CropMatch - Byte Me

Homepage
Allows user to input specific nutrition values
Accommodating new farmers
Displays recommended crops and their ideal conditions

Inspiration

Agriculture acts as the cornerstone of the Australian economy, yet our country is experiencing a decrease in farmers due to high financial and knowledge barriers to enter the industry. The statistics are sobering; over the 30 years to 2011, the number of farmers dropped by 40%, from around 263,200 to 157,000, averaging a loss of about 294 farmers per month. On the other hand, new farmers often face the challenge of not knowing what crops should match with the soil type, resulting in the waste of resources and money when planting crops that are not suitable for the soil.

We believe that a significant barrier is the inaccessibility of agricultural knowledge to inexperienced farmers. In particular, we were motivated by seeing a family member new to the farmer’s world struggling in choosing suitable crops to plant and forbidding their interest in entering the farming industry, or the sheer amount of knowledge required to just maintain or start a small-scale farm, or even garden. Hence, we wanted to create a simple and effective way of distributing information about potential crops to people in the same situation as them.

Our website addresses the knowledge gap between new farmers and experienced farmers, by providing a convenient, simple and accessible way to help new farmers most efficiently match their crops to their soil type.

What it does

CropMatch is essentially a soil analytical website designed to assist new and experienced farmers in solving the fundamental economic problem of ‘What to produce?’. A user has the option to confirm their familiarity with their soil nutritional values, which would direct them into a questionnaire in inputting the exact nutritional concentration within the soil. Or an alternative based upon the appearances of soil types. Based upon the response received from the user, the website will generate an overscroll of crops recommendation based upon the prominent nutrients present, allowing them to acknowledge the type of crops best suited to their farmland.

How we built it

Our UI/UX was designed in Figma. We chose a simple design, cohesive green theme and smooth animations to complement the app’s purpose of creating a simple and accessible way to distribute crop information to newer farmers. The front-end uses HTML, CSS and JavaScript. For the backend, we used express.js and MongoDB and random forest. The model uses Gemini API.

Challenges we ran into

One of the first challenges we encountered was creating a solid product identity/uniqueness - the fundamental questions of “What makes us stand out from other apps and websites?” and “What do we do differently?” Quickly, we realised that our first idea was too ambitious and broad, and had to rethink our approach to this task. This led to us rebuilding our project from scratch with a renewed focus on targeting a specific problem within the agricultural sector - the knowledge barriers required to break into the industry. Next, our original method of collecting information for our databases - web crawling - was too time-consuming, so we resolved it by using pseudodata. Due to time constraints, we were not able to implement the predictions model. Finally, UI/UX design. In order to provide a good UX, the designing of UI on the homepage was indeed time-consuming, where a lot of effort was put into the css coding in attempting to provide a better UX.

Accomplishments that we're proud of

Being able to deploy a real-life website, and providing good UX from the designs accompanied with warm colours in UI.

What we learned

We learnt that farming involves a lot of data-driven decision-making. Also building a product from scratch - from brainstorming to realizing it, and how to use Figma to help us ground our ideas.