Ask Your Knowledge Graph

chevron-icon
Back
project-presentation-img
ivan reznikov
Project Owner

Ask Your Knowledge Graph

Funding Requested

$120,000 USD

Expert Review
Star Filled Image Star Filled Image Star Filled Image Star Filled Image Star Filled Image 0
Community
Star Filled Image Star Filled Image Star Filled Image Star Filled Image Star Filled Image 5 (5)

Overview

Ask Your Knowledge Graph - RAGoverGraph is an innovative solution that integrates RAG techniques with a robust KG to deliver accurate, context-sensitive answers to user queries. This project aims to ingest and structure Deep Funding-related data within the KG by creating a seamless, interactive platform where users can efficiently search, retrieve, and generate insights from the structured data. In our previous beta round https://deepfunding.ai/proposal/metta-driven-kg-service-with-llm-integration/, we successfully demonstrated the power of Graphs and democratized their utilization. See our core modules from DF4-beta round here and our POC https://poc.knai.ai/.

Proposal Description

Company Name (if applicable)

KNAI

Our specific solution to this problem

Knowledge Nexus AI (KNAI) aims to harness the power of advanced technologies including LLMs, LangChain, Graph and Language Vector Embeddings, MeTTa.

We'll convert and upload all the DF data to graph structures and create an application to query the data!

RAGoverGraph leverages a combination of advanced LLM techniques to enhance the Knowledge Graph with real-time data ingestion, semantic analysis, and automated categorization. The solution includes robust APIs for data ingestion, management, and search functionalities, ensuring scalability and future integration across the SingularityNET ecosystem. By employing RAG techniques, our solution will deliver precise, context-aware responses to user queries

Alignment with requirements and objectives: does the proposal meet the requirements and advances the objectives of the RFP. Yes

Pre-existing infrastructure & risk mitigation: is the team starting from 0 or has already built part/all of the proposal, and are other risk mitigation factors at play. We already have some bits built as part of the DFR4beta

Team Competence: does the team have relevant experience. We have all the necessary resources to pull this off

Cost: does the proposal offer good value for money. Our proposal covers all the must, majority of the 'good-to-have' features. Our team has a record of successes, and as a cherry on top, we have some ideas to share for the RFP application besides Q&A - plagiarism and AI detection check, analysis of submitted applications, etc

Project details

Context and Background

The SingularityNET (SNET) ecosystem produces vast amounts of data, ranging from technical documentation to community interactions and proposals. This data is currently dispersed and siloed, limiting its potential. RAGoverGraph aims to address this issue by creating a foundational Knowledge Graph focused on Deep Funding-related data. This KG will be structured and enriched with metadata, relations, and trustworthiness metrics to serve as a robust database for various data applications.

Project Goals

RAGoverGraph will provide a comprehensive solution to ingest, structure, and manage Deep Funding data within a Knowledge Graph, enhanced with retrieval-augmented generation (RAG) techniques. The KG will support multiple applications, offering APIs for data ingestion, search, and management. This will facilitate the creation of a dynamic and interactive platform where users can efficiently query and retrieve context-sensitive information.

Solution Features

  1. Initial KG Functionality: Establish a scalable KG infrastructure representing Deep Funding content, including entities such as documentation, proposers, proposals, milestones, voters, and votes.
  2. Data Ingestion and Management: Enable real-time data ingestion from various sources via JSON-LD format, with data cleaning mechanisms to ensure quality.
  3. Search and Retrieval: Implement robust APIs to facilitate search functionality, allowing third-party applications to query the KG effectively.
  4. Semantic Enrichment: Integrate NLP and ML tools for automated categorization, tagging, and semantic analysis to uncover hidden relationships and enhance metadata.
  5. Community Collaboration: Develop tools and interfaces for community-driven data contributions, corrections, and annotations, fostering a participatory ecosystem.
  6. Trustworthiness and Reliability: Incorporate metadata for reputation scores and flagging mechanisms to track and ensure the trustworthiness of data objects.
  7. Documentation and Support: Provide thorough documentation and support to enable effective collaboration with other teams and ensure the KG's usability and maintainability.

