Breif description about the project
- Problem Statement
- Idea / Solution
- Dependencies / Limitations
- Future Scope
- Setting up a local environment
- Usage
- Technology Stack
- Contributing
- Authors
- Acknowledgments
Our task was to create a conversational AI model using Google Dialogflow that can understand natural language queries for diamonds and generate filtering parameters for a large inventory of diamond data.
Our idea is to make a chatbot for the diamond industry. It will bring a revolutionary change because user's can get faster and more personalized recommendations! There are several advantages of chatbots like:
-
Improved Customer Service - There is no need of waiting on call for the receiver to answer. Customer can get instant assistance regarding the shape, size, cut, polish and pricing of the jewellery. This can help to improve customer satisfaction and reduce the workload on customer service representatives.
-
Personalize the shopping experience: We can use data on a customer's browsing history and preferences to offer personalized recommendations for diamond jewelry. This can help to increase sales and improve customer loyalty.
-
Data Analytics - It provides a well-documented data of the demands of each customer so it is easy to track and accordingly maintain stock of the respective jewelry.
-
Efficient marketing: Targetted marketing helps to increase user engagement and sales.
Overall, chatbots can help to improve the customer experience, increase sales, and streamline operations in the diamond industry.
While we made sure that our chatbot is well-trained and as accurate and efficient as possible, there are a few limitations of the same.
The dataset contains values that are common across multiple columns, eg : diamond cut, polish and symmetry all have common attributes, like EX, VG, G, etc. This sometimes causes confusion when they are not explicitly stated.
Furthermore, our chatbot uses fuzzy matching to match the language to the desired parameters
We intend to integrate our chatbot with virtual reality to enable customers to interact with the jewelry in an immersive way. We would also like to train our model more efficiently in different languages to ensure no language barrier remains!
-
Download or clone the project repository to your local machine.
-
If you want to test the model locally, upload the zip file to Google Dialogflow and use it to test the chatbot.
-
Alternatively, you can use Google Colab to run the project. The code has been implemented with Google Translate API for multi-language support, which extracts parameterized information from the query given to and processed by Dialogflow.
Deployment link - on WebDemo mode
- Note: You may need to set up a Google Cloud account and enable the necessary APIs to use Dialogflow and Google Translate. For more information on how to do this, please refer to the official documentation provided by Google.
- Google Dialogflow - A conversational AI platform used for building chatbots and virtual agents
- Google Translate API - A cloud-based machine translation service provided by Google
- Google Colab - for implementation and testing of the model
This project was developed by:
@devansh-shah-11 - Generated test cases and implemented the project on Google Colab for feature extraction.
@Devasy23: Generated test cases and implemented the project on Google Colab for feature extraction.
@Divycholera98 - Trained the chatbot on various entities and intents and trained on various phrases.
@Aaryan-Chokshi - Trained the chatbot on various entities and intents and trained on various phrases.
@Rahilshah777 - Preprocessed raw data and created code to show analysis final results from the processed query.
We would also like to thank the organizers of the hackathon for providing us with the opportunity to work on this project.