With the help of Fetch.ai's innovative technology, this project creates an intelligent agent dedicated to revolutionizing the shopping experience. Fetch.ai enables the development of intelligent economic agents *capable of executing tasks on behalf of users. By carefully integrating Fetch.ai's documentation and development tools, our project has created an agent that effortlessly handles all the aspects of shopping tasks. This agent not only interprets basic user queries related to **finding products, **comparing prices, and *checking availability but also excels in executing more complex tasks. These include making purchases on behalf of users, notifying them of more economical alternatives, employing *conversational AI for generating shopping item images, conducting **sentiment analysis on product reviews, and *preserving user preferences through context storing for a truly personalized and dynamic shopping assistant. This ensures a sophisticated and efficient shopping assistant that can adapt to the diverse requirements of users giving them the ability to independently carry out tasks that align with the user's preferences in the world of online shopping.
- Workflow: When the user inquires about the availability of a specific product, the assistant utilizes Fetch.ai's decentralized architecture to swiftly check multiple stores. The agent communicates back to the user with real-time information about the product's availability, ensuring a seamless and efficient shopping experience. Also, we can reserve the article for a particular use-case.
- Workflow: The agent monitors prices continuously in the background. If a product the user has shown interest in experiences a price drop, Fetch.ai triggers a personalized alert, notifying the user promptly. This proactive approach ensures users are informed about cost-saving opportunities.
- Workflow: In response to user requests for visual representations of shopping items, the assistant utilizes Fetch.ai's conversational AI capabilities to generate images. By understanding user input and preferences, the agent creates realistic and relevant images, enhancing the overall shopping experience through dynamic visual engagement.
- Workflow: When prompted for product recommendations, the agent performs sentiment analysis on product reviews using Fetch.ai's NLP operations. It evaluates reviews to provide nuanced positive or negative responses, aiding users in making choices aligned with their preferences and expectations.
- Workflow: The assistant stores user preferences and behaviors, creating a personalized profile using Fetch.ai's context-aware architecture. This profile informs future interactions, allowing the agent to offer tailored recommendations and a highly personalized shopping experience for each user.
- Workflow: This integration merges Fetch AI's decentralized AI with Gemini AI's personalized recommendations, creating a Smart Shopping Advisor. By combining Fetch AI's market analysis and Gemini AI's user profiling, this tool provides real-time, tailored shopping advice. Users benefit from personalized insights, market trends, and product recommendations, enabling informed decision-making
- Workflow: When the user decides to make a purchase, the assistant, leveraging Fetch.ai, initiates the transaction on behalf of the user. It navigates through the purchasing process, and order placement, thereby simplifying the entire shopping process and reducing user effort.
- Workflow: Upon receiving a query related to price comparison, the agent taps into Fetch.ai's decentralized network to gather real-time pricing information from various platforms. It then presents a detailed comparison, empowering the user to make informed decisions based on the most up-to-date and accurate data available.
- Integration with Fullstack React Shopping Website: Seamlessly integrate the Personal Shopping Assistant with the created fullstack React shopping website. This integration will empower users to interact with the assistant directly on the website, providing a unified and enhanced shopping experience.
Clone the project repository to your local machine using the following command:
git clone https://github.com/AshrayGattani/hackai-final-230675.git cd hackai-final-230675
We recommend creating a clean Python virtual environment using poetry for package management. Initialize a new poetry virtual environment:
poetry init -n && poetry shell
Now, install the required dependencies from the PyPI package registry:
pip install uagents
Set up your Fetch.ai environment by following the documentation provided by Fetch.ai. This may include creating an account, obtaining API keys, and configuring the necessary credentials.
Obtain API keys for Gemini by utilizing the Google Makers Suite. Configure your Gemini API keys in the project's settings to enable the agent to access real-time pricing and availability information.
To enhance conversational AI capabilities and utilize Hugging Face's model, generate a Hugging Face token by visiting the official Hugging Face website. Replace the placeholder token in the project configuration with the generated token for optimal performance.
Start the Personal Shopping Assistant by running the main script:
python main.py
Once the agent is running, interact with it by providing natural language queries related to shopping. Experiment with different requests to explore the diverse functionalities of the agent.
Dive into the project code to explore customization options and adapt the agent to your specific needs. The documentation within the codebase provides insights into how different functionalities are implemented.
Now you're ready to enjoy the personalized shopping experience with your Personal Shopping Assistant powered by Fetch.ai and the μAgents library! If you encounter any issues or have questions, refer to the documentation and community forums for assistance. Happy shopping!
Access Results on Google Drive: Google Drive Link