Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/anasaito/ircai-submission


https://github.com/anasaito/ircai-submission

Last synced: 2 months ago
JSON representation

Awesome Lists containing this project

README 

        
# Aquascore 🌊💻



## University: UM6P/College of Computing

## Team Name: Neptune-Warriors

## Project nominator: Karima Echihabi

## Team Members: Anas Ait Aomar, Manal Cherkaoui, Yassir Fri, Mahmoud Maftah, Ayman Youss



### Overview 📝



In rural Moroccan communities nestled in the High Atlas Mountains, water plays a vital role in sustaining livelihoods and fostering development. However, these communities face challenges in maintaining and improving their traditional water management systems due to limited resources and lack of visibility. Despite needing modest funds, accessing support remains a significant hurdle. Our mission is to increase the visibility of these villages and facilitate connections with potential supporters using technology and AI-driven solutions to objectively quantify water challenges.



### Proposed Solution 💡



Our proposed solution involves the development of a water map that pinpoints remote villages and assigns them a water score based on accessible data such as satellite images (vegetation, water streams, roads). This score is enriched with textual data crowdsourced from visitors to these villages or scraped from internet documents. Additionally, the rich map will enable ranking these villages based on their water score, highlighting their water criticality to help funders and supporters identify where to direct their assistance. Furthermore, we aim to create a platform for discussing water solutions and motivating engagement through gamification features such as challenges, user badges, and events.



### AI Method 🤖



We have chosen a hybrid approach combining Computer Vision (CV) techniques with Natural Language Processing (NLP). By utilizing CV algorithms to segment satellite images, we can generate automated baseline water scores for each village objectively. NLP algorithms will then be used to extract insights/features from textual data, enhancing the accuracy of our water score assessments. These textual documents will come either from the internet or via reports submitted by our App users.



### Novelty and Uniqueness 🚀



Our AI solution's uniqueness lies in its data generation and refinement approach. Unlike conventional methods reliant on existing datasets, our system doesn't train on data but generates it, particularly crucial in areas with data scarcity like rural Moroccan communities. Our water scoring begins with an automated baseline derived from satellite imagery, then powers an interactive app where user interventions generate textual content. This data enriches our scores, forming a closed loop that employs active learning, progressively enhancing accuracy and relevance.



### Validation Strategy ✔️



Our validation strategy focuses on ensuring the accuracy and reliability of our village ranking system, which determines water criticality and guides resource allocation for users. Drawing from research on metrics used in recommendation systems, we'll employ established validation techniques to assess the effectiveness of our ranking algorithm. Furthermore, to validate the real-world applicability of our rankings, we plan to conduct field trips to select villages with water experts. During these trips, we'll compare the manual rankings provided by experts with the rankings generated by our AI system.



### Alignment with Objectives 🎯



Our AI solution is aligned with the objective of promoting water sustainability, focusing specifically on water management in constrained environments. By targeting this core issue, we aim to facilitate efficient investment of resources, both financial and technical, for potential donors and experts. Our system serves as a tool to sift through the complexity and noise, providing objective assessments that enable stakeholders to allocate their resources effectively. Simultaneously, by increasing the visibility of these villages through our AI-driven approach, we ensure that they receive the assistance they need in a fair and unbiased manner.



---



For a demo of our solution and access to sample village images aggregated by our pipeline, please visit [this link](https://drive.google.com/file/d/13xGuhJgEZSI2lDquKx1J4xV6ro7ZYbMi/view?usp=sharing). 🌐📸



**Note:** The code for our solution is available in this repository. 📂