Grocery Chat Bot

Inspiration

The inspiration for this Grocery Chat Bot came from a personal experience. One day, I urgently needed a specific product and embarked on a grocery store odyssey. After visiting four different stores, I finally found the item at the fifth. This experience made me realize the inefficiency of traditional grocery shopping, especially when you're unsure of product availability or pricing beforehand. I envisioned a solution that could save time, reduce wasted trips, and potentially eliminate the need for multiple grocery store visits.

What it does

Provides Product Availability Information: Customers can ask the chatbot about specific products by name. The chatbot checks the inventory data to see if the product is listed. If the product is found, the chatbot informs the user whether it's currently in stock and tells it's Price and Unit Measure Details. It will also tell total cost of all products a customer wants to buy so that customer can carry enough money beforehand. By informing customers about product availability beforehand, the chatbot can help avoid wasted trips to grocery stores that don't have the items they need. When a product is not found in the inventory data, the chatbot indicate that the store might not currently sell it.

How we built it

1. The foundation of the chatbot is an inventory spreadsheet containing product names, availability, prices, and unit measures.
2. I explored the use of a pre-trained NLP model (like Vertex AI's Chat Model) to enable the chatbot to understand user inquiries and respond appropriately.
3. The inventory data was integrated with the NLP model to allow the chatbot to access product information and provide accurate responses.

Challenges we ran into

1. Choosing the right NLP model.
2. Integrating inventory data effectively with the chosen NLP model.
3. Ensuring data completeness to to get the correct responses.

Accomplishments that we're proud of

1. Identified pain points in grocery shopping - checking for product availability and enough money a customer need to carry to purchase the available products. Project directly addresses these issues.
2. Despite the challenges, I successfully built a chatbot that integrates with an inventory and can answer product availability inquiries.
3. Throughout the project, I learned new skills related to data management, NLP fundamentals, and potentially project development.

What we learned

1. GCP
2. Vertex ai capabilities
3. Gained experience in organizing and manipulating data from an inventory spreadsheet to be usable by NLP model.
4. Building the chatbot involved understanding basic NLP concepts like how a model can interpret user queries and generate responses.
5. The process likely encompassed tasks like planning, building, testing, and iterating, providing valuable project development experience.

What's next for Grocery Chat Bot

1.Connect the chatbot to a live inventory management system to provide users with the most up-to-date stock information. 2.If a product is unavailable, the chatbot could suggest similar items or substitutes that the user might be interested in.

Built With

• google-cloud
• python
• vertexai