Los Angeles County Parks and Recreation A.I. Agent Members: Jeffry Alvarado, Christopher Carlson, Joshua Lazaro, Garin Lee Advisor: Keenan Knaur Liaison(s): Mohammed Al Rawi, Chaitanya Madala
Parks and Recreation Parks and Recreation Background and Webpage The Current Issue Solution through an Artificial Intelligence (A.I.) Agent and Machine Learning
Project Goals Agent Features Low maintenance cost Extensibility Cross-platform Features Park Information Location / Directions Programs / Events Reservation Scheduling
Google Dialogflow Agent
Google App Engine Serverless Application Platform Faster project deployment and management Infrastructure ideal for web applications Testing, Staging and Live environments Cloud Developer Tools
Additional Technologies Programming Languages JavaScript Node.js Wordpress Framework PHP
Artificial Intelligence as Child Development A.I. is like a baby Both are blank slates Raise children and train A.I. Parks Assistant A.I. in early stages Development is training Continuous development Continues to learn after release Artificial Intelligence is like a baby in a way. In the early stages, AI is a blank slate, and it is up to us developers to train it, much like how parents raise a child. As training furthers, the AI gets smarter and becomes more knowledgeable. In it’s current state, our AI is still early in development and requires much further training to be ready for deployment. However, just like a human, AI continuously learns. As it gathers user data, the AI can be further developed to specialized with the needs based on the user data. WIth this being the case, the agent will not automatically know the answer to every question a given user has at first. When it encounters a question that it cannot answer, it will provide assistance where it can. The AI will instead give the user either an email or a phone number to direct the customer to the right place. If this happens often enough, we can take notice and train the AI to be able to answer the particular question for the future.
This is a screenshot of the current parks and recreation website, and right here, is an example of how our agent could be integrated to it. On it, we can see a possible conversation that a user can have with the AI. When it is opened, the AI will greet the user, saying “Greetings! How may I assist you?”. The user then responds by telling it that they want a reservation. It asks for more details in order to accurately answer the question, by asking what they are trying to reserve, and is told “ a campsite”. Given this information, the agents gives out the link to to form to reserve a campsite. However, with the limitless possibilities of artificial intelligence, it is possible to go a step further.
Current Progress and Future Vision Preliminary research completed Working prototype agent Backend infrastructure set-up Virtual Employee Low maintenance 24/7 support Hands-Free Interaction Fully voice operational Our vision for this Artificial Intelligence agent is to effectively have a low maintenance virtual employee that offers 24/7 support to users. Users would be to have a completely hands-free experience with the AI agent, and can even operate it on the go with the future Parks and Recreation app through. We envision a parent driving their car and speaking with our agent through the app, asking it to schedule a swimming lesson for their child. The AI would be audibly responding to the user asking for all the needed information that is needed to schedule the swimming. So far this project has given us the opportunity to learn about new technologies with the purpose of addressing the parks and recreation dept’s business needs and have provided us as a team to apply our years of learning into real world practice.
Q&A