Key Components and Specific Functions

  1. Graph Database (MeTTa): This will serve as the core of our system, where all data relationships are stored and managed. It provides the necessary infrastructure for complex query handling and scalability.

  2. Data Ingestion System: Utilizing LangChain and Python, this component will automate the extraction and loading of data from various APIs and online sources into our Knowledge Graph, ensuring the data remains current and relevant.

  3. Search and Query Interface: Integrated with graph queries, this component will enable complex data retrieval operations, essential for advanced data analysis and user interactions.

  4. Fast API Layer: This ensures secure and efficient communication between the Knowledge Graph and external applications.

  5. Documentation and Developer Support: Providing extensive documentation and support to facilitate easy adoption and integration by developers and third-party entities.

Our team

 

View Team

What we still need besides budget?

No

Existing resources we will leverage for this project

Yes

Description of existing resources

We’ve already completed some core modules in DF4-beta round and will be using it:  https://deepfunding.ai/proposal/metta-driven-kg-service-with-llm-integration/

Open Source Licensing

Apache

Additional links

Our POC: https://poc.knai.ai/

Our Whitepaper: https://knai.ai/#Whitepaper

Additional videos

Our demo: https://drive.google.com/file/d/1teZKE4JPpxQe7gP606WyYs79AEhegdZ0/view?usp=sharing&t=8

Proposal Video

Placeholder for Spotlight Day Pitch-presentations. Video's will be added by the DF team when available.

  • Total Milestones

    5

  • Total Budget

    $120,000 USD

  • Last Updated

    20 May 2024

Milestone 1 - Project Kick-off Initial Setup and KG Design

Description

Define project scope objectives and timeline. Assemble the project team and assign roles and responsibilities. Set up project management tools and communication channels. Prepare the development environment and initial infrastructure setup. Design the overall architecture of the Knowledge Graph. Define data models and schemas for the entities (e.g. documentation proposers proposals milestones voters votes). Establish relationships between entities. Plan for scalability and future integration with the wider SingularityNET ecosystem.

Deliverables

Project plan document. Team structure and role assignments. Configured project management and communication tools. Initial infrastructure setup and configuration. Detailed design document outlining the architecture and data models. Schema definitions for all entities. Relationship mappings between entities. Scalability and integration plan.

Budget

$30,000 USD

Milestone 2 - Data Pipeline and KG Functionality Implementation

Description

Develop the data ingestion pipeline to accept data. Integrate with the Deep Funding voting portal for real-time data ingestion. Implement data cleaning mechanisms to ensure data quality. Develop APIs for data ingestion from additional sources. Implement core functionalities for the Knowledge Graph. Develop search APIs to enable third-party applications to query the KG. Implement data management functionalities for adding editing and deleting data. Create a tagging system to organize content with rich metadata.

Deliverables

Functional data ingestion pipeline. Integration with Deep Funding voting portal. Data cleaning and validation mechanisms. APIs for data ingestion from external sources. Functional core KG with initial data ingested. Search APIs and documentation. Data management functionalities (add/edit/delete). Tagging system with rich metadata.

Budget

$25,000 USD

Milestone 3 - Community Collaboration Features Development

Description

Integrate ML/LLM tools for automated categorization tagging and semantic analysis. Develop algorithms to uncover hidden relationships within the data. Enhance metadata with semantic information. Develop interfaces for community-driven data contributions and corrections. Implement annotation tools for users to add insights and notes to data objects. Create a feedback loop or rating system for users to report inaccuracies or suggest improvements.

Deliverables

Integrated ML/LLM tools. Automated categorization and tagging functionalities. Semantic analysis algorithms. Enhanced metadata with semantic information. User interfaces for data contributions and corrections. Annotation tools for community input. Feedback and rating system for user-generated insights.

