chatbot is growing, dropping, or remaining the same over time.
  • Chatbot response time (how long from when users ask till they get the answer).
  • The opinions, and satisfaction levels of surveyed users. The survey will be conducted in Google form and its results will be shared with the community.

Also, the community can access the chatbot evaluation by questioning the chatbot itself (e.g you may ask questions such as “How many downloads on Appstore?”, or “What is your satisfaction rating on Appstore?", etc).

For development progress, we will track our progress through:

  • The number of "done" tasks (we will use the Kanban management method in Trello)

  • The number of design documents and reports

  • The number of Github commits (we will use CI/CD to ensure the quality of our code)

  • The number of unit tests

  • [AUDITABILITY] What does success for this project look like?

The success of this project is a chatbot mobile app available in Appstore and Playstore. Ideally, the chatbot is frequently used and highly evaluated by users. Users will have a good experience and benefit from the answers of the chatbot.

[AUDITABILITY] Please provide information on whether this proposal is a continuation of a previously funded project in Catalyst or an entirely new one.

This is a new proposal.

Reseñas de CAs (1)

Comments

close

Playlist

  • EP2: epoch_length

    Authored by: Darlington Kofa

    3m 24s
    Darlington Kofa
  • EP1: 'd' parameter

    Authored by: Darlington Kofa

    4m 3s
    Darlington Kofa
  • EP3: key_deposit

    Authored by: Darlington Kofa

    3m 48s
    Darlington Kofa
  • EP4: epoch_no

    Authored by: Darlington Kofa

    2m 16s
    Darlington Kofa
  • EP5: max_block_size

    Authored by: Darlington Kofa

    3m 14s
    Darlington Kofa
  • EP6: pool_deposit

    Authored by: Darlington Kofa

    3m 19s
    Darlington Kofa
  • EP7: max_tx_size

    Authored by: Darlington Kofa

    4m 59s
    Darlington Kofa
0:00
/
~0:00