Inspiration
We are working on a digital market place for Agriculture and Agro Businesses in rural India (http://salesbee.io). The locations are usually geographically scattered, struggle for valuable product information and lack any support or transparency that makes best business practices.
The small businesses and farmers have no way to know about the products they should be ordering, current promotions, accurate pricing and so on. Similarly, they usually don't have any data (in regards to crops they have sowed) for seasonal diseases, how to prepare for those diseases, and other products they should use.
Sometimes, they are not even aware of the best crops they should be sowing (which can depend on soil type, geographical data, climate, weather predictions, etc.) This makes the whole operation very inefficient and far more challenging than it should be. When talking to these people, we realized, they want to use technology to help their businesses, but they have no clue on where to even start.
That is when we came up with the idea of a chat bot that can leverage on data science and API MashUps and help these people to use technology and data to combat with financial and operations inefficiencies.
What it does
The Bot is still in its early stage as we just started three weeks ago and perhaps it is not even a right fit for the challenge. However, we wanted to spread the word about meaningful uses of Bots and how it is not just for ordering Pizzas and booking hotels.
Some Typical Use Cases
1) Users can ask about the diseases for the crops they have sowed. The Bot response factors in the current month, crop, and geographic location amongst other parameters.
2) Users can ask about the products they should use for these diseases or other crops in general.
3) Users can ask for additional information about the product.
4) Users can ask about trending products. The response accounts climate, what other people in their geographic location are ordering etc.
5) Currently, we are working on the part where they can ask what crop should they sow. The response will factor in user location, soil type, climate, last crop they had, water conditions, etc.
How I built it
Amazon Lex Platform using Lambda functions and mapping tables from our SalesBee Platform.
Challenges I ran into
1) natural language processing could have been better. 2) support for more languages would be beneficial as we want to make it global.
Accomplishments that I'm proud of
1) Yet to see
What I learned
Always start with "Why" and that "why" can't be money.
What's next for SalesBee
We have just started, and there are so many ideas we want to incorporate. We want to make it available in more markets and countries so that everyone can benefit from this. We will also want to bring in more companies on board so that farmers can directly get in touch with the businesses and get transparency they need. We also think adding photos and videos will be helpful and we are working on it.
Built With
- amazon-lex
- lambada
- node.js
Log in or sign up for Devpost to join the conversation.