Budget

$25,000 USD

Milestone 4 - Trustworthiness and Reliability Features

Description

Develop mechanisms to assign and manage reputation scores for data objects. Implement flagging features for third-party applications to warn about data trustability. Create an auditable log of interactions and modifications to the KG.

Deliverables

Reputation scoring system. Flagging mechanisms for data trustability warnings. Auditable log of interactions and modifications.

Budget

$20,000 USD

Milestone 5 - Document Test Deployment and Maintenance Plan

Description

Develop comprehensive documentation covering KG structure functionalities data models and APIs. Conduct thorough testing to ensure all functionalities work as expected. Provide training materials for users and administrators. Deploy the KG to a reliable hosting solution and implement the maintenance plan. Establish a maintenance plan to ensure ongoing support and updates for 1 year.

Deliverables

Comprehensive documentation for developers and users. Test reports and bug fixes. Training materials for end-users and administrators. Deployed KG on a hosting platform. Maintenance plan for ongoing support and updates.

Budget

$20,000 USD

Join the Discussion (3)

Sort by

3 Comments
  • 0
    commentator-avatar
    Jan Horlings
    May 19, 2024 | 9:35 AM

    Is this proposal compliant with the actual RFP: https://deepfunding.ai/rfp/content-knowledge-graph/ ?

    If not, please add it to another pool such as Miscellaneous or, in case you are utilizing/creating an AI service, to 'new services.'

    • 0
      commentator-avatar
      ivan reznikov
      May 19, 2024 | 11:33 AM

      Yes, this is relevant. We've updated the proposal to make it more clear

      • 1
        commentator-avatar
        Jan Horlings
        May 19, 2024 | 11:41 AM

        Super. Good luck!

        Reply
        Upvoted by Project Owner

Reviews & Rating

Sort by

5 ratings
  • 0
    user-icon
    Nick
    May 20, 2024 | 8:58 AM

    Overall

    5

    • Feasibility 5
    • Viability 5
    • Desirabilty 5
    • Usefulness 5
    It will be high valuable product on market

    The proposed solution offers significant potential for enhancing research efficiency and discovery. By automating data cleaning and creating a robust knowledge graph, researchers can easily navigate and uncover relationships between papers, facilitating better citation analysis and recommendations.

  • 0
    user-icon
    Max1524
    May 18, 2024 | 1:03 PM

    Overall

    5

    • Feasibility 4
    • Viability 4
    • Desirabilty 5
    • Usefulness 5
    Follow every step the product\'s operation process

    This application is really good and it surprises the people who read it.
    The budget allocation is quite consistent with 4 milestones with a total of $95,000 (unfortunately, the milestones do not have an implementation time element).
    If researched and developed well - it has the potential to become a useful tool for everyone.
    In return, the team must closely follow the implementation process and follow closely how the tool works to gradually improve it - the product created by the team.

  • 0
    user-icon
    IndependentStream
    May 16, 2024 | 11:34 AM

    Overall

    5

    • Feasibility 5
    • Viability 5
    • Desirabilty 5
    • Usefulness 5
    Healthcare improvement

    As a neurologist, it can be challenging to shift to modern, intuitive systems that require some preparation. I am deeply researching the non-obvious correlations and relationships between oncology and stroke using machine learning methods, which are innovative compared to traditional methods that offer only a single perspective.

    I consistently face obstacles such as time constraints and a lack of resources, hindering my ability to improve patient quality of life through accessible technology. This product allows for rapid, high-quality, and structured insights and results. Searching for articles, encountering imprecise formulations, studying vast amounts of literature, and losing focus; having limitations in skills and knowledge to move from simple data collection to a new qualitative level and implementing solutions that will advance prevention rather than merely providing help to those who could have learned about their pathology earlier; spending much time on theories and so little on practice, consuming more and more new information and facing cognitive biases—all this can be avoided by learning to trust and understand new technology.

