High-Level Project Summary
As a prototype, we created an AI-Powered search engine that takes short text input, fetches data from NASA's API, runs a self hosted AI as an API (if requested), then displays results (images) in a creative easy to parse way. This prototype will be part of the main web app that helps people access earth and space data in an innovative and effective manner, focusing more on making and enjoying their research rather than diving deep into filtering and understanding complicated data.People tend to learn fast when the information presented to them in a simple, organized and creative way. Having easy access to this data will significantly add more value to communities via insightful content.
Link to Final Project
Link to Project "Demo"
Detailed Project Description
What exactly does it do?
The current prototype is a demonstration of a search engine that our main web application will use to fetch NASA's images to include in their research or just download it for a cool wallpaper.
The app will basically be a platform where users and non users can search and collect information just for curiosity or for a project and merge them as a bundle with description and a title. The feed will be a public place where you can share that bundle or explore other people's bundles in case they choose to share it.
How does it work?
The Project is displayed in 3 main Websites:
Project Presentaion Website: https://insightinput.co/
The Project Presentation website is an initial interaction for people willing to learn about or contribute to the project. Making the understanding of the project's essentials easier.
Learn more about how the website works and code licenses at: https://insightinput.co/ & https://github.com/Insight-Input/website
Project Resources Website: https://cloud.insightinput.co/
The cloud for now, is the project resources gateway, helping archive and map all project websites, resources and links for better understanding of the project's functionality and roadmap. It will support in the future a multi-user system and will host the dashboard.
Learn more about how the cloud works and code licenses at: https://cloud.insightinput.co/ & https://github.com/Insight-Input/console
Project Demonstration Website: https://app.insightinput.co/
The current prototype is a demonstration of a search engine that our main web application will use to fetch NASA's images to include in their research or just download it for a cool wallpaper.
Learn more about how the app works at: https://app.insightinput.co/ & https://github.com/Insight-Input/app
The Main AI App for our prototype is apython scripthosted in server and runs with API requests. However, the goal is to build an AI Infastructure as described bellow.
Learn more about the use of AI as an API at: https://api.insightinput.co/ & https://github.com/Insight-Input/ai-api
What benefits does it have?
What makes Insight InputSpecial is how beautifully and easly people can conduct research! Making it fun and effective, using AI to help match users input to existing data and display it in artistic backgrounds, cover pictures, icons and illustrations.
It's super friendly, suitable for adults and kids of all levels. It could be used for school projects, professional research or maybe just for fun.
The app offers easy access to earth and science related data which will be then presented to the users in a simple,-organized, creative, innovative and personilized way so they can easily understand it. The next step is to build a multi-user system as described bellow.
What tools, coding languages, hardware, or software did you use to develop your project?
We used a combination of frameworks and tools such as:
- CSPs (SpaceApps & Student Offers: Microsoft Azure, Digital Oceans.
- Domain Provider (SpaceApps Offers): GoDaddy.
- Project Files: Google Drive, OneDrive.
- Code Host : Github.
- UI/UX: Figma, Adobe XD, Canva, PowerPoint, Google Slide
- Video Editing: Filmora, Kdenlive
- Open Source Code: NuroDev, Cruip, Rclip, Neuml.
- Frameworks: NextJs, React, Tailwind, Laravel.
- Programing languages: Javascript, TypeScript, Python, PHP.
- Data and AI Tools: Nuclia, HuggingFaces.
More details about source code can be found at: https://github.com/orgs/Insight-Input
Space Agency Data
NASA API Portal:
The type of data we used are NASA Image and Video Library search that provides images and videos according to a query a user sends. The way we used this resource is that we request the data from the API according to what the user inputs then our app formats the request and adds all the needed query string to the the request link and send a get request to NASA's API and then we get a response as a json object with all the data we need.
Example Images of Earth as Art:
For Testing our AI we came across one of the resources provided which is Earth as art and we used it while testing with Azure Cognitive Search AIto extract data from it according to a queries. Luckily we found a Azure Cognitive Search AI Samples Repos which already had a NASA document sample analysis.
We tried to work with an other beta software as well that analyze files called nuclia we tried to scrape articles related to images that we fetch and then add their URLs to nuclia's software database to be able to fetch all that data from our API in case e we want to add article support to our app.
Collection of Art and Science Galleries, Videos, Publications, and Opportunities to Create Art & European Space Agency’s “Art & Culture in Space” webpage & Mexican Space Agency Resources for Space Apps participants (in Spanish:
We were inspired mainly from NASA's Collection of Art and Science & ESA Art and Culture Website layout that provides a simple frontend that is easy to understand. Also, how foreigh language space data can be displayed from Mexican Space Agency resources.
Hackathon Journey
How would you describe your Space Apps experience?
This is one of the greatest experiences we had so far. We learned a lot, managed to practice what we have learned through the past years and build a web solution from scratch. It's A great feeling to see an app working after deploying it into the cloud.
What did you learn?
We learned a lot! We had the chance to experiment with amazing technologies such as cloud infrastructure. We built ci/cd pipelines and multiple experimenting servers and APIs and we also experimented with AI.
What inspired your team to choose this challenge?
This part did not go well as we tried to generate art on our local machine and it started sounding like a vacuum cleaner.
We love knowledge and we believe that it is the only solution to build anything. However, sometimes you need an artist to give you that crazy plan that has nothing to do with reality to push you to work harder and dream more.
This challenge combines both worlds: the knowledge part where we have to build a reliable, useful solution to share more knowledge and the artist part where we develop the most artistic way to share that knowledge with others.
What was your approach to developing this project?
The approach we took to build this project is breaking it into small parts as much as possible, as I like to call it baby steps.
More details about the Hackathon Journey can be found at: https://cloud.insightinput.co/timeline
What was your approach to developing this project?
First, we described the problem and studied it from all angles. What is it, how much it affects people, how much it will help people if it was solved and how we can solve it. Then we brainstormed about what solutions can fix it, Agreed on the solution that sounds real and doable, split the tasks and started working.
For setbacks git revert and npm -f solved most of our issues but as we described above we do it in baby steps. So, anything we get stuck on any step that won’t get resolved we split it into multiple other micro steps to solve the issue.
Is there anyone you'd like to thank and why?
We would like to thank the people behind NASA's API, the open source community, the people on stackoverflow and anyone who likes to give before they take.
References
- Main Data Source: NASA API Portal
- Large Snippets of code from Open Source projects: NuroDev, Cruip, Rclip.
- Pics and Other Resources Generated with: Figma, Canva
ALL details about credit and licenses can be found at: https://cloud.insightinput.co/credit
Tags
·#search,·#art,··#ai,#insight,#input