    This project resonates with me, and I hope that through its development and implementation, I will be able to make the world a better place for people. Great idea, I look forward to your success!

  • 2
    user-icon
    Joseph Gastoni
    May 15, 2024 | 8:47 AM

    Overall

    5

    • Feasibility 5
    • Viability 4
    • Desirabilty 5
    • Usefulness 5
    RAGoverGraph has potential to be a valuable tool

    RAGoverGraph has potential to be a valuable tool for data management and knowledge discovery. Careful planning and execution are required to ensure the accuracy and efficiency of the LLM, user-friendliness of the interface, and a clear value proposition compared to existing solutions. Demonstrating the effectiveness of RAGoverGraph in real-world applications can be key to user adoption and long-term success.

    This project proposes RAGoverGraph, a system that uses Large Language Models (LLMs) to create Knowledge Graphs from uploaded data. Here's a breakdown of its strengths and weaknesses:

    Feasibility:

    • Moderate: The project leverages existing technologies (LLMs, Knowledge Graphs) but requires significant development effort.
    • Strengths: The concept builds upon proven technologies and the successful completion of core modules in previous rounds.
    • Weaknesses: Ensuring the accuracy and efficiency of LLM-generated Knowledge Graphs requires expertise and data quality control.

    Viability:

    • Moderate: Success depends on user adoption, competition from existing solutions, and the complexity of query functionalities.
    • Strengths: Democratizing graph creation and offering plain language queries can be attractive to users.
    • Weaknesses: Competition from cloud-based graph providers and the need for a clear value proposition for specific user groups are challenges.

    Desirability:

    • Moderate: A system that automatically creates Knowledge Graphs from uploaded data can be desirable for businesses and researchers.
    • Strengths: The focus on user-friendliness (plain language queries) and open-source approach can be appealing.
    • Weaknesses: The effectiveness and accuracy of LLM-generated graphs compared to manually curated ones need to be demonstrated.

    Usefulness:

    • Moderate: RAGoverGraph has the potential to streamline data management and knowledge discovery for various applications.
    • Strengths: Automated graph creation and user-friendly querying can improve data accessibility and analysis for users.
    • Weaknesses: The long-term impact on user workflows and the actual benefits compared to existing solutions need evaluation.

    Besides, the project should consider:

    • Focusing on a specific target audience and demonstrating clear use cases for RAGoverGraph is crucial for user adoption.
    • Developing a strong marketing strategy to differentiate RAGoverGraph from existing cloud-based solutions is important.
    • Building a user community and encouraging feedback can help refine the system and address user needs.

    Here are some strengths of this project:

    • Leverages advanced technologies (LLMs, Knowledge Graphs) to automate data management and knowledge extraction.
    • Focuses on user-friendliness with plain language querying and an open-source approach.
    • Successfully completed core modules in previous rounds, demonstrating progress and feasibility.

    Here are some challenges to address:

    • Ensuring the accuracy and efficiency of LLM-generated Knowledge Graphs compared to manually curated ones.
    • Competition from existing cloud-based graph providers and the need for a clear value proposition for specific user groups.
    • Demonstrating the long-term benefits and impact of RAGoverGraph on user workflows compared to existing solutions.

     

  • 1
    user-icon
    Hanna Boichanka
    May 15, 2024 | 8:41 AM

    Overall

    5

    • Feasibility 5
    • Viability 5
    • Desirabilty 5
    • Usefulness 5
    Great application

    Great application!

Summary

Overall Community

5

from 5 reviews
  • 5
    5
  • 4
    0
  • 3
    0
  • 2
    0
  • 1
    0

Feasibility

4.8

from 5 reviews

Viability

4.6

from 5 reviews

Desirabilty

5

from 5 reviews

Usefulness

5

from 5 reviews